JOURNAL FOR THE STUDY OF THE OLD TESTAMENT

SUPPLEMENT SERIES
21

Editors
David J A Clines
Philip R Davies
David M Gunn

Department of Biblical Studies

The University of Sheffield
Sheffield S10 2TN
England
This page intentionally left blank
COLOUR TERMS
IN THE
OLD TESTAMENT

ATHALYA BRENNER

Journal for the Study of the Old Testament

Supplement Series,21

Sheffield
1982
 Copyright © 1982 JSOT Press
ISSN 0309-0787
ISBN 0 905774 42 6 (hardback)
ISBN 0 905774 43 4 (paperback)
Published by
JSOT Press
Department of Biblical Studies
University of Sheffield
Sheffield S10 2TN
England
Printed in Great Britain by Redwood Burn Ltd.,
Trowbridge, Wiltshire.
1982
British Library Cataloguing in Publication Data

Brenner, Athalya
Colour terms in the Old Testament.
(Journal for the Study of Old Testament
Supplement Series, ISSN 0309-0787; 21)
1. Colors in the Bible
I. Title II. Series
221.8'5356 BS680.C55

ISBN 0-90577442-6
ISBN 0-90577443-4
T A B L E O F C O N T E N T S

Foreword

I. INTRODUCTORY REMARKS AND METHODOLOGY 1-35

A. The phenomenon of colour and

colour terms 3-17

B. The biblical text: difficulties

of interpretation and of semantic
classification 18-25

C. Some semantic terms and their

usage 26-30

D. Etymology: Its value for the

study of the field of colour 31-32

E. The field of colour terms:

a proposed structure 33-35

II. BIBLICAL COLOUR TERMS 37-174

A. Criteria for the classification

and description of colour terms
and of colour allusions 39-48

B. Primary Terms in the OT 49-105

0. General 49-57
1. d(i)TK 58-80
2. -p$ 81-94
3. l(i)iiS 95-99
4. pi-|> T 100-102
5. ah* 103-104
6. Summary 105

C. Secondary Terms in the OT 106-125

0. General 106
1. Within the D'TK sector 106-115
2. Within the l$5 sector 116-120
3. Within the "inS sector 121-123
4. Within the p'll* sector 124
5. Within the 5ns sector 125

D. Tertiary Terms in the OT 126-136

0. General 126
1. Within the tn$ sector 127-132
2. Within the }$ sector 133
3. Within the intf sector 134-136
 E. Man-Made 'Colours': paints, dyes,
pigments, dyed materials,
cosmetics, writing materials 137-155

0. Classification and
general remarks 137-139
1. Colour and paint,
colouring and painting -
general terms 140-142
2. Textiles 143-150
3. Cosmetic preparations 151-152
4. Dyeing agents, pigment
sources 153
5. Writing materials 154
6. Whitewash, plaster 155

F. Indirect Colour Connotations 156-168

0. General 156
1. Proper names 156
1.1 Personal and
gentilic 156-158
1.2 Toponyms 158-160

2. Covert colour allusions

allusions 161
2.1 Under dhK 161-162
2.2 Under itf 162-163
2.3 Under lhtj> 163-164
2.4 Under ph; 164

3. Metals, gems, stones 165

3.0 General 165
3.1 Gems and stones 166-167
3.2 Metals 167

4. Comparisons and Collations:

Items evoked to convey
typical colour
properties 168

G. Terms for speckles, spots, and

other multi-colour phenomena 169-171

H. Appendix: The pa°ol pattern 172-174

III. COLOUR TERMS IN MH 175-203

A. General 177-178

B. Primary Terms 179-185

1. dhN terms 179

2. 13$ terms 180
3. nhB terms 181-182
4. p"P terms 183
 5. an* terms 184
6. Others: A term for
'blue' 185

C. Secondary and Tertiary

Terms 186-192
1. Under Q'TK 186-187
2. Under ^ 188-189
3. Under Ihef 190
4. Under phj 191
5. Under 3'nx
T
192
D. Man-made 'Colours': Dyes,
dyed materials, make-up,
writing materials 193-201
1. Terms for painting,
dyeing, and colouring 193-195
2. Colourful textiles 196-197
3. Cosmetics 198
4. Dyeing agents and
pigment sources 199
5. Writing materials 200
6. Whitewashing 201

E. Summary 202-203
IV. SUMMARY 205-208

V. SOME REMARKS ON COLOUR TERMS IN

MODERN SPOKEN HEBREW 209-218

VI. NOTES 219-260

VII. BIBLIOGRAPHY: 261-280

in European languages 261-272

in Hebrew 273-280

ABBREVIATIONS 281-284
INDEX 285-296
FOREWORD
Scholarship has in modern times become increasingly aware of
the necessity for an adequate theoretical framework for the study
of vocabulary. Words cannot be understood in isolation, but only
as possible choices, as against other words, within the language of
a particular period; and they cannot be understood simply on the
basis of derivation and etymology, for these do not tell us how the
various units acted and interacted in actual use. Although all this
has come to be realized, scholarship has been slow in developing
methods by which entire word-groups and sectors of the vocabulary
could be isolated and studied as a whole. Many difficulties seem
to stand in the way of isolating one semantic area from another.
In this respect colour terms seem to provide a highly
interesting test case. One can, at least prima facie, hope to
identify what is a colour term with some prospect of general
agreement, and the literature provides sufficient evidence of
usage to provide a suitable basis for study. On the other hand,
the terms are not so directly loaded with religious nuances as
many biblical terms are. Moreover, they form an excellent testing
area for certain hypotheses about the correlation (or non-correlation)
between psychological structures and the building of vocabulary in
languages. The conventional glosses used in dictionaries and
translations of the Bible, which offer us the simple English "red"
or "blue" as rendering of a Hebrew term, must be regarded as rough
approximations or wild over-simplifications at the best.
Comparatively little systematic and informed study of these
problems has been attempted in the area of biblical scholarship.
The author of this work, Mrs. Brenner, is a sensitive and
sophisticated linguist, who approaches the subject with a thorough
appreciation of modern linguistic methods and of other scientific
approaches that bear upon the problem. Her description of the
Hebrew vocabulary, both biblical and post-biblical, has great
importance for the entire lexicological analysis of the language,
and may well provide the stimulus for many analyses of other
semantic fields in the future. It is a pleasure to recommend her
work to the world of scholarship.
James Barr
University of Oxford
A PERSONAL NOTE

Colour Terms in the Old Testament was first written as a

Ph.D. thesis, and submitted to the University of Manchester in
1979. At that time I received a grant from the Memorial
Foundation for Jewish Culture (New York), which I now gratefully
acknowledge. The Publishing Committee of the University of
Haifa made a generous contribution towards the expense of
publishing the work in its present form.

Prof. J. Barr, of Oxford, has given me wise guidance and

has been of the greatest assistance to me ever since I started
working on this project. My heartfelt thanks.

Last but not least. My family and friends aided and

abetted, suffered and hindered, while the thesis was written
and then made into the present volume. Without their support
and humour this task would have been much more difficult and
less enjoyable. To Chava and Benjamin (my parents); Shalom
(my long-suffering husband); Iddan (my son); Ruth Stadler;
Uzi Istikowitch; and Shalom Paul - THANKS.

Haifa, May 1982. Athalya Brenner

This page intentionally left blank
IINTRODUCTORYREMARKS
AND METHODOLOGY
This page intentionally left blank
 3

A. THE PHENOMENON OF COLOUR AND COLOUR TERMS

Light is perceived by the healthy human eye quite readily.

It is "one of many conceivable aspects of radiant energy" which
2
is part of our physical environment. The distribution of light
is determined by its source (for instance, the sun), and by the
length of the light waves; when light falls on any material, a
"transformation" of energy occurs. This energy is not lost but
is reflected, absorbed, and transmitted, in proportions that are
different at different wavelengths.3
The result of this physical reality - the distribution of
radiant energy at various wavelengths - is what appears to us as
visible "colour". Energy distribution reaches the eye of the
observer, is then transmitted through the observer's vision and
interpreted and turned into a sensation. Physics alone cannot
account for the appearance of energy distribution - visual and
mental processes are responsible for the recognition of radiant
energy as "light" and "colour". The visual aspect, however, is
4
"often less important than the brain behind the eye".
It follows, then, that while energy = light and spectral
distribution = colour can be analyzed in terms of physical
properties, they cannot be fully understood unless they are
considered as manifestations of features that are both psychol-
ogical and physical, i.e. they are psychophysical concepts.
Consequently, the evaluation of light and colour is of "medial
nature: implying at once an appraiser - the human observer - and
5
an object of appraisal - radiant energy". Objects are visually
accessible because light is reflected/refracted from them and
enters our eyes. Nothing can be seen without - at least - a
minimal amount of light. This is particularly valid where
colour is concerned: without light colours cannot be differen-
6
tiated and, therefore, are non-existent.
Objects which are seen frequently are the most easily
perceived. Perceptions of brightness, colour, shape, size,
number, distance, and movement are determined by the sense
organs and the nerve system. The repeated exposure to spatial
and temporal relationships cancels out individual differences of
4 INTRODUCTION

perception capacities and the manner in which attention and

concentration are applied to the visual environment. As a result,
many phenomena are perceived by "most people belonging to any one
human society" - and are classified, learned, and reacted upon - in
7
a similar manner. It is obvious that any object, even after it has
been perceived, continues to exist externally and independently of
spectral distribution of radiant energy as "colour" neither
changes the characteristic features of this phenomenon nor alters
its manifestations. On the other hand, the perceived object has
now become 'internalized': it has supplied the stimuli needed
for the formation of sensations. These experiences are
classified and categorized according to their appearance and
o
function, so that they can be efficiently utilized. They are
subjected to cognitive processes of learning and of assembling
additional perceptive data. These processes are not entirely
conscious, and their results do not necessarily mirror 'external'
facts faithfully. On the contrary; it would seem that human
awareness tends to exaggeration. A colour experience, for
instance, is remembered so vividly that its unique components
are over-emphasized: a bright red is perceived as even brighter;
a strong green becomes more so; and a yellow-green is polarized,
9
diagnosed, and defined either as green or as yellow. When a
spontaneous original impression is turned into an experience it
becomes a member of a set of similar occurrences that are
arranged according to a principle of their most typical, even
exaggerated, feature. Admittedly, this view of conceptualization
and subsequent codification and inter-person communication is
difficult to prove. What makes it plausible is the linguistic
practice of instituting general superordinate terms and of
using them even when more specific subordinates are available in
a given time or place. In other words: for the sake of economy
and clarity, the focus of the impression serves as a represen-
tation of the whole; the experience supplants the impression and
is substituted for it. The external object has been translated
into an 'internal' concept which is then encoded and stored for
future reference. A chain reaction has been instituted - the
conceptualized notion will assist in recognizing new perceptions.
The resultant sensations will (in their turn) become experiences,
COLOUR AND COLOUR TERMS 5

and will be subjected to various cognitive processes of

classification, arrangement, recognition, encoding, and analogy.
A typical mental image will evolve. This will be employed as a
yard-stick for the purposes of identification of objects, for
the recollection of similar events, and for sensual or aesthetic
enjoyment of our physical surroundings. Furthermore, perception -
especially that of ordinary aspects of the visible world, such
as brightness and colour - is heightened, even generated and
motivated, by previous experiences. "Expectations are
established as to the probability of occurrence in particular
situations of events and objects belonging to these categories;
10
and these expectations facilitate rapid and accurate perception".
Like any other sensory phenomenon, colour sensation is
transformed into mental images. These images are non-linguistic.
Colours are perceived independently of the mediation of language,
and their status vis-a-vis other colours can be defined non-
verbal ly as well as verbally. Babies who are a few months old
can discriminate between bright-coloured objects, especially red
ones, and less bright objects. They show their preference for
the bright object by focusing their eyes upon it, by movements
and by facial expressions. Hence, one concludes that recognition
and conceptualization occur long before the process of naming
takes place. A toddler will probably be two years old before he
begins to learn the names for the primary colours 'red', 'blue',
'yellow', and 'green', although he would have been able to
differentiate colour and respond to it at an earlier age.11
However, once the naming process begins there is greater accuracy
in perceiving and in discriminating colours: verbal symbols,
once acquired, not only aid colour perception but also enhance
and preserve it. Therefore, if inaccurate colour discrimination
exists independently of the appropriate verbal symbols attached
to it within a language system, then an analogy can be drawn
between child perception and adult perception. Further, the
existence of any colour name within the lexis contributes
towards further differentiation and organization of a colour
sensation by defining colours as contrasting entities. However,
the absence of a colour term from the lexicon does not
necessarily denote lack of recognition of this particular
6 INTRODUCTION

colour sensation on the part of the speech community, or of

12
individuals who belong to the community.
Although perceptions exist prior to the terms that refer
to sensations, the arrangement of the sensory 'raw material', as
it were, into cognitive sets depends largely upon the existence
of verbal symbols that refer to them. A verbal symbol is easier
to remember and to catalogue as a component of a structured
order than a mental image or picture. Images represent the
compact essence of sensations: they define the uniqueness of an
experience by stripping away elements that seem less important
than others, thus transforming sensations into concepts. Symbols
are even more economical, abstract, and precise than images:
economical, for they are short; abstract, for they are not used
to reproduce the original sensation but to recollect or to
simulate it; and precise, for they denote the typical qualities
associated with the essence of a given concept.
Language, of course, is fundamentally a bi-lateral exercise,
a "... system of communication by sound, operating among members
of a given community, by which human experience is analyzed and
segmented into units which possess phonic expression and
13
semantic content". Verbal symbols do not only encode an
experience but also convey it in a conventional manner, so that
the listener will grasp its meaning and will be able to respond
accordingly. Language signs are usually arbitrary in their
relation to the phenomena they encode, but can be acceptable
only if there is a basic agreement as to their form as much as
their contents; that is, the references to the 'real1 world of
sensations, objects, and ideas. The physical, non-linguistic
world is not defined and delineated in the way language is,
although language divides 'reality* into demarcated units. As
Ullmann observes, although the spectrum is a continuous band,
we impose onto it a certain amount of arbitrary observations or
definitions that may vary from language to language or even from
one period to another period in the life of the same linguistic
14
unit. The relations that link (verbal) signs, referents, and
15
concepts (meanings) are mutually influential, and various
linguistic factors may determine shifts in sense from one verbal
16
item to another. These shifts may mould, change, or seem to
change the presentation of the (non-linguistic) sensory
COLOUR AND COLOUR TERMS 7

experience of the speakers of a given language.

An illustration of these points can be gleaned by following
the usage of v^flif (Qal and Hif.) and some of its nominal
17
derivatives first in biblical Hebrew and then in MH. In
Lev. 13:30, 32, 36 the term n'ilS is applied to (human) hair. The
context is one of skin and hair diseases and the oppositional
term iW Ijny (vv.31, 37) defines nYtf as a colour term. The
""" T"is •*"
dictionaries describe it as denoting a gleaming, yellow or
yellow-red quality, similar to the quality of njtt = gold19 (in
its shine) and perhaps also genetically related to the latter.
The cognate supplied is c^-l/-* - 'be yellowish-red1, and the
derived (Hof. part.) form (in Ezr. 8:27) SHiTO (nwhj) is
explained as 'polished, gleaming' (bronze).
In MH the basic meaning of 3fi\ - 'yellow, gleaming' - is pre-
served and used in conjunction with *\y\!),but also with tl^a. There
are many forms of Qal and Hif. These can be divided into two
groups according to their syntagmatic relations in their contexts:
A. •/5nS' Qal + D'JJfl,Hull. 7:2 - *mtna TOS inHif
B. v^Hif Qal or Hif. + !?, *?y, TJMO + nominal (object),
Tal., Snh. 105a - .ITt? ,1T l^niS 1»m ... d^D 'Jt!^ "ym
Tal., M. Kat. 24b - T»^y D^IPrttn 0*31 WM33 NSITl
The dictionaries 20 record both the continuation of biblical
usage and, next to it, another sense of 'grieve1, 'be angry,
defiant' for the verb forms in MH and in Jewish Aramaic. At
first glance this second meaning can be understood only by
assuming a figurative usage of 1M^, H»ri2Hl : as Rashi says in
his commentary to Snh. 105a, when a man is angry 'p^-IDif TOO,
becomes tense and pale. But as Ben-Hayim points out, all the
instances of group A. ( \/^^^ + ti*3D ) denote pleasure and joy,
not grief or anger. 21 Thus we cannot assume only a polysemantic
shift from the concrete ('gleaming, yellow") to the abstract
(•anger1); but should we also assume a development by way of pola-
rization in the sense of the abstract, figurative usage itself -
from anger to joy -based on the same extra-linguistic observa-
tion that a person's face changes colour while his emotions are
stirred (without reference to the nature of these emotions)?
According to Ben-Hayim, this is hardly the case. The first
indication that this view should not be adopted is supplied by
the syntagmatic re-occurrence of /UMif + Q>3£J in the sense of
8 INTRODUCTION

'be pleased, glad', in contradistinction to the structure

exhibited by group B. To assume a state of polysemantic
polarization, be it primary or secondary, is risky in most
cases. 22 The solution he proposes is quite different. Group A
demonstrates the shift from the colour term n'flif, through the
observation that a man's face brightens when he is pleased, to
'show pleasure or joy'. Group B and its opposing sense is
etymologically unrelated to A. It is traced back to \/5fM in
ff
/ */ •
Christian Aramaic and /^-v_s-^ in Arabic (against S^«f-*£> ),
which basically means 'shout', 'scream'. This basic sense of
t/Dfi^ is then shifted and widened until it denotes 'quarrel1,
'be angry'.
Therefore, MH dictionaries should record two distinct 3Mif
23 4
sequences: the one derived from bib. Hebrew Shif, while the
other is unattested in biblical language but confused with the
former once it has become homophonous and homographic with it
in MH. A distinction is made in MH between the colour of a
happy or healthy person's face ('bright': D*TK. SMiD T
and that
"*" 24
of an angry one ('pale, greenish' - pi1*; fell in Tal. Ket. 103b).
It is true that biblical IMS is enlarged in MH by being applied
more widely (not only to 1JN20, and there is a greater abundance
of nominal (i.e. Uin^S) and verbal forms apart from the
figurative (group A.) usage. Nevertheless, it will be a
mistake to claim that bib. ^Jfiif is the basis for all /Iflif
T
occurrences in MH. As demonstrated above, a certain sense
originated from a sound shift - not a sense shift - which was
not recognized as such. This error was the basis for the
confusion showed by Rashi and by the dictionaries.
To summarize: the referential scope of bib. UJ12 as a colour
term is basically unchanged in MH. Perhaps we should say that
the wider application of the term, coupled with the appearance
of secondary derived terms, point towards a possible restriction
of it (although this is by no means certain). As will be shown
in the appropriate discussion - see 1M^ in MH, III, B.5, p.184-
even in MH , we cannot posit Ihi? as an exclusive Hebrew equivalent
to 'yellow', which is the slot UtlS belongs to in the "colour"
system of modern Hebrew. In addition, the term has acquired a
"non-colour" figurative sense through its colour denotation.
The position of its derivatives within the secondary field is
COLOUR AND COLOUR TERMS 9

made ambiguous by the existence, within that same (figurative)

field, of a homonymous term which denotes the opposite of /Sfitf I.
An attempt to carry v^MS I to the field of 'emotions and their
visible indications' as a root referring to the external
appearance of both 'joy1 and 'anger', and not 'joy' alone, would
therefore be incorrect and misleading.
Let us return to the psychophysical specifications of
colour.25 As noted above (p. 3), what we'regard as 'colour'
is a physical manifestation of light and illumination. On the
other hand, when we define colour as a psychological 'reality',
we have to list the three dimensions which are normally perceived
as the characteristic attributes of colour sensation. These are:
a) Hue, or the presence or relative absence of
chromaticity. An object which is perceived as 'red' is
immediately felt to differ from a 'green1 object even if the
'red' is quite dark in colour, while the 'green' is felt to
be light in colour. In terms of light, hues or chromaticity
represent dominant wave-lengths (red - long waves; green -
short waves, and so on).
b) Brightness. We distinguish between light and dark
colours even within the same hue range, that is, we apply a
quantitative criterion to the quality of chromaticity. The
leaves of the olive-tree are perceived as different from those
of the pine-tree, although there is an agreement about their
common 'greenness'. In terms of light, brightness denotes the
amount of white light applied to a visible object; or, in the
case of a pigment, how much white or black is mixed in with the
pigment in order to achieve the required result.
c) Saturation. The purity or intensity of the perceived
colour. If a coloured object is considered typical of its kind,
then it seems to possess a high amount of saturation. The
possibilities of confusing it with other colours are then
minimal. In terms of light, saturation is determined by the
existence - or the absence - of homogeneous, spectrally pure
radiant energy.
Apart from these three dimensions there are other factors
that contribute to the colour experience. These are variables
that represent our perception ofyhe visible properties of
physical objects. These additional features do not determine
10 INTRODUCTION

the colour sensation but influence it to a certain degree. A

person who is trying to define any given colour sensation
consciously will probably ask himself at least some of the
following questions:
1) Does the colour seem to be inclusive, like the colour
of blood, plants, or metals? Or, does it look like a surface
colour? (e.g. an exterior coating of a paint or dye?)
2) Is the surface reflective or is it relatively
non-reflective?
3) What are the texture and the shape of the object?
4) Is the surface glossy or matt, transparent or opaque?
5) How far from the scrutinized object or how close to it
is the observer? What are the visible-spatial relations between
?6
the object and its visual environment?
6) What is the scope of the coloured object? Is it
perceived as uniformly coloured or non-uniformly so? In the
latter case, how are the various visual areas set off by one
another? Is there a clear boundary between them or do they
merge into each other? It would seem that when we are confronted
with speckles, dots, stains, and the like, the shape of a colour-
ful object is as important as its colour; or, conversely, the
colour determines the boundaries of the shape.
The complexity of what we call 'colour' has been outlined
above only in a generalized manner. The multi-dimensional
properties and complexity of the sensation are probably at least
one of the reasons for the vagueness and relative ambiguity of
colour terminology in all languages. So many shades exist that
an attempt to convey them in words is bound to simplify,
exaggerate, and/or overlook fine distinctions. The terms vary
from person to person and even ,in the same person's speech,
while only a few colours are generally conveyed by single terms
27
that refer unmistakably to them. Let us now consider some of
the assumptions concerning colour-naming that have been advanced
by European scholars during the last hundred years, and were
repeatedly debated during that time.
a) In all languages, colour terminology (reflecting colour
discrimination) developed gradually and hesitantly. Some
28 29
scholars, notably Gladstone and - more judiciously - L. Geiger,
even concluded that the ancients' colour sense, as reflected in
COLOUR AND COLOUR TERMS 11

philological material, was less acute than that displayed by

modern Europeans. According to them, colour nomenclature is
the product of an evolutionary trend which corresponds to a
similar process of development in psychophysical perceptual
abilities.
b) The difference between light and darkness - and between
the abstract notions of light and dark, black and white - is
easily discriminated and encoded.
c) After the recognition of light as a defining factor
comes the naming of the chromatic, spectral colours. Indeed,
it seems that colour coding in European languages is motivated
first and foremost by hue, while saturation and brightness are
perceived as secondary attributes. The hue component is
perceived as the most specific feature, and is used as the
30
chief differential for terminology building. English illus-
trates this point very well. Spectral colours are referred to
by single lexical units - 'red', 'green', etc. Let us look,
for instance, at the range superseded by 'green': distinctive
sectors within 'green' (which includes a variety of hues, some
of them bordering on blue and yellow) are usually expressed by
a compound ('apple green', 'moss green', etc.) which is based
on the comparison to an object displaying this typical shade.
To all intents and purposes the compound, despite a transparent
structure that points to its secondary nature, functions as a
single unit. However, when variations in the brightness scale
are to be conveyed, the structure of the resulting syntagm is
different: a modifying term such as 'dark', 'light', or 'pale'
is attached to the colour name. Variations in saturation are
similarly defined by using the second structure, albeit with the
qualifiers 'deep', 'rich', 'strong', or 'dull'. Additional
31
attributes of the colour sensation display the same structure,
perhaps because they are felt to be secondary members of a wider
group that defines itself by referring to the same basic (extra-
linguistic) feature.
d) The human retina and nerve fibres respond maximally to
32
the red, green, blue, and perhaps yellow of the spectrum.
Hence, these colours are considered psychologically 'primary'
and are easily recognizable and verbalized. In fact they are
the first to be discriminated by small children.
33
12 INTRODUCTION

Since psychophysical perception and philological development

are analogous, the terms denoting these colours should be the
first chromatically-oriented terms to appear in any given
language.
e) The lack of a colour name in a language might imply the
lack of discrimination of that (chromatic) colour on the part of
34
the speakers of that language. This assumption was held to be
valid mainly until 1880. It was widely applied to so-called
35
'primitive' cultures and to literatures of ancient
36
civilizations, where the absence of a single term for our
37
'blue1, for instance, is conspicuous; colour names are few
(in comparison to modern European languages); and the boundaries
between chromatic colours seem to be only vaguely defined, if
at all.
Let us discuss these assumptions briefly in order to find
out if - and how - they can be applied to colour names in
38
ancient literatures. Magnus showed already in 1880 that
colour perception and its verbal identification are not directly
connected, although there is an evolutionary trend in the
development of nomenclature. Magnus investigated 'primitive'
39
languages, while G. Allen argued for the same approach to
ancient literature; that is, he refuted Gladstone's views of
colour perception and colour terms in ancient Greek. Rivers
40
reopened the question in 1901. Later in the twentieth century
the double-stranded evolutionary hypothesis (of perception =
terminology) was abandoned due to the preoccupation with the
'linguistic relativity' hypothesis. A different evolutionary
hypothesis has been in favour since the publication of Berlin
and Kay's Basic Colour Terms. Nowadays there is virtually a
general agreement as to the lack of correlation between
perceptual ability and the composition and versatility of the
colour lexicon. (See below, p. 14 ff.).
Brightness, or the relative absence of it, is the decisive
factor for naming the distinction between dark and light, black
and white. Black and white are achromatic (and hence are
sometimes called 'neutral'). It is often observed that because
41
they are hueless they are not colours at all, properly speaking.
This leads us to the third assumption, namely, the decisive
importance of the chromaticity factor for determining a colour
COLOUR- AND COLOUR TERMS 13

name. This notion has been repeatedly refuted. Although it is

correct as far as most European languages are concerned, it is
not necessarily so with regards to other languages, 'primitive'
and ancient alike. Heinrich demonstrated the importance of
brightness, or the dark-light contrast, for the colour
M
terminology of Bering Straits Eskimo speakers. 42 Durbeck 43 and
44
Fronzaroli dealt with the problem of misapplying these
chromaticity-oriented Western concepts to written languages.
Berlin and Kay attempted to determine the order, in which basic
colour terms enter the lexicon of any given language (see below).
The lack of correlation between nomenclature and perception is
45 46
summarized by Berlin and Kay, and by Vernon.
I would like to elaborate the objections - listed in the
previous two paragraphs - to earlier views and generalizations;
and, partly by following the Berlin-Kay hypothesis, to suggest
an evolutionary framework for the development of colour terms
in biblical Hebrew and post-biblical Hebrew. 47 It will be
submitted that names for the achromatic 'black/white' originally
had, and - to a certain degree - even retained, a basic meaning
of 'darkness'/1light1; that the motivation for installing further
terms was first brightness and only later chromaticity, while
the brightness factor has remained relatively weighty (See
discussion of 1D^ vs. Ihif*, IT*, and tVi, of the same field; and
TT T -
the distinctions in saturation and in brightness of the same hue
embodied in the contrasting pairs D'lN/D.in'TK, 17V/fni?V, ShB/irnhB*,
a contrast that is expressed by creating a secondary lexical
formation, but not by using a compound or a modifier + modified
nominal construction, as in English); and finally, that colour
terms have probably developed in the language in a partially
predictable order, and that the historical-diachronic description
of this process implies modifications in the traditional hue-
determined arrangement of colour names in Hebrew (as in other
ancient languages, or in modern languages that are spoken by
communities considered to be 'low technology profile'48
societies). Indeed certain terms, especially the achromatic
'black'/'white', preserve their 'light' or 'absence of light'
connotations even after they have developed into full fledged
symbols of colour concepts. Even DTK (the basic term for the
maximally saturated chromatic colour) and its derivatives are
14 INTRODUCTION

sometimes used to denote 'brightness1, despite their clear basic

chromatic denotation. Thus, in the syntagm tittWTK flJS!? (Lev.
13:19, 43) it seems - for considerations of sequential structure -
that na*TOlN modifies fiJ^ either from the hue aspect or from the
brightness aspect, or both; and the corrupt O'O'Jfltt QXy IQTK
(Lam. 4:7c) - by analogy to 7a, 7b, 7d and 8a - seems to deal
with the purity and brightness of complexion rather than with
any actual juxtaposition of 'white/black/red' as precise or
49
specific chromatic entities.
The work of Berlin-Kay has been referred to a number of
times on the previous pages. It seems appropriate, therefore,
to outline the hypothesis and the findings of that anthropolo-
gical-linguistic study, especially since the authors supplied
new impetus to research into colour terminology in recent years.
Their approach and their conclusions have been, and still are,
the subject of many critical evaluations; but, most important of
all, they have stimulated renewed interest in the field.
Berlin and Kay started by speculating that colour terms
translate too easily from one language into another to be
considered 'semantically arbitrary relative to any other
50
language'. This approach was proposed by, among others,
51 52 53
Ray, Gleason, and Nida as an example par excellence of the
non-existence of semantic universals. This doctrine of 'extreme
linguistic relativity' stems from writings by Sapir and Whorf,
who maintain that any language shapes and moulds the thinking of
its speakers in a way that is unique to it and different to other
54
tongues, or families of tongues. Berlin and Kay believed
intuitively that it was not necessarily true that each language
divides the spectral continuum arbitrarily, irrespective of
other languages and the conventional modes of psychophysical
perception. They examined 98 languages, mainly by eliciting
response to standardized colour chips that were graduated
according to hue and to brightness (but represented maximum
55
saturation) from native informants, and supplemented their
material by reading. The languages chosen are mostly contemporary
"living" languages which represent a wide array of differing
stocks or families - from Lebanese Arabic to Javanese, from
modern Hebrew to Hungarian. Therefore, their attempt to find
semantic universals in the colour domain is amply illustrated by
COLOUR AND COLOUR TERMS 15

results from linguistic, geographical, and cultural environments

that are completely different from one another, and in many cases
have had neither substantial contact with nor mutual influence
upon each other. The material investigated included only 'basic'
colour terms, and certain criteria were decided upon for the
purpose of the inclusion of a given term in this class or for
its exclusion from it.56
In essence, the Berlin-Kay conclusions are:
a) There is a total universal inventory of eleven basic
colour terms. These are: black, white, red, green, yellow, blue,
brown, purple, pink, orange and grey.57
b) Not all languages contain all basic terms, but all of
them possess terms for 'black' and for 'white'. When a language
has only these two terms in its colour nomenclature it is
classified as a 'Stage I 1 language.
c) If and when other basic terms emerge in any given
language, they appear in the following order:
Stage II. 'Red' (including orange, most yellows, browns,
pinks, and purles of our terminology; that is, hue,
intense brightness and relatively high saturation in
contrast to achromaticity and lower saturation).
Stage Ilia. 'Green' (including most blues), or:
Stage Illb. 'Yellow1 (including light greens, tans,
light browns).
Stage IV. 'Green' or 'Yellow' (whichever did not emerge
in the previous stage).
Stage V. 'Blue' emerges as a term defining parts of the
area formerly named 'green'. 'Green' becomes 'green' as
we understand it. 'Black' and 'white' are reduced to
achromatic, neutral values. The area covered by 'red'
loses its 'purple' and 'violet' references.
Stage VI. 'Brown'; 'yellow' and 'red' are further
restricted in scope.
Stage VII. One, or more, of the remaining basic colour
terms - 'purple', 'pink', 'orange', and/or 'grey' - is/are
added to the lexicon. Now the foci of the basic colour
terms are quite well defined by the existence of all the
rQ
terms, although the boundaries usually remain fairly fluid.
16 INTRODUCTION

d) There is a high degree of correlation between the

evolutionary stage of the colour vocabulary and between: 1) the
technological-cultural complexity of the society which uses this
vocabulary, and 2) the amount of contact that society has with
other, more technologically and linguistically advanced,
59
societies.
e) These experimental data can be corroborated by
"internal linguistic data" in order to demonstrate the relative
chronological emergence of the terms, mainly on grounds of
60
originality vs. loan words and of analyzability.
During the years following the publication of Basic Colour
Terms several field experiments were carried out in order to
verify aspects of the theory expounded in the book or to
contradict them. "Hill and Hill examined Uto-Aztecan colour
f\ 1 AO
terminologies; Snow, Samoan terminology;
Heinrich, Eskimoan
63 64
terminology in the Bering Straits area; Frisch, Mohawk terms;
65
and Broch, Hare Indian terms. All of these studies accept, to
a greater or lesser degree, the basic Berlin-Kay theory that some
colour semantic universals exist, and from here proceed to re-assert
or to refute details of the theory. Further discussions involve
criticism of the methodology employed, be it the nature of the
colour samples presented to informants in order to elicit the
66
spoken response, or the criteria employed to define the
67
'basicness1 of the terms.
The usefulness of the Berlin-Kay study for the description
and analysis of colour terms in biblical Hebrew (or any other
ancient or modern language) seems to be considerable, although
various modifications are called for.
Although Berlin and Kay discuss only what they call 'basic'
terms, other sectors of the colour vocabulary can be defined by
using their criteria in a modified manner. These criteria will
be discussed in the section dealing with the classification and
the description of colour terms and allusions to colour
properties in Part II of this study (in particular in II, A. 1.).

The 'semantic universals' described above were found to be

relevant mainly to spoken 'living languages', as attested by
informants. The experimental procedures used obviously cannot
be applied to literary languages of the past. This fact might
constitute a methodological objection to the relevance of their
COLOUR AND COLOUR TERMS 17

findings to biblical Hebrew, and to other ancient, 'dead1

languages. Indeed, other procedures have to be applied to the
existing material: a careful evaluation of terms in their
individual contexts, followed by the reconstruction of the whole
system - that is, the mutual relations of the terms as members
of a set - should be attempted. Although the nature of the
material excludes the type of experimentation and eliciting of
information current in the social and natural sciences, the
findings can still be interpreted first by using criteria
internal to biblical Hebrew, and then by comparing the results
with the picture of colour fields in immediate linguistic
cognates. Such a method would look for similarities, or
dissimilarities, in the organization and the structure of the
field rather than for a common origin of lexical items in related
stocks. If such structural and evolutionary similarities can be
found between biblical Hebrew and (some of) its cognates, which
/• o
seems to be indicated by the material, then the results can be
compared to other, unrelated families of languages. In other
words, the conclusions of Berlin and Kay will serve here as a
working hypothesis, to be strengthened or weakened, as the case
might be, by the conclusions of this study. Furthermore, by
adopting this structural-comparative method the language, its
speakers, and the culture it underlines can be seen in wider
perspective and to a greater advantage.
The difficulties involved in assessing the biblical material
will be discussed in the second section of this introduction.
There, the need for preferring internal linguistic reconstruction
to inter-linguistic comparison with cognate and non-cognate
colour nomenclatures will be expounded, and the implications of
our limited knowledge of the history and the lexical stock of
biblical Hebrew will be presented.
18 INTRODUCTION

B. THE BIBLICAL TEXT: DIFFICULTIES OF INTERPRETATION AND OF

SEMANTIC CLASSIFICATION

The problems involved in interpreting biblical material and

in describing it as an organized and coherent language system
are generally well recognized. Therefore, only a few observa-
tions of a general and generalizing nature will be attempted
here. Roughly speaking, the problems fall into three main
categories:
a) The arbitrariness of biblical language.
b) The history and the chronological ordering of the
various texts.
c) The varied contexts of the texts.

Such is our lack of information concerning these problems

that no work can be attempted without resort to conjectural
hypotheses which require frequent revisions and renewed
evaluations.
a) The arbitrariness of biblical language
Because no native speakers of classical Hebrew are available
as informants, procedures of eliciting information under experi-
mental (field) conditions are excluded. There is no possibility
of controlling a sample, be it an inclusive or a random sample.
Thus procedures adopted successfully for research in the realm
of the life sciences or the social sciences can be only of
limited use for the investigation of written languages. This
presents additional complications. Matters of tone and emotive
attitudes largely depend on non-verbal communication - the mood,
facial gestures, body language, tone, and pitch of voice of the
speaker - as much as on the non-verbal response of the listeners.
These are usually an integral part of a verbal exchange, as
revealing as the vocabulary chosen for a particular utterance.
Written transmission lacks the vitality of speech: a certain
amount of relevant data is excluded, unless again, the mood or
gesture is reported by verbal description. Unfortunately,
'psychological' states or observations only appear in biblical
literature because the circumstances demand them, or for didactic
reasons, but only seldom for purely descriptive purposes.
So far we have listed mainly difficulties which exist in any
THE BIBLICAL TEXT 19

written material. Biblical Hebrew presents us with other problems

as well. The student of OT language has no control over the
manner in which his data are recorded. The spelling and the
pointing alike often confound the phonetic realization of the
material because:
1) Only the consonantal skeleton of the text existed until
1
the first half of the first millennium A.D.
2) Fluctuation between plane and defective spellings -
inside the same layer of composition as well as across different
traditions - are so widespread that any attempt to establish a
systematic description of phonetic or scribal developments might
2
be considered an impossible task.
3) The vowel and accentuation system developed by the
Tiberian Masoretes proved to be a great leveller: it disguised
many characteristics of the living language by claiming a
basically unified vocalization for all texts, without distinction
of date or of geographical origin. Good original MSS. are rare.
As a result, what is known as the "Masoretic Text" has for
centuries been the basis for research into the vocalization of
biblical Hebrew but, as Ullendorff has rightly pointed out,
3
biblical Hebrew was never spoken the way the eclectic MT is now.
When we attempt lexical analysis our position is even more
precarious than it is on the phonetic level. The scope of the
texts was never decided upon by the redactors by taking
linguistic considerations into account. The vocabulary of the
OT is rather limited: almost a third of it consists of hapax
legomena, while an additional part of about 8% consists of
4
lexical units that occur only twice. To complicate matters
5
further, some books are richer in hapax legomena than others.
Extra-biblical Hebrew sources, from the Gezer inscription to the
Qumran literature, amply illustrate the fact that the Hebrew
lexicon of biblical times was much wider than the vocabulary
preserved in the OT itself. In these circumstances, if a
lexical unit does not appear in the bible or if it appears only
once, this fact should not serve as a basis for conclusions that
are too hasty. This is particularly valid when the extra-biblical
evidence of the documents mentioned above, and especially of
Mishnaic Hebrew, is utilized. plV (Job 39:8) is a hapax
legomenon. Fronzaroli suggests that the connection between
20 INTRODUCTION

lexemes that are derivatives of pIV - P1PV, 111*V - and

between the parental forms of pT>,
• •••
pT>
TT
('vegetation') is still
so transparent in the OT that it 'seems that p!1> represents a
•T
secondary, non-primary unit whose application and scope were
7
enlarged in post-biblical Hebrew. In the Mishna it is used as
a colour term for vegetation (A11HN - Sukk. 3:6) but also for
'blood' (Ed. 5:6) and a fowl's inner parts (Hull. 3:3), while
N/PT Hif. appears in a syntagmatic relation with human 'face'
(Sot. 3:4), among other things. It would be tempting to argue
•
not only for a later development of the basic colour term p1T>
in Hebrew, excluding it from OT language, but also for its later
appearance as a speech (performance) lexical item, and that even
before a discussion of its colour denotations - that is, its
sense significance - is embarked upon. Nonetheless, the
arbitrary nature of biblical vocabulary, the evidence of MH,
and the existence of the Akkadian cognate argu (which does not
always qualify vegetation, although it exhibits a transparent
etymological connection with the name for 'green things in
nature')8 might point to a different classification of plV *
in
biblical Hebrew. Consequently, the structure of the colour
field in certain levels of OT language should be described
g
somewhat differently from Fronzaroli's description. A more
detailed analysis of PIT, pT, PT, and related lexemes will
T VV TT
be undertaken in II, B.4 (OT language) and in III, B.4 (MH) of
this study (p. 100 ff. and p. 183 respectively)..
Finally, we can supplement the relatively unified tradition
transmitted to us by the spelling, the pointing and the
accentuation of OT texts by using non-Masoretic materials which
exhibit alternative systems of pronunciation - the non-standard
traditions of the Dead Sea scrolls; the Babylonian and
Palestinian pointing systems; the Samaritan traditions; early
Hebrew epigraphy; and post -biblical literature, which might
provide solutions for lexical problems. 10
b) The History and the chronological ordering of the texts
It is generally accepted that biblical literature encompasses
about a thousand years of creative literary activity, and that
many texts had a long history of oral transmission before they
were committed to the written form. Needless to say, such a
long span can hardly be considered a single period in the life
THE BIBLICAL TEXT 21

of any language, even if the MT would have us believe that the

basic lexical and grammatical features did not change much over
that period. Thus no diachronic or synchronic study of the
semantics of biblical language is possible without a
chronological ordering of the texts. Unfortunately, the
linguistic arguments for classifying certain sections of the
OT are often circular. When a lexical item such as MUIO (tribe)
appears abundantly in the Book of Numbers (90 times), it is said
to be characteristic of the usage of Priestly documents. When
it is found to feature 57 times in the book of Joshua, on the
basis of that fact (and admittedly, cumulative similar phenomena)
the two books are taken to be related not only by subject matter,
11
but by date and authorship. Thus the discussion of the
evolution and development of Mtt£ as an isolated lexical item is
determined by linguistic and non-linguistic data pertaining to
the P source, which is - in its turn - delineated by the same
often arbitrary linguistic data of whichfilDttis a part. Again,
because of the unified manner of MT presentation, conclusions
in regard to dialects (= locality-determined developments)
rather than chronological (= time-determined) factors are
extremely difficult to reach.
Despite the above-mentioned considerations, far be it from
me to reject the principles of chronological ordering of OT
literature that are the result of the last 150 years of biblical
scholarship. On the contrary: because evolutionary processes
cannot be described if they are not anchored in a chronological
framework, I shall follow the schematic division that is widely
used for biblical language, albeit with minor modifications.
The scheme, as presented by Gradwohl, divides biblical
Hebrew into four stages. The stages, which are characterized by
historical (and not linguistic) events, are:
1) Pre-monarchic Hebrew (to the beginning of the 10th
century B.C.).
2) Pre-exilic Hebrew (to the fall of Jerusalem, 586/7 B.C.)
c) Hebrew of the exilic period (to Ezra, mid-5th century
B.C.)
12
4) Post-exilic language (to the second century B.C.).
This rough sketch is obviously far from satisfactory. Each
stage sprawls over hundreds of years, and within each one major
22 INTRODUCTION

upheavals - economic, social, political - occurred, upheavals

that changed the language in which they were later described.
Nevertheless, the scheme accounts for the major political events
that undoubtedly influenced the course of the language, while a
more detailed division seems impossible because of the limited
corpus available. The colour terms preserved in the texts will
have, then, to be studied against the chronological background
of every occurrence - in so far as it can be determined in each
case - within the stage it belongs to. Only after the synchronic
stages have been described will a diachronic study of evolution
across the stages be attempted.
Two remaining factors must be borne in mind. Firstly,
differences of dialect between the Northern Kingdom and the
Southern Kingdom (Stage 2) seem to be minimal. This is probably
the result of the Southern scribes' language. On the other hand,
the differences might have been suppressed for geo-political
reasons. Secondly, MH is the natural successor of biblical
Hebrew. OT books which are considered late on the basis of the
events described in them (Esther, Ezra, Nehemiah) or the view-
points they advance (Qoheleth) also exhibit linguistic material -
grammatical and lexical - that is closer to MH than, let us say,
to the Song of Deborah (Stage 1), or the Blessing of Jacob
(Gen. 49; Stage 1 or 2). The language of the Hebrew Dead Sea
Scrolls, the Hebrew Ben-Sira, the early Midrash, and the Mishna
is a continuation of OT language as much as a, new language. It
is impossible to fix its origin in time, although it is probably
concurrent with at least the language of our Stage 4. Semanti-
cally speaking, MH lost some items of biblical Hebrew, but also
preserved and further developed many lexemes of the old stock,
among them colour terms, while incorporating foreign words -
13
Aramaic, Persian, Greek, and others. It even seems likely that
items that were not retained in the OT, but belonged to some
stage of classical Hebrew, reappear in MH lexicon. We are aware
of the foreign, especially Aramaic, influence on MH, but it still
is the most valuable tool we have for the understanding of
evolution processes in OT vocabulary and for filling lacunae
within it. Therefore, the study of biblical colour terms, like
the study of any other semantic field, cannot be complete without
a comparison with the cognate field in MH.
THE BIBLICAL TEXT 23

c) The varied contexts of the texts

DUrbeck argues that a free lexical meaning attributed to
any word independently of its context is not only an abstraction,
14
but a fiction as well. Let us consider, for instance, biblical
DTK. Could the JlfflTK niffl (Num. 19:2) and the O^atX D'XDlD
T TV-: TT ".7:
(Zech. 1:8, 6:2) have the same colour connotations as D'aiK O^nfi
0*13 (2 Ki. 3:22)? Or, is the colour quality referred to in
in:\*»O IK/HI nyrl Im'-OD V?y Wl (BH3:^'»niD:j) (Song. 7:6) similar
or equivalent to the 1Q>1N that is almost always paired with
ntOJl CimiKI tfJDtl, Ex. 26:!)? Clearly, the DTK which qualifies
'cows' and 'horses' differs in its colour denotations from the
DTK of DT
r
= blood, although both appear in the dictionaries
~f 1 15
under the same entry, translated 'red ; and when inJHN of
Song 7:6 is found within the list bearing the translation
"purple, red purple, i.e., purple thread and cloth" and defined
as a "simile, of woman's hair", 16 this does not seem satisfactory.
The distinction we make between the 'redness' of the
objects referred to by M19 and 0«10 on the one hand and 0*1 on the
other stems from two basic factors: 1. We know from experience
that no horse or cow skin tone is identical with the colour of
blood. 2. Because it modifies words that refer to different
classes of extra-linguistic entities, tJTK should be interpreted
separately for ms and DID, in contradistinction to 01. In
other words, the denotation of the colour term is determined
both by its situational context, that is, by its extra-linguistic
reference, and by its linguistic context,that is, by its actual
17
or possible collocations.
Because of the fragmentary nature of biblical language,
(a) and the uncertain chronology of many texts, (b), frequency
of terms and their distribution can be relied upon only in the
most cautious manner. Sums of the actual occurrences of any
given lexical item should therefore be regarded as indicative
of tendencies or, at the most, trends. They are only of
relative importance, for their presence or absence is dictated
by the context. The linguistic context is easier to define
formally by means of describing syntagmatic relations and
grammatical formations, thus taking care of the sense relations
within the linguistic system. The situational context is more
difficult to define because of its non-linguistic nature, and
24 INTRODUCTION

because the written text is far removed from the world of

references and from the actual language event itself.
The problem of understanding a passage such as Song. 7:6
might be helped by adopting a method proposed by Palmer. He
suggests Firth's scheme for the classification of situational
contexts, which is designed to supply the following details:
A. The relevant features of the participants: persons,
personalities.
(i) Their verbal actions
(ii) Their non-verbal actions.
B. The relevant objects.
18
C. The effects of the verbal actions.

When we try to utilize this scheme for the utterance 1WK1

IttATKD Itwa n!ni ^ni3D V?y the result is as follows:
A. The participants are a lover and his beloved; the
general background, a shepherd's life? or, perhaps, city life
(the many luxuries cited)?
(i) The speaker describes the head and hair of his beloved,
(ii) No non-verbal action is reported.
B. Head, hair - depending on the meaning of flt?T - is
described in terms of colour, but the reference does not make
19
immediate sense, for no natural hair is purple-coloured.
C. The effects are not reported.

It is clear that we have progressed only a little: the main

problem - the reference of t^ETO (= ^tti:>) and 1)1*1N in this
context - has yet to be elucidated by determining the level of
speech. Is it formal, colloquial, dialectal, idiolectical,
legal, and/or poetical? And, as we are dealing with literature,
we must also ask what is the literary genre of which the
utteranct, forms a part, and which might influence its level or
style. Indeed, the literature of the OT includes many types and
genres - from folk sagas to historiography and legal documents,
from poetry to wisdom teaching. Lexical items and their meanings
may be peculiar to certain types of literature as much as to
certain chronological periods. Once we recognize that Song. 7:6
is part of a poetical work, in addition to the other information
we possess about its components, we can begin to consider the
appropriate solution for the problem of IttAIN and ^013 in that
THE BIBLICAL TEXT 25

particular context. Poetry is characterized by emotive meanings

- no identity with the technically precise 1J3A1K should be sought:
metonymy of reference (IQAIK as sub-heading of DTK) can account
for this unique usage which belongs to the performance level of
the language, but not to the competence level. 'Incidentally,
this understanding would render any textual emendation of V3A1K
unnecessary.
To summarize: any linguistic utterance should be classified
in terms of its linguistic context and of its situational context,
In the case of 'dead' literary languages the literary context
(the typology of the literary genre) can form the basis for
additional observations. The latter may determine the speech
context, the level of diction, and the meanings that are either
unique to a given utterance or that, through a cross-section of
utterances whose contexts are thus grouped, serve as signifiers
that are widely accepted and recognized as such.
26 INTRODUCTION

C. SOME SEMANTIC TERMS AND THEIR USAGE

"The terminology of the subject (e.g., semantics) is

rich - and rather confusing, since it is used without
any high degree of consistency and uniformity between
different authors. It is inevitable therefore that
the terms introduced ... will not necessarily carry the
same implications as the same terms employed in other
treatments of semantics".
John Lyons, Introduction to Theoretical
Linguistics, p. 403.

a) Sign and Referent, Sense and Meaning

A verbal sign is a lexical unit, be it a word or a similar
speech unit (See b) below, "Word and lexeme") which constitutes
a part of a linguistic entity and communicates a message
concerning some non-linguistic entity, or state. (Verbal signs
like tlN, *yy, 1 have no one-to-one equivalents in the non-
linguistic world. Because of their exclusive existence within
the realm of language their functional status is fundamentally
different from that of 'fully meaningful1 lexical units. The
following paragraphs focus on the latter - that is, the lexical
type - because of the semantic orientation of this work). It is
a sound image reflected in, or deduced from, written forms. Used
thus, our 'sign' is the equivalent of de Saussure's 'signifier',1
Ogden and Richards' 'symbol1,
2 or Berlin and Kay's neutral
3
'term'. The referent is the external event, or object, which
is described by the linguistic sign. As such, it is the
4
equivalent of de Saussure's 'signified1.
In order to illustrate the point let us briefly consider
biblical Hebrew DTK. As it stands, it is a sound image
transmitted by script, a verbal sign which appears in a limited
number of contexts. The purpose of this study will be to define
tjTK in its relationship (i) to other signs, from O-tttTK, *31tt*TK
which have an external formal association with dIK to 13^ which
T TT
has a content relation with it - this line of investigation
deals with 'sense1, "... the complex system of relationships
5
that hold between the linguistic elements themselves"; and
(ii) to the non-linguistic world of experience. Here the
questions that should be asked are: What are the points of
reference of OTK in the visible world? Can it be defined by
SEMANTIC TERMS 27

actual occurrences of the sign? What are the implications for

other sets of 'sense1 and for other 'references'? Palmer makes a
convincing case for the inclusion of both sense study and
6
reference study in the scope of semantic investigation. His
conclusions are here accepted as the basis for discussion. As
Nida summarizes, "The study of semantics is meaningless apart
from the cultural context of usage. On the other hand, semantics
7
cannot be divorced from formal structural units".
b) Word and Lexeme
Dictionaries deal with words, and therefore it seems
reasonable to accept the word as the basic unit for semantic
treatment. Nevertheless, if we are to use 'word' as a semantic
term some problems should be noted. The term is widely used in
at least three ways:
(i) to denote a sound unit whose boundaries are
sometimes quite elusive. This is the 'phonological
word'.
(ii) to denote an orthographic unit that conventionally
has a space before and after it.
(iii) to denote a lexicographical heading, while the
various inflected forms are traditionally considered
Q

as different 'forms' of the same 'word'.

In view of this ambiguity, the term 'word' here will be
reserved for the phonological unit as it is represented by the
orthographic unit (i, ii). The third sense of 'word' will be
replaced by the term 'lexeme' in order to distinguish between a
basic lexical unit and the related inflexional forms which are
9
generated by the grammatical process of the language.
Therefore, different patterns of the same root - whether
nominals or verbs - of one series (like DTK, CfraTK. 'UiniK DTK,
T -T
O'WnrO will be called 'lexemes', while inflexional forms
(like .imK, 0*»7K,JWj»jtt<) will be considered 'words'. Thus
'words' are understood to be derived, through various processes,
from 'lexemes', and to exhibit some formal similarities with
their 'base1.
c) Syntagmatic Relations
A syntagm is a language unit that is larger than a word but
smaller than a sentence. It includes the immediate linguistic
elements that surround an investigated word. By isolating
28 INTRODUCTION

recurrent patterns through the listing of the various syntagmatic

combinations a word exists in within the text the contexts can
be systematically arranged, and both aspects of meaning - sense
and reference - may become clearer. Therefore, whenever a
colour term is approached a list of the syntagmatic relations
it enters into is supplied as one of the tools for defining its
meaning. To illustrate this point: If 1M1K1 n!?DH appear in
biblical literature mostly as a pair, this might be meaningful
in some way. On the other hand, the limited distribution of
IIATK as an allomorph of 1»>1K might be indicative of the
latter's slot within this particular area of OT language.
d) Semantic Fields
Ullmann defines the semantic field as "a closely knit and
articulated lexical sphere where the significance of each unit
.is determined by its neighbours, with their semantic areas
reciprocally limiting one another and dividing up and covering
10
the whole sphere between them".
Donald accepts this definition for his study of the field
11
of folly for two reasons: (i) Its usefulness for the understand-
ing of words, mainly in 'dead' languages. (ii) Its assistance
for the understanding of the internal development of the
language and the history of culture and ideas. 12
Indeed, the definition seems useful precisely because it
is so loose, and because it requires the arrangement of lexical
stock according to both sense and reference hierarchy. According
to it the field of colour should include not only all lexemes
that refer to chromatic qualities, but also those that refer to
properties of brightness and saturation which are felt to relate
to chromatic qualities. Therefore the field includes lexemes
which denote achromatic colours - 'white', 'black', 'brown',
'grey' - and so also partially overlaps with its neighbour in
the semantic mosaic of the language, the field of light and
darkness. In OT language this is demonstrated by the transfer
of lexemes from the latter to the former (cf. TT*, v^TTP, v^T2"1 )»
without losing their foothold in their original field. The area
within the field itself is divided into sectors. The heading for
the sectors are supplied by their referents. QTttTK, Vinn, V*)3N*,
DTK, and p^Kf* form a group that is distinct from *p^, h^, and It*,
T% % T TT ~" i"
or DfiS, IhB; on the other hand, it is the arbitrary sense
T T'
SEMANTIC TERMS 29

relations within the sector itself that decide the meaning

(reference) of each term. Again, a certain overlap between
sectors is not only unavoidable, but also necessary. The
absence of some terms may be overcome by directly invoking the
object which characteristically bears the relevant colour, or
this method may be used as a stylistic device (Isa. 1:15,16).
Finally, some lexemes are not used as direct colour terms
although they exhibit paradigmatic or formal associations with
'full' colour terms (e.g. names for precious stones, personal
names or place names). These too will be included, although
they belong to the boundary of the colour sphere.
e) The Diachronistic vs. Panchronistic Approach to the
Description of Semantic Processes.
Ullmann describes a diachronic (semantic shift) law as the
result of five variable factors:
A. sense (or word).
B. word (or sense).
Ci temporal factor.
D. spatial factor. ,,
E. set of conditions.
In other words, after we have investigated a certain term
we might arrive at the conclusion that A becomes B, or means B,
in a period C, within a spatial boundary D, under circumstance E.
Now, when we try to apply this procedure to OT literature we may
find that our knowledge of the material is so inadequate as to
render the procedure almost useless. The difficulties of
establishing the chronology and the geographical origin of many
texts were touched upon in the previous section (pp. 20-22) of
this introduction; in many instances we shall find that the
variables Ullmann refers to as C and D cannot be determined with
certainty. The lexeme h^ appears in biblical Hebrew four times.
Of the three nominal occurrences two (Isa. 18:4; Jer. 4:11) have
no colour associations, as demonstrated by their syntagmatic
relations (US Oh, hS n«11 respectively) and by their contexts.
The third occurrence is defined as a colour reference by its
k .
syntagmatic proximity - DllKl h^ 'Tin (Song. 5:10) - although
the colour referent is not clear. The verbal lexeme -Ih^
(Lam. 4:7) is defined by its syntagmatic position and its
14
context as an equivalent or partial equivalent of "p^. This
last statement of referential meaning does not solve the problem
30 INTRODUCTION

of the h^ referent in Song. 5:10. Furthermore, even if we date

Lam. 4:7 with relative accuracy (Period 3: after the fall of
Jerusalem), the lack of additional statistical evidence concern-
ing the same lexeme does not allow us to make any generalisation
about the time of the shift n^: "glowing, clear) •—^ n^: ll!? or
similar to "p!?, for we do. not possess enough information about C
(time) and D (space). At the most, a statement about the
meaning of .Ihif in Lam. 4:7 can include the tentative generaliza-
tion that •Ihi?
— and «lS>t— become - or function
. as - rparallels to "p!?
-r-r
when associated with ^V) and/or ±!?n, which are known to function
".••/ TT
as specifications of l^ elsewhere (Ps. 51:9 - "pH^K Atwtk;
TT • y •••••
Isa. 1:18 - .U^^ rtWto; Gen. 49:12 - tfrvb d'3W Vtf). This
• •- vv— TT» «-• %•:
tentative statement is achronistic in nature: it only takes into
account factors A, B, and E of the list above, without attempting
synchronistic or diachronistic distinctions. It is what Ullmann
16
calls the "panchronistic approach". This approach will be
resorted to when a synchronistic - diachronistic description of
a given problem does not seem feasible for lack of sufficient
information. The same panchronistic approach will, perforce, be
employed for the analogous presentation of the colour field in
MH and its scope vis-a-vis the field in OT literature (Part III).
ETYMOLOGY 31

D. ETYMOLOGY: ITS VALUE FOR THE STUDY OF THE COLOUR FIELD

1
In his article 'Etymology and the Old Testament' Barr
distinguishes the following types of scholarly etymology:
A. Prehistoric reconstruction (pp. 4-7).
B. Historical tracing (7-9)
C. Identification of adoptions from other languages (9-11).
D. Analysis of words into component morphemes (11-15).
E. Use of a cognate language to discover the sense of a
word in Hebrew (15-16)
F. Comparison of institutions with cognate names (17).
Gradwohl published his study of colour terms in the OT, Die
Farben im Alten Testament, in 1963. His work leans heavily on
etymological evidence, mainly of the types A and E listed above.
Whenever possible he supplies the cognates for each entry in
other Semitic languages; then he proceeds to discuss individual
colour terms by grouping them chromatically and by listing the
evidence of the ancient VSS (Greek, Latin and Aramaic) as guide-
lines for understanding the referents of the terms.
The material he has collected is no doubt valuable if taken
as a basis for further study. However, his method has many
shortcomings. This type of etymological approach is, of course,
atomistic: it deals with single lexical units, not with sets of
units, whereas it is the existence of other terms - alongside the
term discussed in each case - which delineates its area and
defines its scope. Etymological exercises, especially of the
type Gradwohl uses, are often of a hypothetical and reconstructive
nature. They tend to search for the origins of a lexeme while
semantic investigation tends to look in the opposite direction,
from the historical beginning of a lexeme (as documented in the
text) onwards, inside periods of time or across them. The
history of 11^ in Hebrew, from its earliest occurrences to MH,
depends less on the hypothetical origin of the term, but rather
more on the actual contexts in which it appears. Shifts of
meaning of individual lexemes cannot be described in backwards-
oriented, etymological terminology; nor is it sufficient to
summarize the development of OT colour terminology by way of
dividing the findings into four schematic stages (pp. 91 ff.).
Finally, the cognate terms Gradwohl uses are taken from the
32 INTRODUCTION

standard biblical dictionaries, where the cognates appear out of

their original contexts. The dictionaries are not .always right
in their definitions or etymological descriptions (cf. the
etymology of \/y$X - Gradwohl, p, 61; and Blau's comment in a
2
review of Gradwohl 's work. )
It seems, then, that etymological study, especially of types
A and E, is of limited importance only if not used as a base for
further investigation. Therefore, although the information
Gradwohl has painstakingly collected will be gratefully utilized
here, the emphasis will be away from the origin of single colour
terms and towards their development, in an attempt to uncover the
pertinent linguistic and non-linguistic variables that determine
the structure of the field as a whole and the respective position
of each component within it .
THE FIELD OF COLOUR TERMS 33

E. THE FIELD OF COLOUR TERMS: A PROPOSED STRUCTURE

Part II of this study will deal with colour terms in

biblical Hebrew, while part III will deal with the relevant
terms in MH. The criteria for the organization of the material
into structural units covering the field will be given in the
first section of part II and are applicable to part III as well.
Hence, a short outline of the classification employed will
suffice here.
Gradwohl organizes OT material under chromatic headings,
that is, all words that seem to mean 'red' are grouped together,
and those denoting 'white', 'black', are similarly collected and
dealt with as a group, from so-called adjectives to verb forms
and to place names. In a different chapter he deals with dyes
and pigments; and, later still, with the generalization of the
material into stages of development.
This arrangement seems to me unsatisfactory. It is not
structurally meaningful, although the internal ordering from DTK
to V^ton, Vinn*, IK*, iVl^*, I.IINfl is somewhat indicative of the
T T
relative weight of each term in its sector. The principle of
chromaticity is taken to be the chief motivation for the
creation of colour terms, although this is not valid for every
case. There is no firm distinction between direct colour
• •
references (DTK) and indirect colour connotations (DTTK). Lastly,
the organization of the sectors as mutually exclusive complemen-
tary units is far from clear: 'red' and 'black', for example, are
discussed separately; as a result, there is no awareness of the
non-chromatic denotations of various colour terms.
An alternative method of organization has been chosen here.
The primary terms, those that are recognized to have the widest
scope, are treated first. These "blanket terms" are usually easy
to spot: they enter a great number of syntagmatic combinations,
are widely distributed throughout the OT, and are applicable to
lexemes that belong to many fields. It is important to list and
to analyze them as a group because they delineate each other and
supply the framework for other terms that are their subordinates
•
within the field hierarchy. Biblical QTK (and its direct verbal
. .T .
derivatives), "p^, *\na - and perhaps SMS and i7lT> - belong to
TT T T f
34 INTRODUCTION

this primary group. Their sense is determined by their being

mutually exclusive, while their non-linguistic references depend
chiefly on the listing and on the interpretation of their
contexts.
The next, lower layers are those of secondary and tertiary
terms. These are more restricted in scope than the primary ones
and are often - and quite transparently - derivatives of the former
("UTO-*1*hini2>*, d**TN -? QJprrN). The internal territory of each
colour sector - the combination of hue, brightness, and saturation
- is divided into overlapping sections by these subordinate terms.
All secondary terms will be grouped together under the heading
of the relevant reference. Thus dTQTK, d^IKn, VOK*, pTK;* are
secondary subordinates to DTK; and IHIhW*, d«1h to the "frlty sector.
T T
The same procedure will be applied to tertiary terms.
The fourth category is that of paints, dyes, dyeing materials
and dyeing. These too will be grouped according to their colour
references wherever possible. The fifth group is that of
lexemes that have colour connotations although, in contradistinc-
tion to the first four groups, the latter cannot be considered
1
direct or "full" colour terms. The material that belongs here
ranges from proper names (13^, imp) to names of objects
(3Tiy, ruin^), metals and gems.
•• T *.
The next group consists of terms for speckles, stains and
multi-coloured phenomena. These invoke a spatial as well as
visual reference. Still, they possess sufficient formal features
("fia*, TpJ*, *TpJ>*) or referential associations (MOpI, nil'ia'ton*)
T T -r Tj.
to make them part of the field.
This structural framework - a number of categories of direct
colour terms arranged in a descending order, followed by groups
of indirect categories - will be observed for the description of
the field in both biblical Hebrew (Part II) and Middle Hebrew
(Part III). In addition, an appendix will deal with problems
that are specifically relevant to biblical language, namely:
1) Remarks on the morphological pattern pa ol (the
so-called d'jm nKIl "?i)m in Hebrew) .
2) General considerations concerning the relations between
word classes - 'verbs', 'adjectives' and 'nouns' - in biblical
Hebrew. A cursory glance in any biblical dictionary would reveal
that verbal forms, whenever they exist, are listed prior to the
THE FIELD QF COLOUR TERMS 35

listing of nominal forms that derive from the same root. Within
the colour field this practice is not adequate: nominal forms
precede verbal forms and generate them. The status of colour
terms as 'adjectives' is far from certain in some instances
(Gen. 25:30 - Jim QTKn oWl 1& Nil >:)l3^n), while some 'verbal1
combinations are equal not only in syntactic function but also
in deep structure to the analogous 'adjectival' surface structure.
If we compare IhtJ) my (Job 30:30) to ^K miht2> (Song. 1:5),
especially when these are isolated from their complements and
syntagmatic neighbours, we find that the difference between the
two does not go beyond the surface structure. These considera-
tions cause doubts as to the traditional division of OT stock
into 'parts of speech' and the inter-relations between the
traditional categories. These doubts will be aired in the
appendix (p. 172 ff.) .
In part III the fields of direct colour terms (the first
four categories) in biblical Hebrew and in MH will be compared
in an achronistic manner. Following a summary of the conclusions
of the main body of this study (IV), a final chapter (V) will
present a few of the aspects relevant to the history of the
unnatural, conscious process of renewal and invention of colour
terminology in contemporary Hebrew, a process that relied heavily
on the borrowing of biblical Hebrew, MH, and modern foreign terms.
Thus old word stock was - and still is - brought back to
linguistic life and supplemented by new terms that have become
necessary because of temporal, cultural, and technological
changes.
This page intentionally left blank
II BIBLICAL COLOUR TERMS
This page intentionally left blank
 39

A. CRITERIA FOR THE CLASSIFICATION AND DESCRIPTION OF COLOUR

TERMS AND OF COLOUR ALLUSIONS

1, Primary Terms
I chose the attribute 'primary' rather than 'basic' because
the connotations of 'primary' seem to be wider than those of
'basic': 'primary' indicates both the significance of a term as
a dominant factor together with its genetic originality as a
specific sign for its referent.
Berlin and Kay set the following criteria for determining
1
the "basic-ness" = (primary status) of any colour term:
a) It is monolexemic: its meaning is not predictable from
the meaning of its parts. In English, this criterion will
exclude expressions such as 'bluish', 'lemon-coloured', 'the
colour of . . . ' . By analogy, lexemes like *3(1)BIN (Gen. 25:25;
1 Sam. 16:12, 17:42) and mrnrw (Song. 1:6) are eliminated from
the 'primary' group.
b) Its signification is not included in that of any other
term. 'Crimson' and 'scarlet', which for most speakers of
English mean 'a type of red', are excluded by this criterion.
Thus even before we define ''DW as a 'dye' and/or 'dyed material'
we cannot attribute primary status to it, as Isa. 1:18 -
.,„ ytnnD WIN* DK ... O>JKD DD^KDH vn> &N restricts •>:)» (thus
the ancient VSS and the IQIsa ) to a type of 'red'.
c) Its application is not restricted to a narrow class of
objects. 'Blond', which can be applied (only) to hair and
complexion, is thus eliminated. Similarly, d«in (Gen. 30:32, 33,
35, 40), which is used to only describe the appearance of sheep,
is excluded from the 'primary' category even before considerations
of distribution, reference, or status vis-a-vis other terms are
embarked upon.
d) It must be psychologically salient for informants: the
references of the signs should be easily recognizable and
adhered to without doubt. Border cases like 'bluish1, 'blue
green' are excluded by this criterion as well. Berlin and Kay
state that these four criteria are sufficient in most cases, but
that four additional ones might be applicable to a minority of
difficult cases:
40 BIBLICAL COLOUR TERMS

e) The doubtful forms should have the same distributional

potential as previously established basic terms. In English,
the fact that the suffix '-ish1 can be added to 'red', 'white',
'green', 'blue', defines the whole series as a 'basic'one.
Applied to Hebrew, this would mean that OJTttIK ,inilTO*, and even
|7li7T> point back to primary forms that underline them - tHK, ine),
171V The criterion may be especially helpful for establishing
the status of p!1> (Job 39:8) which - as it stands - is a
hapax legomenon.
f) Colour terms that are also the name of an object which
is characterized by that colour are suspect ('gold', 'silver',
'ash', but not applicable to 'orange'). In other words, the
link of transparency between a colour term and its possible
origin should be severed before it has become a primary term.
•g) Recent foreign loan words may be suspect: they might be
an addition, perhaps synonymous for a more established term.
This could be one of the criteria for deciding upon the status
of the lexeme Tin, which is a hapax legomenon in OT Hebrew
(Isa. 29:22 - -lllh* 1^0, although it is possibly related to 1.1 n-
Esth. 1:6, 8:15; Vifl - Gen. 40:16; and 'lit! - Isa. 19:9), while
• T
undoubtedly of a primary position in the field of colour in
Aramaic.
h) When it is difficult to determine the monolexemic
status, morphological complexity can serve as a secondary
criterion. Thus, this would be an additional argument - if one
is still needed"- for the placing of DTttTK ,inihl!>* ,171171''
outside the range of primary terms.
When applying this procedure of multiple criteria examination
to biblical Hebrew and to its cognates, the differences between
Berlin-Kay's material and ours must be emphasized, Whereas they
deal principally with 'living' spoken languages, we are dealing
with ancient and, for us, literary languages. Therefore, some
modification of their criteria is called for if it is to be
adopted as a basis for the discussion and analysis of biblical
material and its cognates.
Hebrew lexemes which belong to the nominal or the verbal
classes can hardly be considered "monolexemic". A Hebrew
lexeme has at least two basic components: a consonantal root; a
pattern that includes vowels; and, in many cases, prefixes and
CLASSIFICATION AND DESCRIPTION 41

• 4 «

suffixes. Because the colour terms tiTK ,1fuy ,113* follow the
-C- T T T
pattern pa ol we are relatively certain, even without previous
lexical or contextual information, that they are nominals, for
the membership of verbal lexemes in a category that follows the
same pattern is restricted to a small group - tO* ,1A> ."JOp
'yya - of the Qal formation. Hence, at least a certain amount
of information - the syntactical 'meaning', the clue to the
possible ways according to which a given lexeme might function
in any sentence - is inherent in every form. Secondly, shades
of meaning might also be predicted from the formation of the
word: verbs in the Hif. formation might have a causative meaning
(\2^\, Dan. 11:35, if this is a Hif. form), or that of 'enter
a state', or 'become' (.l^PSttfl - Joel 1:7; .i:P3^ - Isa. 1:18;
•:' •i 2
l*:it>N - Ps. 51:9). It should be stated, then, that 'full'
Hebrew lexemes - be they verbs or nominals - are partially
predictable from the meaning of their parts (pattern). Therefore,
criterion a. of the Berlin-Kay list should be modified to
exclude structural or 'pattern' meaning from the determination
3
of monolexemic status for primary colour terms.
Criterion b. should be brought to its ultimate conclusion.
A primary term functions as a 'blanket' term of a wide and
general scope. While its full range of reference is not
included in other terms, segments of the range of sense and
reference it occupies can be specified by secondary or tertiary
terms.
The wording of criterion c. should be emended on semantic,
although not biblical, grounds. A term cannot be applied to an
'object'. In other words, a term is primary if it would
potentially enter syntagmatic relations with a varied and
relatively unrestricted amount of lexical items.
Criterion d. is of no real use for the research of an
4
ancient, 'dead' language. Instead, the frequency of appearance
of a term, together with its distribution within texts of
5
various types, should be examined cautiously. In the absence
of information gleaned directly from the speech community, the
picture the texts convey may be distorted and incomplete: our
sources are limited, and the motivation for preserving them was
6
not influenced by linguistic considerations. Lastly, it must
be remembered that Berlin and Kay investigated synchronic
42 BIBLICAL COLOUR TERMS

linguistic states, whereas the biblical material requires a

judicious mixture of synchronistic, diachronistic, and
panchronistic approaches.
To summarize: a colour term will be categorized as primary
when, upon analysis of its occurrences, it is found to:
a. be semantically monolexemic (although not necessarily
so from the morphological, structural aspect).
b. function as a 'blanket' term denoting a wide range of
referents.
c. be applicable, or potentially applicable, to a varied
amount of word signs of various types.
A primary term has to comply with all three criteria; in
cases of doubt, additional criteria (e-h) can be employed to
clarify the structural status of the term in question within its
set.
2. Secondary Terms
Next in the structure of the field come the secondary terms.
The criteria used for determining which colour terms are of a
secondary, and sometimes supporting nature are:
a. When a term is monolexemic but does not comply with
criterion l.b (unlimited signification) and/or I.e. (unrestricted
specification) - it will be classified as secondary. Thus VBK*
(Zech. 6:3,7) or pity* (Zech. 1:8) cannot be considered 'primary',
b. When a term is morphologically derived from another
term which has already been established as 'primary' (in
accordance with criteria l.a, l.b, and l.c above) and is
transparently connected to the latter, it is classified as
'secondary'. (Thus D2mx<0'TK ,dnKtt from the non-attested O-TN*).
c. The signification of the secondary term is included in
that of a primary term. Combining criteria 2.b above and this
one, OTOTK must be included within the d*1K sector and IhlhtS*
-, T «-;
within int», quite apart from the problem of their particular
references within their respective sectors. If 'snow' (Isa. 1:18)
and 'milk' (Gen. 49:12) are used for specifying 1^, then
U^nn ... -ins / At>W» ... .1ST (Lam. 4:7) might function - within
TT«» - VV ~ 7
that particular comparative structure - as specifications of 15t>.
Hence, the reference of a secondary term is a segment of that of
the primary term to which it is subordinated genetically
CLASSIFICATION AND DESCRIPTION 43

(DnmK<DTK) and/or semantically (.IDT ,.1hV to 13^ in Lam. 4:7).

T - - TT
d. The distributional potential of a secondary term is more
restricted than that of a primary one: it is found in less varied
types of syntagmatic relations and its linguistic contexts are
restricted. plpl* defines VjO (Lev. 13:49, 14:37) and qualifies
V.lin in \nin plpl* (Ps. 68:14). It is applied to lexemes that
refer to deterioration of materials (cloth or leather - Lev. 13:49;
a building - 14:37) on the one hand, to a type of gold on the
other hand. The restricted usage could have been considered a
decisive argument for classifying plpl* as a 'tertiary1 term.
However, the weight of cumulative evidence - together with
criteria b. (derivation, whether directly from pll* or from pi*
through analogy with DTQIK) and c. (inclusiveness) - is in favour
o
of placing the term in the secondary strata of its sector.
This technique - the employment of multiple criteria - minimizes
the danger of drawing too hasty conclusions from factors of
frequency and distribution. Conversely, as long as the term
appears in at least two different linguistic contexts, the need
to resort to arguments of accidence and possible artibrariness
of biblical stock is checked somewhat. Because secondary colour
terms are more restricted and specific in denotation, they tend
to be numerous: primary terms have no complete, or nearly
complete, synonym that can be substituted for them in all their
applications, but secondary terms either divide the territory
of their superordinate term among themselves or share it with
9
tertiary terms.

3. Tertiary Terms
The terms which constitute the next layer of the field are
labelled 'tertiary' in order to convey their structural relation-
ship to the first two, and higher, categories. This label,
however, is recognized as not fully satisfactory because it does
not define the class adequately. As the members of this group
are a motley collection of lexemes of diverse semantic and
morphological status, they are difficult to categorize decisively.
Possible criteria, or rather guidelines, which apply to this
group are:
a. A tertiary colour term is a relatively rare term. Any
hapax could be a natural candidate for such a classification -
44 BIBLICAL COLOUR TERMS

cf. 1h2*, Jud. 5:10 - unless other criteria override the argument
of infrequency, in this instance the related proper name IHif,
which is the name of three male persons from different periods
and descent (Gen. 23:8, 25:9, 46:10; Ex. 6:15; and 1 Ch. 4:7,
K "imp, Q insi) and the doubtful nn* (Ezek. 27:18). The related
" A * •
forms might have a bearing on the classification of Ihif* as a
secondary rather than a tertiary term. In other words ^i?*, like
other rare terms, is a borderline case. At the same time it
should be noted that in such extreme cases the argument of rarity
is more important than monolexemic status.
b. Compounds such as the English 'blue-green' or 'greenish-
blue' are categorized as tertiary, especially when one of their
components (or both) is (are) a secondary term. QitmK "p^*
(Lev. 13:19, 24, 42, 43) is a combination of a primary (l^t?)
and a secondary (QTQTK) term, both of which belong to separate
sectors of the colour field. tWOTK 13^ itself is juxtaposed to
"pt? (v. 24) when the latter appears on its own in the same
context of 'skin lesions'. Hence, dtmK "p!? has a specific
denotation which is expressed by a combination of lexemes that
are otherwise placed higher up in the hierarchy of colour
terminology. It follows, then, that OTttTN "p!? is a subordinate
both of "pt> (primary) and ofti*TO*TK(secondary).
c. The signification of the tertiary term is limited and
its application restricted, '^'^h. (Gen. 49:12) is defined as a
possible colour term by its parallel 3^hQ d^tJW 1^. As it is
used only in conjunction with tPJ'y and occurs only once as
">!7>b:>n and once in the derived form JVtWDtl (Prov. 23:29, again
in the same linguistic context), it is listed as tertiary on the
grounds of criteria a. (rarity), and c. (restricted scope).
Quite often the demarcation line between secondary and
tertiary terms is quite elusive, mainly because of the limited
volume of OT literature and the huge amount of hapax legomena.
The actual instances in the texts are sometimes not sufficient
for placing any given term within the proposed structural
framework. In such cases of doubt (see 1tl2* above) the relevant
forms are cited and discussed under either the 'secondary' or
the 'tertiary1 headings, and the doubts concerning them duly
noted.
CLASSIFICATION AND DESCRIPTION 45

4. 'Man Made' Colours: Painting and colouring, paints, dyes,

dyed materials, and pigments.

Terms for the activities of painting and dyeing are

extremely rare in biblical Hebrew. They are treated here
because: 1) v'Hin, y/nwn, and \fyzx* (not attested as a verb in
the Hebrew OT) are general in application, and hence cannot be
included under the heading of a colour reference in any level of
the field; and 2) Other terms, although more specific in colour
reference (*nW3 ... tnoi, Dt. 27:2, 4), are quite rare. Each
term will be discussed separately. Structuralization will not
be attempted because of the scantiness, the low frequency, and
the doubtful interpretation of the relevant lexemes.
In regard to terms for dyed materials, dyes, and pigments,
the situation is different. A few basic assumptions concerning
the classification of this category are given below.
a. Some of the lexemes referring to 'dyes' and 'paints'
have a dual sense. They refer to the dye or paint itself, the
colour properties of which are its most significant character-
istic, but also to the dyed material, the end product of the
process of using the dye or paint. This is especially true in
the case of textiles.
b. Occasionally one or the other of the names acquires a
wider referential meaning, especially in poetic diction ("OW in
Isa. 1:18). Broadly speaking, these names of dyes and paints
are potentially secondary or tertiary colour terms.
c. The existence of these terms in the language is
generated, by (non-linguistic) cultural influence to a greater
degree than in the case of other strata of the field. Like other
technology words' they were often borrowed, together with the
technique for preparing the product they denote, from the
language of the people who made the product known to the Hebrew
speaking community. Thus the terms often transcend linguistic
boundaries, while the product they denote travels across geo-
political borders.
d. Terms denoting paints, dyes and dyed materials are
monolexemic. They might function as subordinates within the
referential sphere of a primary colour term (*3ty under DTK) .
Their application is restricted - 1XU1K and other lexemes which
46 BIBLICAL COLOUR TERMS

refer to colourful textiles will mostly be found in the environ-

ment of words for garments and cloth, while the occurrences in
Song. 4:3, 7:6 are special cases of poetical emotive usage. As
terms for textiles are sometimes employed to convey colour proper,
and because colour properties are significant for the distinction
of the concepts these terms refer to - they are considered colour
terms of a unique type. Because of their restricted applicability
and specific nature, their place in the field hierarchy is below
the first three groups (unless they are not used to denote dyes
or dyed materials, as in Isa. 1:18, Song. 4:3, 7:6). Because
non-linguistic evidence suggests that their references may be
included in primary colour terms - e.g. "OK/, 1KU1N, ^'Q^ as
types of DTK dyes - these lexemes will be internally arranged
under the DTK, "p^, and so on referential headings.
T TT
5. Indirect colour connotations
This group includes lexemes which are not colour terms, in
the sense that they do not refer directly to any colour quality
of the concept they denote. Nevertheless, they are associated
with colour notions through phonetic similarities or semantic
links.
a. These may be: Proper names which are phonetically
related either to established or to lost colour terms. This
group is subdivided into: 1. Personal names ("pt?).
2. Place names (|7lU) ^ru).
b. Some names for objects (or concepts) exhibit phonetic or
semantic links with a colour term (fl313j>, ?TD3h, CPS3). In some
cases (?CPif3) these words preserve a colour reference derived
from a term which might have once been a colour term in Hebrew
or in a cognate language, but which does not function as such in
biblical Hebrew.
c. Terms for metals and precious stones (D^TN, flp'ia, finoa).
Strictly speaking this is a sub-group of b; however, these lexemes
are often borrowed from a source language into another as a
result of the introduction of the object they name into a new
society. Therefore, I felt that they require a separate
category. Finally, while the motivation for indigenous Hebrew
words (such as DTK) is phonetically transparent, in order to
find the 'colour' lineage of borrowed terms (such as il*rca) one
CLASSIFICATION AND DESCRIPTION 47

should turn to etymological operations of the types A, B, C, and

E of Barr's classification (pp. 31 f. above).
d. Entities such as 1**, I1?!!, tfW, 1»S, 'B.10 are associated
. - TT
with colour notions through their most typical visible feature(s).
They are employed - with or without a direct colour term - for
communicating colour perception or colour imagery: thus we have
•n^* A^O / T»?T> 1B5O (Isa. 1:18); ^UD jn'^O (2 Ki. 5:27; and
T:
twice more the fern., >^3 mn>m); V»tt D^y >^Dh (Gen. 49:12);
and others. These descriptions of colour by way of a comparison
to a lexeme referring to an object of a well-known colour can
become a literary convention, even an idiom. Such an idiom may
replace a direct colour term, especially within the framework of
poetic diction - as, for instance, in Jer. 13:23, >K/1D ISD^M
1* 11*13*15'"' 1QJ1 Illy. Furthermore, these same terms might serve
as the base of proper colour terms at some later stage in the
development of the language. Gradwohl, for instance, argues for
10
regarding *p!? as, first and foremost, the colour of milk.
•TT
These lexemes supply us with additional information about colour
perception, information that can be as valuable as the information
that can be gleaned from direct terms. Therefore, these analogies
are fairly important for determining the referents of 'full*
colour words.
In the previous sections terms were arranged under the
heading of the primary term in which their reference is included,
insofar as this was possible. This principle of internal order
will be adhered to in this section too. Lexemes that relate to
the same root will appear together, unless one or more of them
belong to different categories.

6. Terms for speckles, stains, and multi-coloured phenomena.

These refer to the appearance of an area of colour, the
identity of which is not necessarily specified, against the
background of a different colour space. There is no colour
reference that could be employed as a common denominator for
all of them: rather, some (like Tpi*) seem to be linked more
transparently to the field of space and shape than to colour.
Where specific colour references seem to be denoted by the term
(Y12*?), they are difficult to define because of their relative
rarity and limited applicability. For want of better criteria
48 BIBLICAL COLOUR TERMS

the terms are arranged in two morphological groups :

a. pa°ol forms (Sl2*, TpJ*, *TpV*) •
T T T
b. Others: (N.lt>D*; perhaps qiioy* and lufrfc*; IttJ and its
derivaties; niT-l-n^h*; pKa; inna; OpV and nap/1.)
The two most important questions that are relevant to each
term within the b. group are: 1) Is it a colour term at all, or
is its reference more indicative of spatial than of colour
arrangements? 2} If - and when - a colour reference of some
sort is established, can we define the referent, so that the
term be assigned to its proper place under the suitable primary
heading?
PRIMARY TERMS 49

B. PRIMARY TERMS IN THE OT

0. 0(1)™, "p!?, and 1(1)hW (and their verbal derivatives) are
T T.r -r
the only lexemes that meet the triple criteria of (a) monolexemic
status; (b) exclusive signification, and (c) unrestricted
applicability. dTK appears in texts which are attributed to
stages 2, 3 and 4 of biblical Hebrew - that is, the monarchic
period, the exilic period, and the post exilic period. "p!?
possibly exists in all four stages: this depends on our dating,
on other grounds, of Gen. 49:12. "ihW is restricted to the last
two periods. Statistically, although all three are the most
frequent colour terms in the OT, their number is quite small: in
descending order, 13^ appears 24 times, its direct verbal
derivatives - two lexemes, a Hif. and Hit. - 5 times; DTK -
— T
9 times, its direct verbal derivatives - three lexemes, Qal, Hif.,
and Hithp., each of which is employed once - 3 times; Ihty -
2T
6 times, together with one occurrence of a Qal lexeme. In
addition, it must be noted that a sizeable proportion of the "p!?
occurrences - 14 times - is clustered in Lev. 13. Given these
data, we may suppose that "p!? and 0*TK are the most dominant colour
terms, while ItlKJ lags not far behind. This indeed is supported
by other data. DTK and "p^, or subordinate terms that belong to
their sectors, appear as oppositions (Isa. 1:18, Gen. 49:12);
• *
and the three terms are brought together (1flt!> through Tlhtftt {l^h)
in Lam. 4:7-8. DTK has the greatest number of colour specifica-
tions through references (by analogy) to substances whose colour
is well known - blood, wine, scarlet, and corals - and "p^ is
. 3 TT
compared to snow, milk, wool, and implicitly to n^Ta. 1hl9 is
compared to the less transparent (for us) llhiy, to a raven, and
to YTf? ^nK.
4
All three terms serve as bases for derived terms. The
latter, apart from direct (active) verbal formations which belong
to the primary stratum, are placed within other levels of the
field. When we use the criteria set in the previous chapter for
comparing the types and the number of these derivatives, the
resulting (achronistic) picture is:
50 BIBLICAL COLOUR TERMS

primary . dTK , "p!? iVlw'

T TT , r
secondary dWQ .dntiTK — .imhK>*
TT: ;—: :-:
tertiary .dT»*W "P^ ,*J10TK .dTOTK "pt>

Indirect a. TdlTK ?dTK .K/rmb .1^

... JT TT; TT
"T^'TN^ mt>h • ">?^
1. personal/generic names ?dlTN T^IJ)
2. place names .d^BTN nt>ytt .d'lTN .nJ^ . IDb
* *."'. '.*; T; • TT
?naiK ?im^> .n^i^
T:~ -•, • T :
Indirect b. , ^""r^
?dnT ?naiK
TT-•
?dtK
TT
.n^(i)nt?
T I
.njn!?
TT .
.iiniy"
C
names for objects or j?^ n:i?^* ?nnnw*
.n^^ .(vs&ji ?inV
concepts 4—'p^«-MJ!lb
-T T-" ;
?iat»n
Indirect c.
metals/precious stones abv

d*TK has five terms which seem to be directly derived from

it. One of these - the compound dTOTN "p!? - probably refers to
a non-linguistic entity it partly shares with 13^. The lexeme
TT
dTK
T itself, in the m.pl.
r word d»B"TN,
...-. is part of arplace name,
which brings the total up to six. ' In addition, six lexemes -
dTX, dinK, •'OIK, nOTK, n»TK, and dn - are phonetically (the
r
TT •;; • -; Tt" TT~ T
first five) and also referentially (chiefly d-T) so close to the
dTK concept as to merit an investigation of possible root
etymology links.
1^ has no secondary or tertiary terms which are derived
directly from it apart from the dTtiTK 1i^ which it shares with
a segment of the d*TK sector (see above). On the other hand, it
is the basis for eight personal, generic, and place names; four
terms for objects or concepts, and through the fifth - MJ^ -
perhaps the ultimate origin of v/1^7 II ('make bricks') and of
l^n. If we accept that "pt? = personal name is so different
in denotation from "p!? = place name as to warrant giving each of
them the status of an independent lexeme, then the total of the
derived lU^ terms - even if the v"p^ II series is excluded - is
TT
thirteen. It seems to me that this is the correct procedure to
follow: although "p!? (colour), ^^ (personal name), and 12^
TT TT * TT
(place name) share the same form - and a similar lexicographical
PRIMARY TERMS 51

problem is presented by series such as M33!? (moon), K/n3!l!j

(personal name); fiJIl!? (place name), ftte'y*, VDOfi JU3^;
i131!l^ (place name), fl3(1)3t> (frankincense) - each element in the
series refers to a different 'reality'; hence they are
homophonouSiperhaps etymologically identical but semantically
separate lexemes.
Finally, inty is the most limited: it has one secondary
derived term, one certain term for object or concept, and two
terms that are hapax legomena but seem to be related to it.
Thus, the total for Ihia is, at the most, four directly derived
lexemes.
To summarize this point: the existence of colour terms
that are employed as structural sub-components of different
functions within the area designated by the primary terms, when
the subordinate terms can be shown as morphological derivations
of the primary terms, strengthens the status of the latter as
primary governing elements within the same framework. From the
point of view of applicability, again 1^t> and OIK have the
widest range, while intU is more limited. Whereas secondary and
tertiary derivatives of tilK are more numerous than those of 1^!?,
T TT
the latter is more prolific in the area of proper names and in
applicability. Ihe;, although by far the most limited of the
three on all counts, retains its primary status during the
exilic and post exilic periods because its distribution and
range are wider than that of other, non-primary terms whose
applicability is narrowed to one lexeme denoting one object only
(like J71B* or VOX*). In the following table, the range of each
of the three terms is compared with that of the other two, again
regardless of chronological factors and ordering.
52 BIBLICAL COLOUR TERMS

Lexemes for items that are Number of occurrences

qualified by a primary term *p^ QTK IflW
I 'physical' :
a. human: 1) flesh, skin, 1 (nom.) 1 (nom.) 1 (nom.)
of complexion 1 (verb.) 1 (verb.)
2) skin tumor, 7 (NOM.)
sore
3) hair 7 (nom.) 3 (nom.)

4) teeth 1 (nom.)

b. animal (skin, hide,

wool):
5) sheep 1 (nom.)
6) cows 1 (nom.)

7) horses 3 (nom.) 3 (nom.) 2 (nom.)

c. inanimate, food, drink:
8) lentils 2 (nom.)
(Gen. 25:30,
x 2)
Ti\ yyft 1 (nom.)
water 1 (nom.)
wine 1 (nom.)

9) garments 1 (nom.) 1 (nom.)

10) others: 2 (nom.)
wood, bark 1 (verb.)

II 'abstract1:
11) sins 1 (verb.) 1 (verb.)
6
12) sinners 3 (verb.)

TOTAL 29 12 7
PRIMARY TERMS 53

Notably, the three primary terms - CHN ,"\^ ,1I"H2J - are all
applicable to lexemes denoting 'human skin' or 'complexion', and
'horses'; "p^7 and DTK also share the description of sins and of
garments; IhB and ll!> - of hair, while the third term in this set
T T~f
of oppositions is derived from DTK, but is of secondary status
(^IttTN, Gen. 25:25 and 1 Sam. 16:12, 17:42.7) These recurrent
appearances of the same colour term within the same area of usage
strengthen their status beyond, and apart from, the recognition
of their versatility. Further, the wider scope of 'p1' seems to
be of limited significance: while in the tHN sector we find a
number of dTK - derived subordinate terms (apart from personal
names or toponyms), in the "pt> sector terms that are etymologi-
TT g
cally non-related to the chief lexeme are brought in. As
Fronzaroli points out, although the number of associations and
possible substitutions in the corpus is limited, the series is
9
probably a 'closed' one in the actual living language. A more
detailed description of possible sets of oppositions within
sectors of colour references and across them will be given under
the headings of the relevant terms, whether primary or otherwise.
10 11
0.1 Both Gradwohl and Fronzaroli classify £fl^ as a sub-
member of the DTK sector. At first glance their statements seem
to be correct. In the existing corpus £fW is restricted to
Lev. 13, presumably of the exilic or post exilic period, and
its frequency is limited (three occurrences - vv. 30, 32, 36).
From the point of view of versatility and applicability it is
associated only with 'hair'. Its etymological relationship with
3MT is far from proven; 12 therefore, the phonetic similarity
cannot be a guide for defining the actual (extra-linguistic)
reference of the term on etymological grounds.
On the other hand, various arguments can be cited in favour
of including 3fl^ in the primary set of colour terms of the last
two stages of biblical Hebrew. While it is applicable to
'(human) hair' only, it stands in opposition to - and in its
situational context can be substituted for - either synchronic
primary terms associated with hair (IHW, vv. 31, 37;,or
potentially with terms like "p!? 1JH9 - vv. 10, 25), or diachroni-
cally with *31}3*TK, if the latter is to be interpreted as denoting
13
hair, and not complexion, colour. Following the criterion of
• * •
form, 3MS shares the formative pattern ofti*TN,IhW, and other
T -r T
54 BIBLICAL COLOUR TERMS

colour terms. Whereas this last obvious observation is not

particularly weighty as an argument - the same pattern is shared
by subordinate terms like VON* , Ylfl* ,p1W* as well as by non-
-r T -r
colour terms like pny ,Pftn - it does contribute towards the
T T
recognition of the status of 3tl^ as a primary term, an equal
member in the set ll^-inw-Dri^ fas applied to a substantive
TT T T
denoting human hair]. Etymologically, a connection or a starting
point might be gleaned from the partial phonetic similarity,
within Hebrew stock, of /arw and the /5JT^ ,1/IHS groups and their
respective (extra linguistic) references, rather than from a
presumed sound shift *d>s, which is unusual for Hebrew. UMi!
also has a derived subordinate term, ^J3^3 (Ezra 8:27). Although
14 '•
this hapax legomenon is dated as synchronically parallel to
the afiif occurrences, it describes bronze or copper vessels, not
hair; thus its range, although limited to one association, is
different from that of the parent term. Finally, the reference
of the term, although not self-defined by the text itself, is
understood as belonging to the tHN group by the Aramaic
translations, but not by the LXX and the Vulgate. Even more
significant is the fact that in MH 3!"l!i has an independent
reference that may overlap with 0*TN at their common border (as
indeed they should). Basically, though, the two terms are
distinct from each other, independent, and of almost equal
status. All the points mentioned briefly above will be taken
up again when ^M^ is discussed later in this chapter, and in
the appropriate place in the section dealing with colour terms
in MH. Meanwhile, the conclusion adopted here is: Whereas in
the period of the monarchy (Stage 2) there are only two extinct
primary terms which can be defined as such by the multiple-
criteria method employed (135 ,tiTN), both IhW and 3Mif join the
TT T "T T
basic system at the end of this period and feature as primary
terms from then onwards.

0.2 Another term with a dubious status is pTT>. It appears

only once - uh"P plT> 53 "IhNl (Job 39:8) - and as a substantive.
Its derivation from PT or pV seems to be quite straightforward,
for it is still transparently associated with the name of the
object which has the characteristic feature of 'green-ness'
denoted by plT>. On the other hand, one must reckon with
PRIMARY TERMS 55

derived lexemes on various levels. The secondary formation P1|7T4

which is associated with cloth and leather 'disease' (Lev. 13:49)
and building material 'disease' (14:37) is defined as a secondary
form derived from pll* - not directly from P^./PT - by its
analogy and opposition to QWJK (from DTK). The lexeme is
further employed in a completely different context in the syntagm
Vllh p1p*P (Ps. 68:14). The last instance is highly significant:
it has an almost exact parallel in Ugaritic - 1 Krt. 126: yrg hrs.
- but there with what looks like a primary, not secondary, form.
The antiquity of the roots of Ps. 68 makes p*lpT> less dependent
on DTttTK (by way of analogical formation) and at the same time
enhances the possibility that p1V featured as a primary term in
biblical Hebrew, and that its being a late hapax is perhaps
merely accidental. Other \/pT derivatives are of non-direct
types: the toponym llpl^Cn •»») and possibly 1*1 P"l». The latter
has two significations:' as a name for a cereal disease it is well
distributed throughout the last 3 stages of bib. Hebrew (5 times
in the pair lip^l 113TO), while in Jer. 30:6 (d>33 ttt I^QiUI
llpl*!?) it either represents a shift in meaning - not only
application - or else this is the basic sense of the lexeme. To
be sure, llpl*. and IIPTJ. could have been directly derived from
pV; their usefulness as evidence for the status of p1T> is
limited, and the admissability of the latter depends upon the
cumulative momentum of other data. These other data fall under
three headings: 1) The early usage of p'lpT and its sense within
its context; 2) The ample evidence of ancient cognates of the
same consonantal skeleton as colour terms in their respective
languages (Akkadian, Ugaritic, Arabic); and 3) The wide employment
of PIT and its derivatives, old and new, in MH - that is, the
amplification, along newer but similar lines, of older stock.
These three aspects will be discussed later under pll* ,p1pT
15
in the appropriate sections.
Meanwhile let us just say that Fronzaroli's suggestion -
that p"lpT> existed in the sub-system with no archetypal p1T> in
16
the basic stratum - seems to be undermined by all three
arguments; hence, pIV will be considered as a late (last two
stages; but possibly, after a re-clarification of VTltl P1PV
and the literary and chronological context of Ps. 68, even
earlier) primary term, alongside iHJf ,tiTK ,13^ ,*ihW.
T T TT T
56 BIBLICAL COLOUR TERMS

0.3 As reported in the Introduction, Berlin and Kay found seven

universal stages in the evolution of basic (= primary) colour
terms. In essence, a synchronous linguistic state may contain
lexemes that will place it in any one of the following stages of
lexical colour development:
Stage I. lexemes for 'black' and 'white1; or
Stage II. lexemes for 'black', 'white', and 'red'; or
Stage Ilia, lexemes for 'black', 'white', 'red1, and 'green'; or
Stage Illb. lexemes for 'black', 'white1, 'red', and 'yellow'; or
Stage IV. lexemes for 'black', 'white', 'red', 'green', and
'yellow'; or
Stage V. lexemes for 'black', 'white', 'red', 'green',
'yellow', and 'blue'; or
Stage VI. lexemes for 'black', 'white', 'red', 'green',
'yellow', 'blue1, and 'brown'; or
Stage VII. Stage VI lexemes plus either 'purple' or 'pink',
'orange', 'grey', or all of them.
According to this scheme, the visible colour factors in the
non-linguistic world are all referred to by the existing terms:
in stage I 'black' includes all dark colours, 'white' - all
light ones; in stage II 'red' includes most highly saturated
colours, and by its existence makes the area previously covered
17
by 'black' and 'white' smaller but better defined; and so on.
Let us apply these conclusions to the colour systems of
various historical stages of biblical Hebrew. So far, we have
found that:
- in stage 1 (pre-monarchic Hebrew) we have no material in the
corpus to support any system of colour discrimination.
- in stage 2 (pre-exilic) there are two terms, DIN and "pt>,
T T-r
which roughly correspond to 'white' and 'red'.
- in stages 3 and 4 (exilic and post-exilic) there are three
established terms - ]3fo ,DTK ,irny
T
(roughly corresponding to
TT T
1
'white', 'red , and 'black' respectively), and two 'young'
terms - (71 T> ,5>\X (roughly corresponding to 'green' and
~r T
'yellow').
It seems, therefore, that by the end of the OT period the
primary colour vocabulary of biblical Hebrew was at the beginn-
ing of stage IV of the Berlin-Kay scheme; possibly, if we insist
that 3M2 is peculiar to Lev. 13 and that arC&J (Ezr. 8:27) should
PRIMARY TERMS 57

be discounted because of the difficulties it raises, we should

classify the evolutionary status at this point as a transitional
link between stages III and IV. The material available is too
scanty to make any firmer assertion. The situation is much more
obscure as far as Stage 2 (Hebrew of the monarchic period) is
discussed. Not only does it not correspond to the Berlin-Kay
stages I or II but, which is much more important, it does not
tally with basic common sense observations: light and darkness,
'black' and 'white' are a basic contrast in reality; it seems
inconceivable that any linguistic system will ignore this
18
contrast. Fronzaroli suggests that in earlier (stages 1 and 2)
• • 19
Hebrew the term Qin filled the slot later occupied by inu/.
This is possible but cannot be proven from the text. Alternately
we might surmise that "intf only accidentally does not appear in
any text prior to the exilic period, or that a lexeme from the
'darkness' field - perhaps a formation from y/IBh or /YTp - was
the primary term later supplanted by IhBJ. At any rate, in the
interest of common sense we should presume that in stage 2
Hebrew lexical colour categories correspond to stage II of the
Berlin-Kay set: it has the primary established terms - DTK ,1^ -
T TT
that are sometimes juxtaposed against each other (Isa. 1:18),
and a third term that is the equivalent of 'black', but which
either lost its primary sense early and was not preserved as
such in the text (in the case of Din), or is simply not extant
in the corpus.
0.4 So far in this section we have discussed the primary terms
from the aspect of their sense relations on the primary level
itself and across levels and sectors. In the following sections
the occurrences of each primary term will be analyzed in an
attempt to find what they stand for. In other words, we will
now look for the extra-linguistic references of the linguistic
symbols. The three chief problems this investigation will be
concerned with are: a) What is the reference of the term in
each range? b) Adding the total of these references, what is
the colour range that the term covers? c) Is the term a 'primary'
one in the referential, and not only the linguistic, sense? Do
the results emanating from the investigation of the references of
any term justify, or reinforce, its classification as primary on
)
the basis of interlinguistic analysis?
58 BIBLICAL COLOUR TERMS

B.I. OCOTK 1 .
T
1.0 In most biblical dictionaries, translations, and commentaries
d*TN is conveniently equated with 'red1, while the verb formations
T
derived from it are interpreted as references to one aspect or
the other of becoming 'red'. A quick survey will illustrate this
point. BOB 2 defines QTK
• as 'red, ruddy'. The entries in KB 3 and
T 5
in Klpnn TIP!? 1S1K are similar. The Jewish Aramaic translations
invariably render DTK with lexemes derived from the root pfcb, and
this seems to strengthen the notion of a fixed, or well-defined,
reference of DTK. The LXX and V, though, do not use one consis-
*r £
tent term for all instances. A possible reason for this practice
might have been the recognition that DTK does not necessarily
have a constant equivalent in Greek or in Latin. (We will return
to this point below, in 1.4). In other words, the referents, as
denoted by each and every context, are variables within a
framework that is too wide to be communicated by a single common
lexeme. The concept of a one-to-one relationship between terms
referring to the same extra-linguistic colour phenomenon at first
seems to be reasonable. If the phenomenon is constant, it should
be perceived by most viewers in a similar manner. Consequently,
it is to be expected that different speech communities will encode
it in a similar fashion; that is, will coin a term that would
'cover1 the same - or a similar - area of reference, displaying
the same - or similar - boundaries. That this concept is not
always valid will be apparent as soon as we begin to analyze our
first instance (Gen. 25:30). The difficulties inherent in this
orderly - albeit not necessarily adequate - approach are far
greater than its comforts, and the resulting attempt of defining
the extra-linguistic references is sometimes confused. Thus
Gradwohl, who discusses tJTK under the heading 'rot', is forced
to summarize that the term covers 'brown', as much as different
7
types of what we call 'red' and even 'pink'. Therefore his
heading, as much as his organization of the material, are highly
misleading: our modern 'red1 is more restricted in scope than the
biblical &*TN. Furthermore, as Berlin and Kay point out,
"Whenever we speak of colour categories, we refer to the foci of
categories rather than to their boundaries or total area, except
8
when specifically stating otherwise". Berlin - Kay were unable
to explain this process which was nevertheless empirically proven
PRIMARY TERMS 59

by them. However, it seems that their position should be

accepted: we do tend to use lexical definitions which refer to
the typical (= exaggerated) essence of what we describe, for
communication might be impaired otherwise. Now, when we equate
tHN with 'red', and our 'red' is perceived as referring to an
area which has the colour of blood as its focus, what is easier
than to attribute the same focus - with the same frame of
reference - to biblical DTK? The consonantal similarity between
4 ~
QTK and OH appears doubly tempting in the light of the presumed
~f T
focus of the field, and attempts are made to find an etymological
9
connection between the two. Further, biblical passages, as
will be shown in the discussion of Gen. 25:30, are interpreted
in the light of the 'blood-red' focus supposition; but when we
do that, do we not read into the text our own preconceptions,
attributing to it more - or less - than what was originally meant?

1.1 Gen. 25:30 MTil dTKh OIKM 1» NJ 'OO'y^n

TT TTt
The double occurrence of DTKil presents both grammatical -
syntactical and textual problems, and those should be cleared up
before we approach the referential question. Most commentators
propose to emend the text by eliminating the firstti*TKftand
10 11
substituting something else for it: BH , Skinner, Dillmann,
12 '
and Caster suggest an emendation in the pointing to read DTK
(Cf. Arabic ^dam) = 'seasoning or condiment for bread' on the
grounds of unnecessary, even awkward, repetition. Others, like
Ball,13 read 'DlNtl man 1 , following the LXX, V, T, and Neophyti I.
Ball says, "There is no special reason for repeating the epithet,
while on the other hand, a substantive is necessary to sense, and
14
even to grammar". To his first argument we may object on the
grounds that Esau is depicted as being in an impatient, excited
state, and thus might have easily been described as repeating -
15 0
his words. Furthermore, because the T (as in both Sperber
16
and ni^lTA tllXIpn ) contains the same repetition, a simple
accident of dittography should be considered less plausible. As
for the LXX, V, and Neophyti, where the equivalent of T>TJ
replaces the first DTK!"), it might be not so much a part of their
Vorlage as an attempt to correct the apparently unnecessary
repetition.
Finally, as Blau says, the borderlines between noun and
17
adjective in Semitic languages are quite fluid: there is no
60 BIBLICAL COLOUR TERMS

reasoi> for emendation on grammatical grounds - the utterance

makes sense as it stands. On the other hand, any one of the
proposed emendations would rob Esau's utterance of the sense of
urgency and the oblique, too-tired-to-care attitude. Let us
leave it as it appears in the MT.
The reference of dTKM DIKM depends, of course, on the
TT TT
referents of "PtJ (vv. 29, 34) and to a greater extent on d'ETry
•T -T-;
(v. 34), for this 0»Bn)7 T»n is the entity referred to by the
colour term. T^TJ - like its related Hif. formations - is well
.T
established through the occurrences here, in 2 Ki. 4: 38, 39
(where it clearly designates 'something cooked in a pot'), and
in Hag. 2:12 (where it again appears in conjunction with ont») .
So, the translation 'dish', 'pottage' is justified - the Aramaic
Targum consistently uses i/^KO - although the 'soup', or "blood
18 19 20
soup', referred to by Daube, Henton Davies, and von Rad
21
looks less accurate than the 'pottage', 'dish' or simply 'stuff
which other scholars mention.
The second part of the syntagm - Q">uny - is universally
T
22 •
recognized as 'lentils', in OT language as much as in post-
23
biblical Hebrew. The consistent Aramaic translation "ptlOlbO,
and the Arabic cognate of the same meaning, strengthen this
recognition. It is true that the version parallel to (2 Sam.23:11)
O'Jmy MK^n flTEKl npt>n has instead (1 Ch. 11:13) ilK^tt fVTOn flp^h
O'TiyBf. However, as both LXX and the T haveti">vnyin both places,
and the meaning of the word is well known in MH, there is no
reason to attach any significance to what looks like an error in
1 Ch. 11:13. So, what colour are lentils? The question can be
asked only in the present tense: advanced agricultural methods
change many aspects of any harvest, but we have no choice apart
from relying on our own experience if - as in our case - no
additional information is available. Gradwohl describes the
colour of lentils as 'yellowish-brown', but adds that they are
24
perceived as 'red'. As his statement stands it is inexplicable.
Let us add that, at least one type of lentil one gets in present-
day Israel, (and, for that matter, in England as well) is best
described as of a sharp orange colour which turns brownish after
the lentils are cooked. Nobody today would consider the colour
of these lentils 'red', for other terms are available. In the
absence of contemporary terms for either 'orange' or 'yellow1 or
PRIMARY TERMS 61

'brown', most of us would indeed define the colour of lentils as

'red'. However, this (for us) hypothetical practice would not
mean that we equate the colour of lentils with that of blood, or
wine, or human complexion; on the contrary - it would mean that
the term 'red' would be an expanded 'blanket' term which is
inclusive of many more references to chromatic shades than those
we attribute to our modern 'red', or the DTK of contemporary
Hebrew. In other words, dTK in Gen. 25:30 has a reference value
of 'orange', 'yellowish orange1, or 'yellowish brown'. What
should be done, then, is not to bring our sensibilities to the
biblical text and declare that OTK is 'red1, but rather to accept
that the scope of the term indeed includes 'red1 (although not in
this context), although here the reference is to its possible
boundaries on the 'brown-yellow'side. While ti"TK is indeed 'red'
T
it is also wider in range than 'red', covering areas that were
later excluded from this sector as a result of the introduction
of new terms.
18 19 20
Daube, Henton Davies, and von Rad raise the possibility
that the Q">vny TT3 was mistaken by Esau for 'blood soup'.
• T~t ':
According to this theory Jacob cooked the lentils in such a way
that they looked like a red blood soup, not only appetizing but a
source of life and potency to whoever drinks it. Thus Jacob
tricked Esau through the colour of the soup; and as his brother
forswore before discovering the deceit, he has no reason for
complaint apart from this trickery.
The theory is attractive on a few counts: it enriches the
tale with folkloristic elements of the advantage of drinking
blood to the person who consumes it. It explains Esau's hurt
after eating the pottage, whereas if no direct allusion to a
blood soup that is not really what it seems is presumed, he really
cannot blame Jacob too much: the latter exercised an advantage
over him but did not force his hand in any way. At a pinch, this
can also clarify Esau's contention that he had been deceived
twice.
However, the whole theory rests on the notion that QTK is
equal to 'red', whose focus is - or might be - 'bipod-red',
and/or on an etymological connection between d*TK and OH. As
discussed above, these two arguments are far from established.
If we bring common sense to bear upon the question we will
62 BIBLICAL COLOUR TERMS

discover that making a lentil dish resemble 'blood soup' through

simple cooking is a far from easy feat, for lentils tend to lose
their distinct colour and become a faded brown or yellow through
cooking. Therefore the 'blood soup' theory, together with its
enriching symbolism, should be regretfully discarded.
Finally, a word about the 'DTK' element in Gen. 25 in
general: it is obvious that DTK, ^i'lQTK^nK^are phonetically
and perhaps even etymological ly related, and that the 'DTK'
element and its derivatives form an important element of the
25
etiological story of Jacob vs. Esau, Israel vs. Edom.
Nevertheless, it must be remembered that as far as the listener
is concerned the paranomasia of the repeated consonantal sequences
is probably as important as the colour allusions they possess, or
do not possess. There is no reason to suppose that the listener
or reader immediately asked himself why Edom is thus named
(although the story-teller did). As the word stands, it has no
colour denotation whatsoever. On the other hand, DTK and ^JIBTK
can play their part in the grotesque presentation of Esau only if
the speaker and his audience know that their colour references
are quite loose, and hence they are considered to be constituents
of the same general framework.

1.2 Num. 19:2 no»ntt n&TK HIO

T • ; TV-J TT
No natural cow hide is indeed 'red'. If we stick to the
DTK = 'red' formula, we shall be in trouble here too. On the
T
other hand, the retention of the DTK = 'red' definition would
introduce the question of symbolism, especially that of blood and
of cleansing or purification, which is particularly appropriate
here.
Gradwohl tries to side-step the question by saying that the
unfortunate heifer is 'reddish-brown', and therefore rare. He
cites Mish. Par. 3:5, where it is said that even two black or
white hairs make the heifer unsuitable for its destiny, and
claims that the rarity of such an animal is recognized there.
He adds that according to the above-mentioned source, Moses and
*}£i
Joshua offered only one such 'red' cow for that purpose. Upon
investigation, this rendering of the Mishnaic text appears worse
than inaccurate. Joshua is not mentioned at all. Not the rarity
of the 'red cow' is discussed, but that of the ceremony prescribed
in Num. 19:1-21. There is agreement about the information that
PRIMARY TERMS 63

from Moses to Ezra the ceremony was never enacted; on the other
hand, there follows a discussion about the number of the 'red1
heifers offered from Ezra to the destruction of the Second
Temple. Therefore, there is no basis in our Mishna for the
notion of the 'rarity1 of the 'red' heifer as such (and see
below). In actual fact, Gradwohl does not give up the 'red'
definition, but attempts to modify it somewhat, while misreading
the Mishna along the way.
Even Snaith, who recognizes that different shades of 'red1
might be designated by DTK, retains the traditional 'red heifer'
T
27
translation. Indeed, he is forced to do so because of the
importance he attributes to the 'red' element (relating to blood
28
rather than fire ) as a cleansing, desinning agent. Naturally,
if 'red' is DTK then the origin and visual symbolism of the
ritual would be lost if the heifer is discovered to be 'non-red'.
Snaith evades the problem by declaring that it is not always
possible to be precise in matters of colour - which is perfectly
true - and by citing the Mishna (mentioned above) concerning the
rarity of such an animal.
G.B. Gray admits that "no unnatural colour is intended; for
though the word DTK at times denotes a brilliant red colour (as
of blood), it is also used where we should rather speak of a
brown, or reddish brown". 29 So far so good. But Gray is looking
for the symbolism of the rite too, so he seems to retreat to a
safe "Why the cow had to be red is uncertain", and suggests the
30
colour of 'blood1, or of 'ruddy, golden corn'.
Clearly, the cow cannot be 'red'. Whether its skin has a
reddish sheen or not is beside the point: today we would probably
term it 'bay' or 'brown' in English, d^n (= *brown') in Modern
Hebrew. There is no contemporaneous term referring to 'brown' in
OT Hebrew. In the absence of a specialized term that refers to
the chromatically neutral area between 'red' and 'black' (or
'dark'), this territory is divided between the two polarized
terms. (Berlin and Kay refer to this phenomenon when they
describe stage II, e.g. three-term languages: the term for red
in this stage "includes all reds, oranges, most yellows, browns,
pinks and purples (including violet)" ). We cannot argue that
the lack of a specifying term for 'brown' points to a lack of
its identification as a specific entity: as there are no 'red'
64 BIBLICAL COLOUR TERMS

cows or horses, when the term is applied to lexemes denoting

these animals it refers to their visible, 'real' colour, inasmuch
as when applied to 'blood' it means 'blood red'. As a matter of
fact, the system works because of the versatility of range the
terms possess. And when, for some specific reason, a term ceases
to be adequate, the new term introduced infringes on the territory
of the old one (as with tHN and DW7N, or dTK and SHS) .
^ -:"T T T
As for Mishna Par. 2:5, one must realize that it is bound
by the wording of the biblical text (na'TK),
T
although the actual
• X"'
frame of reference for DTK in MH is much more restricted because
T
colour terminology in MH is much better developed. Hence the
extra-biblical legends about the rarity of the 'red heifer' - it
can be found, but is not common. 32 Underlying this explanation
is another one stressing the qualifying iin>ttfi which, as Gradwohl
rightly points out, is understood to refer to the purity of the
colour, not the perfect body of the cow. 33 The Sages must have
realized that the MHTK fl13 mentioned actually refers to a 'brown
TV: TT
cow', not a 'red' one; and this type of cow must have been known,
if not common-brown locally bred or 'Arab' cattle, as it is
colloquially called, is still abundant in Israel today; can this
fact assist in the recognition that such cows were not so rare
after all? The problem facing them was: In what way would this
apparently quite regular animal be so special as to warrant its
destiny? The answer was probably born out of the observation
that whereas brown cows are not a rarity, one that is completely
uniform in its colouring - without speckles or spots of white or
black, without even a couple of neighbouring black or white hairs
(Par. 2:5) - is rare indeed. Thus again, na'TK refers to a
'brown' or, at the most, a 'brown' colour with a 'reddish' or
'yellowish' tinge. Unless a footnote to that effect is appended
to the text, a translation utilizing the lexeme 'red' - with its
modern limitations - is not only inadequate, but also misleading.
Finally there remains the question of the colour symbolism
(if any) of the rite described in Num. 19:1-10. It is conceivable
that animals with reddish or brownish hide have been used for
ritual purposes by various communities because of the resemblance
of their hide to blood.34 In addition, normal linguistic usage
points to the fact that 'brown', 'orange', etc. are included in
the sector covered by the term 'red' before specialized terms
PRIMARY TERMS 65

are assigned to them, together with 'blood red1 and other 'reds'.
But, whether there is a colour symbolism in our passage or not
cannot be decided on the strength of external evidence alone.
Within our text blood is indeed used (v.4) and so is fire (v.5).
However, there is no hint, no clue in the text in regard to the
colour significance of the heifer's skin.
Were the legislator, and/or the writer, aware of the 'hide
colour'-#'blood1 associations? Therefore it would seem, on both
semantic and interpretative grounds, that no colour symbolism
should be attributed to our passage.
1.3 2 Ki. 3:22 TUB 3K1)3 1KT1 d^On ty niYlT BOBm 1pn3 in'OBPl
o-otmn ininj mnn nr an i-ttwi) BIO CPQTK tpnn OK
iron nK urn* ID»I 23.
The reference of O^mN in 2 Ki. 3:22 is defined by its
contexts. The situational context is that of war. According to
the story, the morning sun was reflected in the sudden flood
water which, as a result, looked 'red as blood'. The Moabites,
who knew there had been no water the night before, imagined they
saw blood (not water tinged with 'blood red' colour), jumped to
conclusions and consequently lost the battle. As a narration of
events, the story is full of holes: can water become so intensely
red from the sun as to resemble blood? Why were the Moabites so
incautious? 35 But this, as Gray points out, is a prophetic
36
story, one of the great Elijah - Elisha cycle. Tne miraculous
event corresponding to Elisha's earlier prophecy (vv. 16-19)
should not be rationalized for, as it is told, it is a miracle;
if explained and described as a natural, although rare, phenomenon,
it is demoted to a sphere of every-day life and that, surely, was
not the writer's intention!
The lexical context, however, is very clear: the water is
DnD D^BTK. The term DTK here designates the colour of blood;
T- • •-'. T
hence the translation 'red as blood' is fully justified.
Nevertheless, no far-reaching conclusions can be drawn from this
instance. Gradwohl goes to great pains to stress the connection
between 'red1 = the colour of blood, and the importance of blood
37
in religious history. However, our instance in 2 Ki. 3:22 is
unique, whereas DTK defined by other materials appears in other
places (Isa. 1:18, 63:2). This singular syntagm, even if only
accidentally singular, cannot prove anything apart from the fact
66 BIBLICAL COLOUR TERMS

that it exists; one cannot fondly imagine on this basis that 'red
as blood', so usual in our modern languages, was an often-uttered
38
coinage in OT Hebrew. The Akkadian coinage of 'water looking
1 39
like blood might be helpful as a cognate but not as conclusive
evidence of the regularity of d«*T3 OIK as an idiom embedded in
f T
ancient Hebrew lexical performance. On the contrary: had DTK
T
been considered the 'colour of blood' par excellence, that is,
had the latter been perceived as referring to the focus of the
area covered by d*TK, there would have been no need for the
qualifying Dfr5 - tPQTN would have been sufficiently specific!
T— ' V"?
As it is, almost all the ancient Versions display the same text
40
here. Perhaps DTK was felt to have too wide a scope, too many
referential meanings, to stand on its own: it can refer to 'brown'
(1.2) or 'orange-brown' (1.1) as effectively as to other shades;
but when the exact shade is needed for the continuation and the
advancement of the plot (v. 23), a specification is needed and
supplied by the qualifying element QVT3. Thus Gradwohl's
reassertion of many scholars' remarks about the symbolism through
the equation 'red' = 'blood' and its historio-religious signifi-
cance might be true, but cannot be deduced or proven from biblical
material. This assertion depends on the preconceived notion that
OT Hebrew, like modern German or modern English, posited 'blood
red' as the focus, even the starting point or the origin, of the
referential sphere of 'red1 or DTK.
T
1.4 Isa. 63:2
In order to understand the referential allusions and the
colour connotations of this verse it should be cited as part of
its immediate word context (vv. 1-4).
v.i ... rmaa a^-nii ygan a ma K^ MT »n
T •;:
v.2 jnri jTr^vnm "luna^ D'TK yna
T

v.3 •'nans aoa-iKi >o?o tpYttn_ ... ^TJ!? mrn rni3_

?ftfrOK ^i^a !oi *T\3 'yy pnvj T;>I

v.4 nto ^INA njun 'n^ apj DT> >^

(And so on; v.6 closes the section).
Let us first summarize Gradwohl's treatment of v.2. He
attributes the verse to trito - Isaiah and proceeds to quote it
•7
(with the proposed BH° emendation - the omission, presumably
because of an error of dittography, of the first ^ of IWinW);
PRIMARY TERMS 67

to translate it into German; and to cite the LXX and V renderings

to the first part of the verse. He then goes on to remark that
YHWH is depicted anthropomorphically as a warrior whose red
clothes might be red battle dress (cf. Nah. 2:4 - tPyMjn, 'crimson
coloured'), possibly stained by his enemies' blood - 'yy k1"1?-1 T M
''IAS (v.3). Finally the free Aramaic T to this section of v.3 is
quoted together with Stenning's translation of the latter into
English.41
Now, whatever Gradwohl has said must be deemed correct,
valid, and relevant. The material is arranged so that in his
summary to the "Rot" section he would be able to state categori-
42
cally that tHK is, first and foremost, 'rot, blutrot (Gewand).'
T
The problem is that the description is atomistic - it takes the
circumstantial context into account only in a limited fashion,
ignores the rich imagery of the passage, and does not analyze the
significance of the facts mentioned. As such, this partial
treatment is not only misleading but also impoverishing: it is a
disjointed collection of items that cannot be termed 'an analysis',
for it does not enhance our appreciation of the passage as a
poetic whole. Let us try and examine the passage differently.
It is constructed as a dialogue between the dramatic personae, a
watchman-sentry and someone who is approaching the sentry's
post. This is a rhetorical convention with which the listeners
were probably familiar. 43 The watchman inquires and the newcomer
answers obliquely (v.l). The watchman has meanwhile seen that
the approaching figure's clothes are stained with tiTK, like
someone who had been treading the wine press; he inquires why
(v.2). The figure answers by utilizing the watchman's simile of
wine press (v.3a), and gradually moves to a new strand of
imagery - if until now we have imagined that his clothes were
grape-stained, by now we realize that they are soiled with blood
(v.3) and that he is no ordinary person but God the Avenger. Only
now do we understand, through the shift in imagery, that the
approaching 'figure' is not agriculturally employed, but intent
on a mission of destruction. The allusions to the colour of the
figure's garments are three: 0>*n:i Y.ltill (v.l), Itm!? Q*TK
T
(v.2),
3 44 •
as in BH ), Jv>a 1TD VUm (v.2). Other colour allusions are
implicitly expressed by ''TAn *yy Qtlitt T'l (v.3), and possibly by
the phonetic similarity of tm& (v. 1) and DTK (v. 2), ima (v. 1)
68 BIBLICAL COLOUR TERMS

45
and the "i*X|l situation. The twin thread of 'grape juice' and
'blood1 imagery is present throughout the passage: the first
cannot be ignored on the grounds that the lexeme OTK does not
appear in a syntagmatic relationship with 3Jy ,d»S3y.
On the whole, d"TK seems to imply various types of 'red': it
covers an area that is - within the text itself - dissected into
sections covered by words or expressions which are more specific:
46
V<inn, possibly a 'bright red' but a hapax, perhaps corresponds
to the 'blood-red' indirectly referred to by v.3 (dn^D t'l), or
the purplish 'red' of grape juice which is implied by vv. 2b and
3a.
True enough, the whole point is that God's garments are
covered with blood, not grape juice; but again, if we do not see
DTK (v.2) as a superordinate term which has partial synonyms or
further references in the passage, and which may refer to various
shades of 'red', not necessarily 'blood red' - we would miss half
47
the significance of the mixed metaphor and its connotations.
Ibn Ezra (to v.2) says: 1^ nftlTJVdTri dyoni - the resemblance of
the colour of blood to the colour of wine is the basis of the
prophetic hyperbole. Let us add that the similarity in reference
is well covered by the basic term dTK: we cannot say that the
lexical item DTK refers more typically to 'blood colour' than to
'wine colour', and that - it must be remembered - in a post-
exilic passage, one that was composed during a comparatively late
period of biblical Hebrew!
By way of rounding the subject, let us return to the LXX and
the T and see what can be gleaned from them. The LXX has eruthra
for d*TK (v.2), an item which Gradwohl records but does not
T 48
comment upon. As a matter of fact, this is the first and only
instance where dTK or a derivative thereof is thus rendered within
T
the LXX corpus. Otherwise, lexemes derived from purros are used.
In a similar manner, V translates various instances of dTK either
with rufus (Zech. 1:8) or, as in Isa. 63:2, rubrum related
lexemes. The different translation for our passage, especially
49
in LXX, might be significant. It is conceivable that the
rendering eruthra here - and only here - is accidental or
idiosyncratic, unique to the translator of this passage.
Therefore, no conclusions can be drawn from this unique occur-
rence. On the other hand it is possible that the dTK reference
T
PRIMARY TERMS 69

in Isa. 63:2 is, as defined by its contexts, closer to the focus

of the 'red' sector in LXX; and secondly, the mere fact that mor
than one term is needed, both in LXX and in V, to cover the
range of reference of the single Hebrew term d*TN points to the
primary, general, and superordinate character of the latter,
even if differences in translation - the outcome of varying
methods, periods, approaches, or lexical stocks employed - are
taken into account.
As for the T, i/|7nO derived lexemes are always used as the
equivalent of DTK ; thus we must conclude that the Hebrew and
Aramaic terms cover roughly the same range of referents. As a
footnote, let us just complete the quotation of the T (for
Isa. 63:3) which Gradwohl quotes only in part. The translation
is a 'free' one, but preserves the double metaphor and the
50
double colour reference of the passage:
<
.KI^JW^ iflhD ni7£)'» •piuPni T>^op tnn Tnio iip&o* VT •JNO
1.5 Zech. 1:8, 6:2

1:8 tm DID *?y am KPN njrT> n^n ->n>Ki

Q'j^i d'fniy c^niK tPoiD T>iriKi
6:2 tPfcHN d - J O I D imtyttin niD1)33

(6:3 D'oiD my^in nayttDi d*3^ d^oio rpiu^n roaiKDi

d^sttN d»m -[BH : omit, gloss.])
In order to illustrate the point in general Fronzaroli says
that the association of 'white' with 'horse' is clearly interest-
ing because an adjective which appears in a given corpus can be
51
defined by the substantives to which it is linked. As a
principle his words make sense; however, the fact that he picks
the 'white' horse - and not the horses called ,d>inb ,d">Tia .d^iWK.
D^^K - is symptomatic of the fact that some colour terms (his
"adjectives') are easier to define by their contexts than others.
d'"TK (sg. and pi.), like the other colour terms in the
visions, is applied to lexemes denoting certain animals. If so,
its point of reference should be similar to that of the haiK
T ma
. •-."•' -*~r
(Num. 19:2). There (1.2) d^K was interpreted as referring to
the sector that today will be termed 'brown'. Therefore it
seems reasonable that the same meaning be assumed for the present
passage: 010 (d^D-ID) ,d1K (d^fcTTK) should be rendered 'brown
horse(s)'. Indeed, some scholars have taken this path. Hertzberg
70 BIBLICAL COLOUR TERMS

even remarks that the Arabs call 'brown horses' 'red1, thus
explaining the occurrences in 6:2.52 G.A. Smith 53 renders 'brown'
in both visions. The ancient versions are of no particular help
here - the LXX has purros, V - rufus, and T - plQD , pnilD.
T:
54
Gradwohl, as usual, states that the colour is 'red', or
'reddish brown'. Mitchell is the most specific: he talks about
55
•chestnut coloured' or 'bay' horses. He further remarks that
perhaps the colour description should not be understood too
literally. The vision, he thinks, is an imaginary scene set
against a concrete background and inspired by the appearance of
riders (divided into troops) whose signs were perhaps dyed in
different colours for easy identification. He claims, and rightly
so (for 1:8 ff.), that the rider's actions are the crux of the
vision, not the horses' colours. Therefore, he concludes, no
symbolical significance should be attached to the colour of the
56
horses themselves. Rashi, for instance, connects dlN with
T
blood and war in his 'midrash' to 1:8, but hints beforehand
(l!? Kin naun nmn d^n* am wi) that one should not belabour the
referential meaning of the epithet DTN here. Ibn Ezra is even
more definitive: for him DIN is apparently 'red' proper but, as
he reminds us, we are dealing with a vision, and not with
reality, and there is no reason why an 'unreal' horse should not
57
be red. On the other hand he rejects any symbolical interpreta-
tion of the colour as superfluous - tnn IttD dlTNl ... tnonni
.Dn I9VT »i:i*O h^Nl d*»T lists!? In 6:1 ff. the horses and their
colours are possibly more important than in 1:8 ff. Rashi again
interprets the colours as symbolical, assigning each colour to
a nation hostile to Israel. However, Mitchell's point seems to
be valid: the horses' colours function as distinguishing marks
for the chariots; to attach a wider meaning to them would be
58
unjustified by the text. Let us summarize: the horses
described are either 'real' ones, and in that case the colour
denoted is 'brown' or 'chestnut'; or, the horses are imaginary,
and in that case d*TK means 'red' and the latter interpretation
is utilized for symbolic representations of the vision. Again,
the flexibility of d*TK as a wide-range primary term - set
against the later limited sense of it - is the cause of the
difficulty.
PRIMARY TERMS 71

The given date of the visions is 519 B.C.59 = stage 3 in

the history of biblical Hebrew. By this time OTN has quite a
few partial synonyms that are subordinated to it - P1K> (1:8),
VJ3K* (6:3) and V.inn (Isa. 63:1),60 not to mention ^in^IK_ ,&rrK)3
OWJK, - and distinguishable from it as specific sectors within
its territory. Yet even in this late period, because of the
absence of a synchronous term for 'brown' and the definitely
specialized use of 5iUl (as applied to hair, Lev. 13), ti'lN
still retains its general referential status although this
is detrimental to the clarity of communication. The most we
can say is that by stated opposition tHK is neither in!? nor
l'nt27 ,VnN*, i?*W, or Y12*. By analogy to the na*TK mfl (Num.
T T T T TV—. TT"
19:2) and by using extra-linguistic observations one arrives
at the conclusion that d*TK probably means 'brown1, 'chestnut',
or 'brown with a reddish tinge'. Alternately we could describe
DTK both as a primary term (on the first level of opposition)
and as a hyponym of itself, but with a more restricted
referential meaning.
A schematic representation of the last alternative will
look like this:

(brown,
bay,
chestnut)

For examples of single terms that can function both as

superordinates and as hyponyms (albeit in Modern Hebrew and
61
in Modern English) M. Dagut's doctoral thesis can be
consulted.
1.6 song 5:10 n:mn tnvranKi m? »Tn
(v.n - 62 n-iiyD mints d^nfti -pn-mi? TO &nD iK&a).
Song. 5:10-16 contains the girl's description of her
beloved's physical appearance. Within its literary context
it constitutes a reply to the question, (v. 9) what is your
beloved that you are so love-sick? One has the impression
that the daughters of Jerusalem (vv. 8, 16) are slightly
sceptical, or world-weary, and that what they really ask is:
72 RIBLICAL COLOUR TERMS

what is so special about him that you feel and express such
violent emotion? The girl proceeds to describe him from head
to foot, invoking extravagant images of doves (12), scent and
sweet smelling flowers (13), precious metals (14, 15), and
nature (15). The imagery is complex - it evokes visual,
tactile, and olfactory sensations - and extravagant in
application, but precise in reference. Against this background,
v. 10 (and vv, lla, lie) should be interpreted as possessing
specific colour denotations, ones that are outstanding, peculiar -
in the eyes of the speaker - to this singular young man, not
the usual physical features of any ordinary young man, and in
the sense that the beloved's appearance is better than average.
What is meant by DllKin^? Much depends on our grammatical
classification of this syntagm. If it is a hendiadys, then tltf
(in this context) should have the approximate referential
meaning of D1*TK, and this meaning should be substantiated by
other occurrences of D!tf (notably Lam. 4:7 - D^hfi Iflij!) and of
DIN and/or its derivatives. Alternately, h^ and dllN can be
considered separate attributes, that is, oppositions. In this
case, what is the point of reference specified by each one of
them? Syntactically both modify *T1?T - but what is the topical
entity each of them refers to? Finally, while &1N has already
been established as a primary colour term, hi? has not. In
Lam. 4:7 .Ihif. (the verb form), because of its syntagmatic
association with 3^hO, must be considered a colour term,
T
although not a primary one;63 but this cannot be applied to
the present passage without prior examination of the term in
that particular context.
Ihi? in Lam. 4:7 is defined by !)t?htt as an equivalent of
•p!? (Gen. 49:12 - 3t>nn D'-ie; p!M) at least for that
TT TT" • - • v:
context. Therefore the question now is: what is the reference
of "pi? and, by analogy, of .IDS and hif? It is usual to define
pt> and consequently h2 as 'white'; the comparison to 'milk1
or 'snow' (Lam. 4:7) seems to strengthen this approach. Thus
hX and D1*TN seem to be separate attributes. What do they
actually describe? Rashi is quick to recognize the problem.
He says, 1*331 Vtf Ninw "lin^ »i:i Ol^>p D11K1 ^h» 103 pt? MS
ni'JIlTTN - hif means 'white'; white is the colour of his skin
(all over $he body), while tmK (as equated with >310TK)
PRIMARY TERMS 73

modifies, or reports the colour, of the young man's complexion.

Indeed, this explanation is the only valid one for the 'white
64
and ruddy' approach (so also Currie Martin ) . Rashi and
Currie Martin equate Q1TK and ''JlfcnN. This is by no means
certain: each "OimN occurrence should be examined against its
context in order to determine whether it describes 'complexion1
or 'hair'. Here, however, the beloved's hair is described in
the verse following ours. In fact, vv. 10-13 deal with the
lover's appearance in the following order: i. His beauty in
general (v. 10). ii. Various items of his head - hair (11),
eyes (12), cheeks and lips (13). Structurally, item i can be
assigned to 'complexion' only; hence both Q1TN and hi? designate
complexion colour. This recognition, although similar to
Rashi 's and Currie Martin's, rests upon a literary examination
of the text but not on a D11K = 'JlttTK formula.
Even if we agree that hif means 'white', what colour does
D1TN refer to? Snaith talks about 'dark and ruddy from
65
exposure to the sun'. This interpretation is highly unlikely -
the idea of burnt and tanned skin is expressed in Song, by
/Thiy-derived lexemes - thus mine; (1:5), and JlSrnnw (1:6).
Rather, D1TK probably means (here as in Lam. 4:7) the colour
of healthy complexion which is pleasing to the eye. Ullendorff
points out that /'dm derived lexemes in Hebrew as well as in
Arabic and Ugaritic (and in some Cushitic languages) mean 'red',
'brown', and the colour of human skin. In Ethiopic, he says,
/'dm has lost all its 'colour' contents (which is carried by
the root gyh) but means 'pleasant, beautiful, pleasing to the
66
eye 1 , as healthy or (what would be approximately termed) "pink*
complexion really is. Similarly, the Egyptians called themselves
67
'red1 in contradistinction to the 'black' foreigners: the
range of their DIN- equivalent must have been similar to that
of the Hebrew term, and the same concept probably underlies
both linguistic usages.
Let us return to n^ and its status vis-a-vis DTK. As will
T
" 6 8
be discussed in the relevant section and as argued in the
introduction to this study, *pX flif, or any other colour term
TT
are usually defined by us according to the chromatic quality
we think they denote. The colour sensation, however, is the
product of the brilliance and saturation, as much of the
74 BIBLICAL COLOUR TERMS

chromaticity, of the object viewed. We might feel that the

'white1 element is the most significant feature of ,-iriif. ,-13£
"PX especially since they are compared to 'milk' and 'snow1.
Alternately, though, the element of brilliance, lustre, and
light might be equally important for the ancient poet and/or
listener. If we understand fTO as 'brilliant, dazzling' (= the
most prominent quality of snow, what we call 'pure white1)
rather than 'white', we cannot claim that Q1*TN1 hif is a
hendiadys: ho instance in the OT points to the usage of QITN
or its immediate derivatives as markers for brilliance rather
than some (although rather wide) chromatic value. Consequently,
the 'hendiadys' solution is not acceptable. However, the
syntagm can be interpreted thus: the lover's appearance is
pleasant indeed - his skin glows, presumably with health
(instead of 'is white'), while his complexion is pink. This,
as Rashi says, is the conventionally proper colouring of a
healthy young person's complexion.
It is tempting to view D1TK1 hif as a compound (and thus
subordinate) colour term referring to a boundary between *pt?
(as the superordinate of US) and 01 TN with the above-mentioned
denotation of a 'pinkish' colour (of complexion), but defined
by means of using a compound, rather like the dWtN "]$*? of
Lev. 13. Against this view it should be said that the
structure - two lexemes with a conjunction between them - is
different from that of d*TO*TN "pb which, because of its reference
as well as its form, should be considered a single lexical item
69
(although formally it too is a compound) .
To summarize: two interpretations are possible:
i. hS = 'white'; dllK = 'pink1 (or healthy complexion
colour)
ii. CIS = 'glowing'; D1TN = as above.
I prefer the second interpretation for literary reasons. Further,
v.ll presents a similar problem: how are we to understand IWi
311 3D ni*lhO...Tfl dtt3 if, unlike BH , we want to retain the third
part of the verse? The only solution is to say like Rashi for
this verse: his head glows, is brilliant, like gold, while (the
colour of) his hair is as black as a raven. The structure is
of course different from that of dllKl hS (v.10), but the
emphasis on the dazzling appearance of the beloved is one of
 75
PRIMARY TERMS

70
the leitmotifs of this passage.
1.7 Isa. 1:18 71UBD tD^KOtl 1*n' dK nifi» ION' nhDUl K3 1^
•pn» "HMD ytnro -itt'TK^ 72dK .1^-73^ ^uo ,
The Hi£. lexeme derived from DTK is considered the most
T 73
ancient example of DTK-related verbs in the OT. BDB defines
T 74
the Hif. as 'show redness'; Gesenius - 'be red1 or 'make
itself red';75 Klpnn IIP1!? 1X1K - 'be red1 (as for the Qal
76
formation). Gradwohl leans on the Arabic cognate and
explains the form (together with the Qal) as'be or become
77
red'. However, as Fronzaroli iustly says and as we have
previously seen, the fact that O^NiVbased forms are usually
translated or explained as 'red' depends not only on the
reference but also on the system available in our own
7o
language. To say that .1J3*"TK* in this instance - the only
one of its kind in the OT - means 'be/become red' does not
explain much; even our 'red', which is much narrower in scope
than an ancient corresponding term, is still too wide a term
to be specific. We can narrow the term further by forming a
compound of 'red' + the name of an entity whose typical colour
we want to convey ('blood red', 'tomato red"), or we can resort
to an analogical construction ('red as blood') that would make
the point of reference more explicit. (A third way is, of
course, to coin new lexemes). Here the second possibility is
utilized, together with a juxtaposition of the colour with an
oppositional shade. y^lBD •ltt'*TK'> is self defined: the general
term is narrowed, and modified by y^lIO to mean 'the colour of
crimson', 'scarlet' (D*3EO. „ „ 1TP DN)79 as opposed to that of
80
'wool' and 'snow'.
The primary, general nature of DTK becomes manifest again
in this example: the immediate verbal context gives the lexeme
its focus, a focus that is valid for this context only, y^lfl
can be considered a kind oftiitf,but a substitution test - is
T
DIN a typeofy^in? - is not satisfactory. The literary context
of Isa. 1:18 ff. is, of course, prophetic and poetical. More
difficult to determine is the circumstantial context of the
passage. The previous verses represent the people of Judea
and Jerusalem as totally corrupt. In 11-15, which might be
imagined as uttered against the background of the worship in
76 BIBLICAL COLOUR TERMS

the Temple, no colour term is explicitly mentioned. However,

blood is referred to twice: 0*T of sacrificed animals (11.) and
D>QT (15), probably a double allusion to the 'blood' of v. 11,
but also - in the figurative sense - the 'blood1 of social
injustice (16-17). The visual picture underlying 11-15 is
that of hands that are bloodstained from sacrifice (which makes
the owners of these same hands morally tarnished). The condition
requires cleansing, both physically (.l^hl, v. 16) and
spiritually (.IJWri), in order to remove the blood-red stains.
This implicit picture of blood colour and the call for its
removal is in a sense parallel to the 'crimson-white' opposition
of v. 18. It is tempting to view vv. 18-20 as an elaboration
QI

of vv. 16-17, whether y. 18 ff. is a formal invitation for a

lawsuit, a sarcastic remark, a straight-forward interrogative,
82
or a rhetorical question. On the other 'hand, "it certainly
seems doubtful whether Isaiah would, in immediate succession,
first represent the people as red-handed criminals (v. 15) and
Q
then treat the redness of their sins as hypothetical (v. 18)".
In the light of vv. 19-20 (if the latter are not a fragmentary
saying but a continuation of v. 18), obedience and consequently
forgiveness are still possible and the reversal of fortune can
be achieved by making the right choice, whereas the call for
repentance in vv. 16-17 does not give out much hope. Therefor
it is perhaps better to regard v. 18, or rather vv. 18-20, as
an independent prophetic utterance whose theme is that of moral
choice and its results (the reversal). The utterance, although
unconnected to what precedes it, was probably inserted here
because of the associative value of dT and Itt'"!^, blood and
crimson, and the parallel idea of 'cleansing' in v. 16 and in
v. 18.
84
1.8 Prov. 23:31 O^S im »D dWll1; O 1" Kin "?K
86
anB»'aa n^nii* 85i:i»j>
87
The dictionaries as well as Gradwohl explain the Hithp.-
• 88
lexeme derived from QTK as 'redden', 'grow or look red',
T" QQ 07
'look red', or 'have a red sparkle'. Gradwohl goes on to
note that there existed a poetical convention of calling wine
or grape juice 'the blood of the grape' - not only in OT
90
Hebrew (as in Gen. 49:11, Dt. 32:14), but also in Ugaritic
 77
PRIMARY TERMS

c 91
(dm. sm. ~ 'blood of the vine'), and a comparable dam erinni
'
'cedar resin' - in Akkadian. 92
It is the last idiom that
Gradwohl catalogues which is quite significant: the 'blood of
the grape' is not necessarily named thus because of its 'red'
colour (see below); rather, the wine is seen as the essence in
liquid form, the 'life1 of the grape, and as such it is
comparable to blood. The colour association does exist, of
course, but is either secondary or else equal in importance to
the 'essence' notion. Even in Isa. 63:1-3, which is based on
the blood letting = grape treading analogy, the image of
forceful physical treading and subordinating is at least as
important as the colour imagery.
The modern noun phrase 'red wine1 (opposition: 'white
wine') which immediately springs to mind as an equivalent of
the biblical syntagm is far from specific. It refers to a
relatively large class (more purple, brownish, or yellowish red
than 'red' proper) of dark coloured wines, in contrast to light
coloured wines that are called either 'white' or 'Rose' but are
never 'really' white.
Similarly, the difference between white (milky) coffee or
tea and black coffee/tea is that of brightness - or the lack of
it - more than that of the chromatic quality of the object
described. In that sense OWn* is only a very loose colour
term - it denotes the boundary of the sector where it merges
with 'dark/black'.
Another line of investigation should be considered as well.
The clarity and gleam of the wine are as telling and as
93
fascinating as its colour. Further, the meaning of /'dm in
94
Ethiopic - 'be pleasant', or 'attractive1 - is perfectly
suitable for our context: although wine seems pleasant, the
result of excess drinking is disastrous. It must be pointed
out, though, that if one accepts the latter explanation of DINfl
one should not classify it as a colour term but rather as a
derivative, or perhaps a lexeme which has colour connotations.
87
To summarize: Gradwohl writes that the occurrence of
DTKn' is probably pre-exilic. The dating fits the description
of DTK and its direct verbal derivatives as primary colour
terms denoting not only a wide range of chromatic values, but
also other colour properties (saturation, purity). On the
78 BIBLICAL COLOUR TERMS

other hand, if tnN]T> here means 'attractive, pleasant' (as in

Ethiopia), then it is not a colour term at all, although its
etymological link to the colour field is fairly transparent. I
think that more decisive conclusions cannot be drawn - both
interpretations are suitable for this particular context.
or
1.9 Lam. 4:7 n^MO 1h$ tfVQ fm»n 13T
96
dmn "pas d'jpioo dsy IOTK
8 .. 0diKn -nnuna iwn
97
.1Q"JX is a verb formation, but stative in meaning. As a
matter of fact, an alternative construction of d>3T I'M* or
•PM d>3T*) d*h2 "Pil*. d»mN *PM* - that is, a copula + noun -
would have served equally well. This suggests that so-called
colour adjectives and their derived Qal lexemes, wherever these
derived terms feature in Biblical Hebrew, exhibit the same deep
syntactic structure. It is perhaps the reason why we do not
' 98 The phenomenon is not
find a Qal formation for *p!? and IhB.
unique to colour terms. There is no difference in deep
structure between the trttA-'n of 1 Ki. 10:23 = 2 Ch. 9:22
(VIKil »Di>n ^DQ nrttw !m>1) and the ntm> of 2 Sam. 13:15, flKJK
nun* *WN rar»Kn ww *«»< iVOBn ntm* »D TK» ftf'm mow I'vjnK
A fuller discussion of the subject will be undertaken at the end
of part II (Appendix, pp.172-174).
As an oppositional colour notion •Itt'TfK belongs to the same
series as the earlier Isa. 1:18, where we find the double pairs
•»i»yD/1J'»^> Jl^KD, y^inD WrNVPn* "10S:> which illustrate the
same idea of colour contrast (see 1.7 above).
Song. 5:10, if dlTNl MS are to be understood as two distinct
terms (see 1.6), should perhaps be added to the list. In other
99
words, these "pVdTK or "^-subordinated lexemes serve as a
-rr T
conventional contrast, to be found in poetical compositions of
three stages of biblical Hebrew (from stage 2 through to stage 4).
Hence, the convention must have been very well rooted in Hebrew
idiom.
From the referential aspect -1)3*JK belongs with Song 5:10.
Although in both places the grammatical subject is a person
('in, Song.) or persons ("pTTJ, Lam.), the topical subject
is the complexion or skin colour of the person(s). In both
occurrences dTN and •lO'TK denote the pinkish colour of a healthy
PRIMARY TERMS 79

100
complexion. Gradwohl, although hesitantly, remarks that
perhaps a reddish, light brown, or suntanned colour is meant.
It seems to me, though, that the concept of this type of skin
101
is denoted by the usage of ItlW ,1FTlfHj>* as in Song. 1:5-6,
and is far from being considered as the conventional beauty
102
ideal. Rather, as Gesenius says, 'whiteness' and 'redness1 -
or the light-coloured complexion that is the result of glowing
health - are part and parcel of the description of youthful
good looks. Although the referent of D»}'>3£) - rubies or
103
corals? - is far from clear, the exact identification of
D^J^O cannot influence the connotation of .IftTK in the present
context. The preceding metaphors - H!tf10 .IDS >>^W3.. .-13t_ - and
the following one - lihera ,T.Bffl - are clearly intended as
exaggerations, emotive descriptions used in order to enhance
the effect of the poem. In this literary context whether »inTK
is described as more 'pink1 (or 'reddish') than rubies or than
corals is immaterial.
1.10 Ezek. 27:13. A suggestion has been made emending DTK K/£)3
TT VV
to DTK Bfij.daw nwna ^i DTK aaan v^:n nan nuni bmn 11*
104
*DTyn). Y. Yadin, following an identification of the Akkadian
105
nabasu as 'dyed (red) wool', suggests the emendation of
DTK Bfli (Ezek. 27:13) to DTK WSJ = 'red (expensive) wool cloth1.
^""r V* V "^ •* *• i n£.
In this he follows Mazar's earlier suggestion. Although the
ancient VSS and the commentators have always understood DTK KJQ3
TT •• v
as referring to the slave trade, the context - IH9h3 *^D and
types of dear cloth - such as IttATK, not?"!, and V-13 are mentioned
(v. 16) among other luxury goods - would benefit from the
correction. If we accept this correction, although it is not
corroborated by additional biblical evidence, this instance
will belong with Isa. 63:2 (which is later in date), because
DTK modifies 'clothes' or 'cloth' in both passages. The
T ,
precise reference of DTK here will remain impossible to
determine.

1.11 Summary
After analyzing all occurrences of dTK and DTK-derived
direct verbal lexemes it was found that bib. DTK
T
is not the
exact equivalent of our 'red'. The total sum of DTK references
in its various contextual environments points to a range that
80 BIBLICAL COLOUR TERMS

is much wider than that of 'red'. Thus, the use of only one
term to translate the OT Hebrew DTK is not always possible.
This statement is perhaps borne out by the practice in the LXX
and the V. The cognate Aramaic terms used for rendering VO"TK ,
VTW30 and its derivatives are never varied, probably because
the scope the two groups cover, each within its own linguistic
framework, is similar; that is, the structure of the same sector
in both languages is, roughly speaking, parallel. In that
sense /DTK and /J7KJO are symmetrical inter-dialectal equivalents.
The area of reference QTfK covers is:
T
(a) 'brown* (of animals' hide) in Num. 19:2 (1.2) and in
Zech. 1.8, 6:2 (1.5)
(b) 'yellowish brown' (of lentils) in Gen. 25:30 (1.1)
(c) 'blood colour' in Isa. 63:2 (1.4), and perhaps in
2 Ki. 3:22 (water - 1.3)
(d) 'crimson' (metaphorically, of sins) in Isa. 1:18 (1.7)
(e) 'wine colour', or non-chromatic colour properties, in
Prov. 23:31 (1.8)
(f) 'pink1, healthy flesh colour in Song. 5:10 (1.6) and
in Lam. 4:7 (1.9).
The range of QTK in Modern Hebrew, even when the term is
loosely used, probably covers the area denoted by items (c),
(d) and (e) above. In other words, the primariness of biblical
dlK is enhanced by the fact that it is the chief (chromatic)
colour term extant in our text. Therefore, its references are
less restricted and much more given to manipulation and
flexible usage than a comparable term in a language where the
colour field as a whole is better developed.
PRIMARY TERMS 81

B.2 "]&
TT
2.0.1. "p^ is the most frequent colour term in the OT. It
TT
appears 24 times as a nominal, another 5 times in directly
derived verbal lexemes. Although secondary and tertiary terms
in its sector (fTS ,*U*) are genetically unrelated to it (apart
from the compound Oim*TN 12tO , 1^ appears as an element in at
Tt— . TT TT
least eight names (be they generic, personal, or place names),
and in other lexemes (M3!l^ ,h31l!7) that have indirect colour
! TT; T :
connotations. As a colour term it features in texts dating
from the period of the monarchy onwards. Its value as a primary
term, then, is amply documented. Moreover, its basic semantic
meaning is - judging by the evidence available - opaque. No
connection between "p^ as a colour term and any lexeme referring
to an object whose typical colour property 13^ has come to denote
TT
is to be found. This, of course, serves as further proof for
the primariness of Vl!7, although it does not offer any clue as
TT
to its range of reference. Gradwohl does not share this opinion
about the opacity of "]^'y. Even though he accepts that the
biblical lexeme is far removed from its origin, he suggests -
with a considerable amount of confidence - that 13^ is derived
TT
from a prehistoric Hebrew word denoting 'milk'. This view, he
argues, makes sense: it is 'natural' that a nation of nomads and
shepherds would see the colour of milk as 'white' par excellence.
Later, "p^ had fallen out of use as 'milk' and its original
reference came to be signified by 3t>tl. In order to substantiate
TT
his argument further he cites Arabic, where lexemes from both
_ _ 2
\Jlbn and \/hlb denote various types of milk.
Tempting as it may be, Gradwohl 's semantic reconstruction
seems improbable on etymological as well as intrinsic Hebrew
grounds. T. Fenton suggests that no North Western Semitic
language has a / li>n-lexeme denoting 'milk', whereas classical
Arabic, which has /IJbn-derived lexemes signifying certain types
of milk (see below), does not have a \/ Ibn = 'white'. Apparently,
he says, no Semitic language has the same consonantal base
3
carrying both references to 'milk1 and 'whiteness'. In fact,
_ _ 4 ^5
in Aramaic /MJb is reserved for 'milk', /Mr for 'white'.
In Ugaritic, like in Hebrew, hlb means 'milk', while Ibn (an
element in names) is derived from an underlying, albeit non-
6
attested, Ibn - 'white' lexeme. In classical Arabic halib
 82 BIBLICAL COLOUR TERMS

is 'fresh milk', laban or liban 'sour milk1, and compounds like

^•jJJl L—-Jj> are also attested, while the lexeme denoting
'white' is 'abyad. In Akkadian there are two labanu roots, but
o v 9
neither denotes 'milk' or 'whiteness': 'milk' is usually sizbu,
while 'white', 'light coloured' (even 'light grey') is denoted
10
by pesu. The hlb root of Hebrew, Aramaic, and Ugaritic
appears in Akkadian in dialectal = Assyrian occurrences only
11
as 'halabu' or 'halapu'r meaning 'to milk'. It would appear,
then, that the concepts of 'milk' and 'whiteness1 are never
defined by the same lexeme, be it /MJb-derived or /lim-derived,
in the Semitic languages cited. When these observations are
applied to OT Hebrew, internal considerations should decide the
issue by reinforcing, or over-riding, the cognate evidence.
This would be our task in the following paragraphs.
"p!? imagery in OT texts is concerned more with snow than
TT
with milk. This is primarily a quantitative judgement which in
the case of biblical Hebrew cannot be conclusive. 12 More
important is the fact that the 'snow' similes are not context-
bound, and/or limited to a single chronological niche. A^^
•IJPabl (Isa. 1:18) and "pa^K >J?WO (Ps. 51:9) are explicit
13
examples. >!?ew h'TTJ IDT (Lam. 4:7) does not seem lesser
in 'whiteness' than its parallel ^ho in* (both -13T_ and -in^
function here as subordinates to the superordinate l!Jt> ).
TT
Although the formula A^WS yiStt (2 Ki. 5:27) / *"pW3 nyY»3 (applied
to 'hand' - Ex. 4:6; to a person - Num. 12:10) does not contain
the term *p!S> and probably refers to the texture of Psoriasis foci
TT 15
rather than to their colour, it is clear that 'white' can be
signified by a comparison to snow. Other oblique references to
'whiteness' are specified by T»n* IQiD (Isa. l:18d, parallel to
•mn^ A^KD) and n^nn PJO (Num. 11:7).16 On the other hand,
where 1H5 is specified by 'milk' the context-bound usage does
not necessarily indicate a typical quality of 'whiteness'. The
'milk' imagery of Gen. 49:11-12 might have been conditioned by
17
the 'wine1 hyperbole which precedes it. In Lam. 4:7 u!?h is
TT
not superior to >^y as a colour specification for 1^!?. In
short, we have no evidence - either in cognate languages or in
literary Hebrew usage - for treating *p5 as a semantically
semi-transparent derivation of a base lexeme denoting 'milk'.
PRIMARY TERMS 83

2.0.2 As in the case of QTK, it would be inaccurate to define

T
13^ as the equivalent of a single term - in this case 'white' -
and to translate it accordingly, for its primariness is
reflected in its wide scope. It refers to areas which are low
in chromaticity but, at the same time, relatively high in
brightness. This feature of 'loose' reference is precisely
what makes the term highly operational. On the other hand,
when greater specifications within this same sector are
necessary, other linguistic means have to be resorted to. A
case in point is the Mishnaic comment in Neg. 1:1,2 (for
Lev. 13), where various shades of 13!? are distinguished on
both scales - that of light/darkness and that of chromaticity:
^•>nn "PD3 ft1? ww A^KD nry mm ,ny3iK orM tpjo d»y^ niton
d>»Dm ."pNo ^31 mn - 13!? in>D n!? h^w ,niP3 DIIPD Mam
18
.n>P3 diipD nt> rpjw 13!? -I»*D nKu/n onoiK
2.0.3 In the following sections I intend to discuss first
the nominals and then the verb occurrences. The usual order
of the Hebrew text will be adhered to, apart from the joining
of the Zech. instances (1:8; 6:3,6, where 13!? modifies 'horses
to Gen. 30:35 (where 13!? modifies 'goats'), and that of Ps. 51:
TT
to Isa. 1:18 (where formal as well as literary criteria dicta
this arrangement).
2.1 135 as colour marker for domestic animals:
TT iq
2.1.1 Gen. 30:35 D^OiTl ^Wfi d^Tlil HK Kinn d1»3 1CP1
.d*3eo3 o^n toi in in!?
TT
IBK !?D'niKt»om n'npjn d'ryn !?D HKI
If 13 13t> 1WK !?D is understood as an explanatory apposition to
niKboni Jm|7jn onyn *J5, the expression probably refers to those
that have 13!? spots, or marks, on their skin. Thus Neophyti I
21
has 20.KTPh NB1W rT»3 fl'KT !?D that is, a spot or mark, l whileE
Rashi uses the only biblical word that might approximate 'mark,
23
spot', 22nj3!? mi3*l3n 13 nn^n IBK !?D. Sa'adia Gaon 3 understandsS
D*1py_as those whose ankles are of a different colour, as if they
'%"' 24
had been bound (from /npy, Gen. 22:9 ). Therefore he
interprets the appositional phrase thus: l!?!!) d>TlpJ?MB 'fl!?
. 1?1'!?>1 >!?I1D1P3 K!?K 13l!? dn3 1>K1 dnihiy Clearly, the sense
of 13!? here is defined by Tpl* ,M!?D* and np^* (if we do not
dispose of the last one as a corruption of d'*Tp/5) . Still, its
reference is far from clear. As stated above, 13!? may denote
TT
any colour property characterized by both a high value of
84 BIBLICAL COLOUR TERMS

brilliance and a low value of chromaticity. As such it covers

the slot which in modern languages would be filled by an
equivalent of 'light coloured', 'beige', or something similar.
The lack of biblical terms which are peripheral to 'white' or
'black' makes the utilization of *p^ for 'light coloured' and
1HW for 'dark' almost inevitable. On the other hand, the
consistent interpretation of "P^, wherever it occurs, as its
TT
modern parallel 'white' is misled and misleading - because
'white', "p!? in Modern Hebrew, or any of the contemporary
equivalents in languages spoken by members of industrial
societies are much narrower in scope than biblical Tp^. Perhaps
TT
25
Skinner is right in saying that the employment of "p^ here
may have been motivated by the wish to have fun on Jacob's
father-in-law by using a word-play on his name. Although our
text refers to *pt> in goats (tiny ,0>Wtl), perhaps a short note
on Isa. 1:18, 1>M' 1&2p, is not out of place here, inif per se
*/ • 9&
refers to wool in its natural state. It is difficult to
compare the 'whiteness' of natural wool before it has been
cleaned and bleached to the 'whiteness' of snow (U'D!?' >^KO,
or in Ps. 147:16, injp >>U>) . Nevertheless, there is a literary
convention of treating A^KJ and 1Q^ as synonyms for designating
the same, or approximately the same, colour phenomenon. Now,
that this is a literary convention only and not an equation of
referents can be gleaned from the Mishna Neg. 1:1 f. quoted
above (2.0.2). In the Mishna the semantic sector "]^ covers is
divided into graded segments, and the segments are designated
by means of colour specifications. It is implied that an
unspecified *p^ is inadequate for diagnostic purposes because
TT
it refers to such a wide area that colour entities typified by
snow, the whitewash of the temple, eggshell and wool ("pt> "H3JO,
in that order, are all included in it. This primariness of the
term - which makes it too general for medical-cultic practice -
is what allows the snow/wool colour parallelism.
However, when 1JM is employed as a specification of "p^
apparently bleached or cleaned wool (not natural) is meant.
The Mishna (above) has "p^ 102 , and the T to Isa. 1:18 10^3
27
•pro has 1in> VJ *in^5. The difficulty arising from the
limited scope of "p^ in post-biblical language, together with
the fact that the colour properties of both ^V and l»^ are
PRIMARY TERMS 85

referred to by 13^ in biblical language, prompted Qimhi to

state explicitly that we should interpret the 1tt2 of Isa. 1:18
not only as ^ "W2* (as in the T), but specifically }& *ltt^3
'T *.' V 98
"»i?J1 (as in the Mishna). Another Jewish commentator who sees
the problem and points it out is the Vilna Gaon 29 who, following
the above mentioned Mishna, says: 10 MOOt? KinB 1tt£3 - "PiT* 1»^
3
°.A^n
2.1.2 Zech. 1:8 Kim Q°l$ 010 !>y 331 B»K fUni h^!?n ^1T>K1
t^in'tj) d'>B'TN<d'>Dio •pinKi...d»b'rnn v^ lay
' '"• d'jnb.i
• f°".
Zech. 6:2 dnrns d^oio...d»tfw: d'O.io
3 °*^31 °*Tp d>oio...d>inV d'oio
6 d»triini...d>j^r)i...d»SnBn d^oion
7 . ..d>a»Kni
"pt?, when applied to '(d*)01d ! = 'horse(s)', is defined
2
as a colour term by I'tlttfT ,thNT ,VOK*
1
T
and Tia*,
T '
all of which
serve to modify '(d*)01o' in the same context. In contradistinc-
tion to the syntagm DTK 0-10 or d^iJttK. (d'TIIl) d>p.1D, the extra-
linguistic reference of \^f} 010 does not depend on the question
of the existence or non-existence of symbolism in the two
33
visions. That the term means 'white' here is made probable
by the fact that ancient extra-biblical sources maintain that
'white' horses were not unknown in victory marches - perhaps
because 'white1 was sometimes associated with victory. 34 In
his commentary to Rev. 6 Charles cites Herodotus, who attributes
this practice to Xerxes and his general Mardonius, and Dio
Cassius, who says that Julius Caesar's triumphant car was drawn
35
by four white horses. We must conclude that Zechariah could
have known that white horses, although rare, did exist; and
that whether the symbolism meant in our passages should be
interpreted in political terms, or (at least in ch. 1) abandoned
36
altogether, makes little difference to the understanding of
•p!? within the two visions.
TT
2.2 Gen. 30:37 ^O^l TlQlyi Tl!n ft1? ilJ^ t>pn lp)P 1^ Hp^l
m!?pnn ty -MX iat>n iform rim!? ni^s inn
TT™ T; T;
Jacob performs an act of sympathetic magic - the streaks
37
peeled in the bark of the Styrax, or the Populus alba, and
the other trees are expected to exert visual influence on the
38
pregnant sheep and on their yet unborn offspring. Does 13!?
here mean 'white'? /5^D and its nominal derivative ni^9 appear
86 BIBLICAL COLOUR TERMS

only here, but are well documented in cognate languages - in

39 40
Aramaic dialects, Arabic and MH. iWltt, the nucleus of the
~ 41
phrase which is appositional and explanatory tofllja!?ni^S,
is a hapax legomenon, but the root it is derived from is well
attested in biblical literature (in Josh., Isa., Jer., Hag.,
42
Jo., and Ps.). There is no doubt that the layer under the
bark of any fresh rod is paler than the bark itself. However,
the manner in which we define it - as 'white' proper or simply
as 'pale' - depends upon the semantic range we attribute to
'white', ^'y or their parallels in modern languages.43
2.3 Gen. 49:12 irthti d^tf 441^-1 l^ti d*:py *?**?H
The designation of *p^ in this verse should be defined by
the juxtaposition of '^^tfVp^ , V/^bh as opposite members
of the poetic parallel structure.
Traditionally, Jewish commentators understood »!>*toh as
a colour term. In both T Onkelos and Neophyti I the colour
opposition p'MO (= ^^h) and 11 T\ (= 1^) features in the
" ' to this verse.^ Rashi, 45 Ibn Janah (JMn'QTN"),
haggadic expansion 46

and Qimhi 47 follow the Aramaic translation. On the other hand,

modern scholars prefer to interpret ^^Dh (and fll^ton - Prov.
23:29, again in a syntagm with tPJ^y and in the context of
wine-drinking; n^'Dflh (Jljm) - 1 Sam. 23:19, 26:3; and perhaps
the personal name fP^>n) as 'dull', 'vague', assuming - on the
basis of cognate evidence - a shift from a basic meaning of
48 49 50 51
'dark', 'dim' to 'dull'. So Gesenius, BOB, KB, Skinner,
52 53 54
Henton Davies, Ball, and others . Indeed, d>:py ^tttn
here is used in an ameliorative manner (in contradistinction
to Prov. 23:29). Dullness, or blurred vision resulting from
excessive drinking seems more appropriate to our metaphor than
a bloodshot (= DIN) condition of the eyes.
T
Now, if I'yi'yyn is 'dark' or 'dull', 'vague', we would
expect its parallel to mean 'bright' rather than 'white'; in
other words, the crux of the metaphor would be placed on a
brightness/lack of brightness scale rather than on a chromaticity
scale. On the face of it the juxtaposition Dbh/V prevents
such an interpretation, especially in the light of expressions
like OTKIP "O V N1fl !>N (Prov. 23:31), D'tuy CTT (//!», Ge
49:11), and 133? dl//inn (Dt. 32:14), which have been interpreted
55
as defining ordinary wine as typically 'red'. Wines which
PRIMARY TERMS 87

were not designated DTK or 'red1, even after considering the

loose manner in which "red1 was used for the classification,
were known in the ancient world: in S.M. Paul's list of
correlating terms in Mesopotamian and Rabbinic literature
there are terms denoting 'white' and 'dark' wine as well as
56
'red' . In addition, the figurative usage of Q^/UJJ dT might
have stemmed from the equation of 'blood/juice' as essential
liquids, not only from the colour similarity as in the case
of some types of wine.
What about U^h, then? The colour properties of the
TT
latter cannot be disputed. On the other hand, the occurrence
of the term here should be seen against the background of its
word context. The basic theme presented in vv. 11 and 12a is
that of plentiful quantities of wine: the repetition of the
metaphor (five times!) turns it into a hyperbole. The
introduction (v. 12b) of a new abundance element (3t>h) and
consequently a new colour element seems surprising from the
literary-critical point of view.
There is a literary convention in biblical poetry of
pairing off S^fl and Vas components of a parallel (or joined)
pair signifying agricultural abundance. Although l!?n is more
57
frequently paired with KD*T or MNtih we find !>' and 3^D in
Song. 5:1 - ^h dj> ^^ TPnty and Isa. 55:1, V T»nn N^m
3bni. 0*0y well established as a componental equivalent of V>
58
in poetry and in the prophetic books, is paired with 3t?n in
59
Jo. 4:18. In all these instances "pV/^bh serve as symbols
of plentiful food. In other words, the introduction of the
'abfl' and 'colour' aspects in our verse were dictated by
literary convention, not by subject matter alone; and as 'wine'
is not necessarily 'red' (DTK), and >^*tOfl means 'dark', the
contrast 'dark/dazzling' (= 1^) is secondary, literarily
TT
speaking, the consequence of the organization of 3^h//T»'> into
a unit of parallelism.
The exact meaning of the ti of Vfc .U^hn has been much
60 •» - • ' T-r-
disputed. The best solution is probably to render with Ibn
55 61 54
Ezra, Dillmann, and von Rad, from abundance of wine/milk,
or: with wine/milk. Because Gradwohl brings to the discussion
62
his preconception of IS^ as a derivative of a lexeme for 'milk',
he has no alternative but to champion the comparative usage of
88 BIBLICAL COLOUR TERMS

the Q.63 His interpretation can be rejected on literary grounds:

the hyperbole is centred around the idea of abundance and
fertility rather than that of colour analogy, which is only
incidental to the main theme.
2.4 EX. 16:31 (JE) tym nn»fl>o ittytn pt> TA yiTD
TT T —V
Kim
(Num. 11:7 - h^Tjiii 'PJD i:pyi Kin *n yin torn)
There is virtually a general agreement that "r> refers to
'coriander', and that "p!? 1A yiT refers to the yellowish-grey
exterior of the coriander seed.64 If we agree with Kohut, who
65
says -Pittn Mtnnn Kin l^, there is no conflict between our
description and that of Num. 11:7. The colour of h^TS - 'a
resinous substance, transparent, gelatinous, and commonly
66
yellowish in colour' - would then be included within the
referential range of ll!?. However, if we narrow the range to
13t> = 'white', problems of interpretation can hardly be avoided.
Thus Rashi is obliged to comment that the first part of the
comparison - *TA y*lT!3 - describes the shape which is, in addition,
67
white in colour. Ibn Ezra does not decide whether 'coriander'
is the correct identification for TA and prefers to declare
f.0
that he 'does not know'. Gray finds a contradiction between
the two verses, and says that two different colours are
69
described and mentions how Josephus overemphasized the 'white'
70
element, to the point of comparing it with snow (Ant. Ill, 1:6).
However, this apparent difficulty is removed when the designa-
tion of *p!> is understood properly.
2.5 Lev. 13: (7 times) "p^ Clfln) 1)M (w. 3,4,10,20,21,25,26) .
(3 times) n^ inna (vv. 4,24,38).
(2 times) n:n!? JlNta (vv. 10,19).
(2 times) "\tt
T'.
Iflnj , "Utf
TT
13n y*S
— -,'
(vv. 13,17).
(i time) *pw
TT ;
isnai »n im (v. 16).
2.5.0 Four variables have to be considered for the diagnosis
of skin and hair diseases which may render the sufferer
ritually unclean:
a. The colour of the sore (y£3). That 'colour' includes
factors of brightness and not only hue is apparent from the
linguistic coinage of prp and 0*1(13, which within biblical
~" '•"'" 71
literature are unique to Lev. 13-14, and from the syntagms
rb:tf
T; nins•• nina
TV (Lev. 13:39), y^n Tnns (v. 6), and cnina)T-*
nns
(v. 28).
PRIMARY TERMS 89

b. The texture of the sore and its position on the skin.

72
flXty is an 'eruption, swelling';nt130 - an 'eruption' or a
.": , , 73 74
'scab'.
c. The colour of the hair sprouting from the sore or
scab.
d. The spreading of the disease over wider areas of skin
during a certain period of time. This last factor lies outside
the scope of the present thesis.
2.5.1 For diagnostic purposes, 1^ 1^ is ah indication of
skin disease. Like ifiS On.} - vv. 30,32,36) - but in contra-
distinction to *ihB (vv. 31, 37), whose renewed growth is a sign
that the disease is receding - it represents a deviation from
the norm. Gradwohl is probably right in asserting that the
75
actual colour designated is 'greyish-white', or else it is
simply 'light-coloured'.
2.5.22 The ty of ttm™ "p^ (vv. 19, 24, 42, 43), 7 6 which is
opposed to HJl^ fl'i'1!! H~li^ ( v - 39) , is of similar diagnostic
value. Mainly because of their obvious secondary structure
(syntagmatic vs. monolexemic) tnttTK *p^ and n^S1? ftlfiD filfp
will be assessed as non-primary, and their denotations discussed
elsewhere. For our immediate purpose, however, it is significant
that dTttTN "p!?, although an independent colour designation, is
presented as visually close to 1^.
TT
According to Gradwohl the reference of "p!? in these vv. is
determined by the understanding of the syntagm ->^KD (n))Tiin3
(Ex. 4:6; Num. 12:10E; 2 Ki. 5:27), which he interprets as an
77 70
indirect colour indication analogous to "p^ in our chapter.
This equation of 'snow' = 'white1 is not in accordance with
Mishna Neg. 1:1, 2 (cited above79), where four degrees of
'whiteness' are mentioned, only one of them comparable to
snow. (We must remember that the cultic (cleanliness) context
of the passage would exclude the possibility of a loose
comparison such as that of the parallel pair >^W//*l)3^ in
poetic contexts). Furthermore, theftjnifmentioned here and
elsewhere in the OT usually signifies 'psoriasis' as well as
80
severe cases of other skin diseases. In his authoritative
article on biblical t\y*\X Hulse states that "snow-like
desquamation with underlying redness was the chief character-
Q1

istic of nyi^"> as is gathered from the text itself. In his

90 BIBLICAL COLOUR TERMS

own words, "The biblical evidence of the nature of nyiS is more

extensive than is at first apparent. The disease was compared
to snow, not because the skin was white ('white as snow' is a
mistranslation) but because the most characteristic sign of
the disease was the presence of scales which rubbed off the
82
surface of the skin like flakes of snow". The poetic idiom
U^!?'' >!?UD (Isa. 1:18) should not, then, be used as a yardstick
for the prose-anchored A^KJ3 yi'^ti; and the extra-linguistic
referent (psoriasis) indicates shiny-silvery scales, together
with inflamed areas (dTOTX 1^) and 'bleeding spots' C>n 112D),
which are characteristic features of the diseases grouped under
07
the blanket-term fijns.
To conclude: l!l^ in this chapter is not 'snowy-white'. It
TT
denotes a property of brightness more than of hue, and its
application was therefore conceived of as rather inaccurate
colourwise by successive generations (see the Mishna). Perhaps
it should be translated as 'light-coloured', 'pale', which is
much more appropriate, unless again we always bear in mind that
biblical 13t? is of a wider reference than our 'white'.
TT
2.5.3 Noth says that "at least the basic form of both chapters
can be claimed as'pre-priestly1 ... A more exact dating of
the first draft and of its further amplifications is no longer
possible". 84 If we accept his opinion, or that of others who
85
champion the relative antiquity of P or its sources (against
the Wellhausen school), the general, wider scope of 1^!? is
TT
further explained as being quite early, possibly from pre-exilic
times, when the focus of the reference is still concentrated on
OfL
the dimension of brightness rather than on hue or lack of hue.
2.6 Qoh. 9:80^^3^7 VUS I^M* Jiy fcto : d*^ d»*m = 'white
87 88
clothes' were looked upon as representing purity, festivity,
or an elevated social status. 89 Interestingly enough, the
syntagm d'J^ D^IAll does not appear in the bible again. Its
topical equivalents are wp' 'T>3 (Gen. 41:42 etc.), V-13 1*1311
(Esth. 8:15), or V-ia ^y» (1 Ch. 15:27); 1-1 HI H*M1. . .Win!?
(Esth. 8:15); and d'fltja *033» (Ex. 39:28). Of these terms
denoting 'white clothes' (made of certain materials), MH has
V-13 (Bab. Yoma 7a), W3 '1>^ (Bab. Mets. 29b), but mainly
T:
90
pt> 'TAil, an idiom that is similar to that of our verse. It
seems, therefore, that d'J3!? d>1>3 is of a late usage, properly
PRIMARY TERMS 91

belonging more to MH than to biblical vocabulary which has

other terms that the syntagm "p!? ">1M supplants in later
91
language, inasmuch as V'l^ is a later substitute for WB. Thus
the syntagm Q^J3^ d">T>3 (although its individual components are
quite frequent in all strata of biblical literature 92) is an
indication - one of many - for the relative lateness of the
language used in Qoheleth, and for the close affinities between
the latter and MH. And, lastly, the term's closest equivalent
in the OT is to be found in (late) Biblical Aramaic, Dan. 7:9 -
"inn tfi\3 bmi1?.
T- -:• " ;
2.7 1^ Hif. (Isa. 1:18; Ps. 51:9; Jo. 1:7) = 'become, turn
PV.
2.7.1 Isa. 1:18 -l^a!?! >^3 tPJKD d^NOn "PH* DN
•prp in>D jtfntt) wno DN
Ps. 51:9 1*??* J^Wni >JW3H
The Isa. 1:18 metaphor of repentance or reversal is based
on an DTK - l^!? contrast. The occurrence of Via!? in a verb
formation which appears twice in the Hif. conjugation within a
passage that can clearly be attributed to classical Hebrew of
stage 2 (pre-exilic), constitutes a proof for the productive
potential of the nominal form lit* and its status within its
sector although, admittedly, the occurrences of "p!? verb forms
are far from numerous. 93 The specification of is!? the verb
TT
.IJ1*^1^ designates is stated directly: >^3. In contradistinc-
tion to the syntagm tfWS (n) yiXti - where the texture, not the
colour, is the focus of the comparison (2.5.2) - wherever the
actual colour of snow is evoked, an adjectival or verbal form
of "p!? was felt to be necessary. Two of the three Hif. forms
(Isa., Ps.) are contextualized in this manner. A similar idiom,
this time with a nominal, occurs in biblical Aramaic - RUttU^
1-1 h A!?fl3 (Dan. 7:9).
T- T: •
The Isa. 1:18 'pV/dTN contrast (of various specifications:
94 TT T
IQito, J\!?t!D // (d)>:ieO ,yt?in3 WTK'O is particularly suited
to the context. If the passage (vv. 18-20) is a continuation of
the previous section and not a randomly appended independent
section, the visible contrast between blood (of the slaughtered
animals, vv. 11 ff.) and white garments (of the priests?) in
the court of the Temple is perhaps adequate motivation for the
95
prophetic utterance.
92 BIBLICAL COLOUR TERMS

The figurative equation of snow-white colour and purity,

or lack of sins, is repeated in Ps. 51:9 and in Dan. 11:35
(the latter with another verb formation, see below, 2.8.1).
This usage is probably the link to the semantic shift apparent
in the Pi.(?) and Hithp. occurrences in Dan. 11:35, and 12:10
96
respectively. However, the Hif. form in our verse should not
be considered as 'late', or as belonging to the post-exilic
(stage 4) period. Although Gradwohl dates the psalm to the
exilic or post-exilic period, other scholars do not necessarily
97
agree with the opinions he cites and accepts. As other
scholars point out the psalm, or at least its original nucleus
(excluding vv. 20-21, or 19-21), reflects prophetic teachings
of the pre-exilic (stage 2) period. 98 Perhaps the fact that
l/ISt? Hif. forms* (the comparison to) >^t>> are attested only
twice is an additional indication that they belong to the same
chronological stratum, inasmuch as they are embedded in the
same ideological context, that of repentance. On the other
hand, one must admit that the usage in Ps. 51:9 could be a
literary borrowing from Isa. 1:18.
2.7.2 Jo. 1:7 hfl^ »tlJKm hn^ ^£3A W
qq . *
rm^iiy -iJ'a^ i»!wii notyn <v^
The colour described is that of vine branches whose bark
I ] QQ
has been stripped by locust. As in Gen. 30:37 - "pfcn qwhtt-
101 102
'grow white', 'bleached', or 'pale' are all possible
definitions for the reference of 13*3^M.
2.8 \/T55 Pi. ? (Dan. 11:35) and /l55 Hithp. (Dan. 12:10).
103<
Dan. 11'.35 pWl Yia^l Ditt qiltf> l^KD* D^DlWan V31
nymt? inn Tiy -o VP ny iy
12:10 d">:n iin^?i •mtjri'n imm
2.8.1 13!^ is written as a Pi. but punctuated as a Hif. form,
although the preformative M and the "> which are characteristic
104
of the Hif. construction are missing. Is the 'double anomaly'
of the presumed Hif. form an indication that the punctuation is
wrong, and that the original consonantal text represented a
Pi. formation? In favour of this identification we can cite
four factors. One, that other Hif. forms in our text have the
105
intransitive force of 'be', 'grow', or 'become' 1^, whereas
TT
a Hif. form here would be of the transitive ('make', 'cause to
106
become' *p^) type. Secondly, the Hithp. form (12:10) can
PRIMARY TERMS 93

be more easily connected to a basic Pi. form rather than to a

107
Hif. one. Thirdly, there is evidence of a properly punctuated
1 OR
Pi. form in this verse in some Hebrew MSS.
Perhaps more weighty is the semantic evidence: /"p^ Pi.
appears quite frequently in MH in contexts of metal-smelting,109
cleaning, bleaching (wool), and - figuratively speaking -
110
cleansing of sins. Both the word context of 11:35 (and
111
of 12:10) - the lexemes qil^ and Ttt!? 11 - and the literary
context point to a metaphor of metal-smelting and polishing,
which stands for the principle of moral trial and selection.
Consequently, IpW here cannot be considered a full-fledged
colour term. Properly speaking it belongs to the field of
metal-processing (and, according to MH evidence, wool-processin
more than to that of colour: it designates a derived sense of
'purity, cleanliness1 that is the result of the equation lUt?
112
(see also 'tPDn!? D'»'TA3') = purity.
2.8.2 Similarly, cleanliness and purity are denoted by .IJa^JP.
Therefore a translation based on the lexeme 'white' or on an
equivalent thereof will be inadequate. 'Will make (themselves)
pure' or 'clean' is a far better rendering of the meaning of
113
the lexeme. On the other hand, this obscures the fact that
133^fl* carries an 'abstract', figurative sense that relates to
the symbolic properties attached to the extra-linguistic
referent of "pXTT
and especially
e J
to the focus of the term
('white' -» 'clean' —» 'pure').
2.9 Summary
2.9.1 "p^ is a primary term which designates any colour
TT
property from 'clear, light in colour' to 'white' proper.114
Therefore, its exact denotation is specified, when required, by
a comparison to an object whose typical feature is 'whiteness'
(>t>l!0, or by a syntagm of a secondary or tertiary nature
(nmttlK rtf:tf, Lev. 13:19 ; im!? niro, Lev. 13:39).
"115
There is no reason to believe - with Gradwohl - that
"p^ originally referred to 'milk'. Neither the text nor the
evidence of cognate languages support this assumption (2.0.1).
116
On the contrary: if - with Gradwohl - we accept that ftZ^1? =
1
'brick and its verbal derivative v/l^ II = 'make bricks'
represent a polysemic development from our "p!? through metonymy
TT
of the referent (= the 'light colour of sun-baked bricks'117),
94 BIBLICAL COLOUR TERMS

118
and is not a loan-word from Akkadian - then what the colour
"p!? refers to should be understood as 'light' in general, whereas
the colour of milk is but one segment within the area covered
by the primary term. Gradwohl himself is aware of the fact that
*p^ is wider in scope than "weiss", but stops short of drawing
the right conclusions.
2.9.2 Berlin and Kay have established the synchronous
!
appearance of 'dark'//'light' — white'//'black' as basic
antithetical pairs in most, if not all, world languages. 119
It would have been tempting to try and reconstruct a basic
meaning of 'light' for "p!?, 'dark' for Iflty (section 3. below).
TT T 0
However, this hypothesis is not supported by the text nor
by cognates as far as "p^ is concerned, although it might be
approached a little less hesitantly in the case of IhW.
PRIMARY TERMS 95

B.3 iconta
3.0.1 int!> derived primary colour lexemes appear in the OT
8 or perhaps 9 times (see below). Their distribution is as
follows:
inty - 6 times: Lev. - twice (13:31, 37); Zech. - twice
(6:2, 6); Song. - twice (1:5; 5:11).
"inw - (Qal lexeme) - once: Job (30:30).
T)t"H27 - (nominal, name of colour quality): once in
Lam. (4:8).
In addition, 1HW3 (Jo. 2:2) should probably be read
l'ne>3 - inu) = ' d a w n ' , ' i n the word context of DT> n!?OK1 "|tyn D"P
' ""* 2
^fliyi 13)7 (v. 2a) and within the description of the descending
locust multitudes, does not make much sense.
As far as syntagmatic relations are concerned, Ihty and its
immediate derivatives exhibit little variety. They modify
Iyj9 = 'hair' 3 times (Lev., Song.), 'human skin" or 'complexion'
- twice (Song., where the grammatical subject is the pronoun
"OK, once; Job, where the direct subject is »1iy - once), and
'horses' hide' (Zech.) - twice. Actually, apart from the
amended nVlWD (Jo. 2:2) and perhaps Tlhty (Lam. 4:8, section 3.4
below), the term refers exclusively to the appearance of human
hair/complexion and human or animal skin. This range is quite
limited - in MH *itny modifies 'clothes' and more. Barring
accidents of text preservation, "inty does not seem too versatile
for a primary colour term.
3.0.2 Productiveness: secondary/tertiary lexemes related to
"int; are irnniy* (Song. 1:6, in fact the equivalent of rmnty in
the previous verse) and n-TlhW (Qoh. 11:10), if the latter is
not to be classified under 'colour allusions'. In MH the term
is much more productive: JnVlK/fl, IPYnnta, mimhW, lin»U> (= 'coal')
are a few of the new nominal creations, while the Hif. verb
3
pattern is extensively used.
3.0.3 Gradwohl states that 1tWT appears only in texts of the
4
exilic and post-exilic periods. The question arises: is there
a lexeme which fills the same slot in earlier biblical
literature, and which complies with the criteria for primary
colour terms (II, A)? DIM, which is Fronzaroli's candidate for
5
this post, occurs only in Gen. 30 and is limited in its
application. y/ivyh and its derivatives seem to occupy a mid-way
96 BIBLICAL COLOUR TERMS

position between the 'light/darkness' field and the colour

field - the interpretation of .1W1 in Lam. 4:8~depends, to a
large extent, on the comparison to 11HW. Even v/TTp derived
7
terms, which cover an area of distribution wider than ^IhEf,
are not specific enough: Trj? '^tVO and >3K nilMB (Song. 1:5)
T •' T ;
support each other's colour specification, but the dominant
factor is the explicit mihiy. Therefore all three terms -
Din, the verb v'lUTl and the verb t/Y7p (and their derivatives)
are assigned 'secondary', 'tertiary' or 'indirect' status in
the hierarchy of the "int2> sector.
3.0.4 Etymologically IhW seems to be fairly opaque, which
is another indication of its primariness. Gradwohl points out
9
that TirPB^KIirPW means 'coal' in Aramaic and in Syriac.
Therefore, he concludes, the basic meaning of IhtJJ is derived
from linty - 'coal' '(discussed below, 3.4) and denotes the
'colour of coal1. He also maintains that v/^nt; in Hebrew is
a loan from Aramaic, although the 'colour' signification as
10
such is original to Hebrew. This hypothesis, however, is by
no means certain. IhW, as the Versions show, is far from
T
denoting 'coal-black' exclusively: Gradwohl himself lists the
evidence of the LXX, which has 'soot' for YinB (Lam. 4:8),
although the V has 'coal'.11 The Targum has Kllh^W for Vina,
-r
but usually D3-1K for IhW, although t/1hty does feature in
Aramaic and - according to Gradwohl - might have entered Hebrew
through that language. 12 On the other hand, it is conceivable
that 11 f\\0 is a derivative of inu)
T and not vice versa, like DIN
' . "
from OIK, "pt? (MH) from "pt?. Alternately, although 11 HK) -
which is a hapax legomenon in the OT - is possibly a loan word
here, IhB is common to the stock of both Hebrew and Aramaic.
r
Kllh'W in Aramaic, Syriac and Mandaic could constitute a
secondary development (from verbal y/1fW) in those languages.
Finally, an alternative basic meaning which is much more
general in scope, and therefore much more suitable for 1hK>
as a blanket term, is proposed by Tur-Sinai. According to
him both V/1HK7 and /TTp basically signify 'to burn1, and the
colour term 1HE' primarily refers to the colour of burnt objects;
any other interpretation would make the syntagm t'yyn IhW
(Job 30:30, section 3.5 below) impossible to explain grammati-
13
cally. This wider definition entails references to a range
PRIMARY TERMS 97

of colours from 'ash-grey' to 'black' proper. As such, its

scope is wider than that of 'black', or Ihly in Modern Hebrew,
where Dlh and 10KT cover
,
sections that in OT Hebrew are
inclusive to the IhB scope. Indeed, the usage of flirnhW
(Song. 1:6) as a stylistic variant of TTTinw (1:5), together,
with the extra-linguistic reference of both nilhW and nihlhW
in this context (3.3 below), fits in with Tur-Sinai's analysis.
3.0.5 Psychologically speaking, 'black' is a specification
14
of the concept of 'dark' or 'darkness1. Has this process of
differentiation of psycho-physical phenomena left any
recognizable traces in the language of extant OT texts? One
of the clues is the usage of /HWh and /"Ihty derived lexemes in
parallel formations, viz. as approximate literary equals. So
15
is the case in Lam. 4:8 - lihBW ,1K>h and,
AND if we accept the
amended vocalization, in Jo. 2:2 - where H^h, fi!>SK v(and 13)7
, \' -T"~:
^Siyi) have references similar to that of the conjectured iruy.
The line of development 'dark' —? 'black' is further illustrated
by /YTp-derived terms and usages (classified as subordinate and
16
indirect terms) or by the lexeme 311)7 (from ^Uiy = 'dark').
«
3.1 Ihty as modifier of 'hair'.
T ,
3.1.1 Lev. 13:31, 37 IHiy 1)>ty
iVliy -IJ7KJ is a sign of health, in contrast to aih*
T
lyb
T T- 17"r"
(vv. 30, 32, 36) which is a symptom of hair disease. As in
the case of other Ihvy occurrences, there is no reason to
T
restrict the reference to 'black' only: a 'dark to black'
range is much more suitable. The opposition in^/VlhUJ basically
signifies a 'light-coloured' = 'sick' vs. 'dark coloured' =
'well' contrast. The same semantic range ('dark1 —t 'black')
is reflected, for instance, in the usage of the Aramaic cognate
10

CD1N and verbal /DDK constructions.

3.1.2 Song. 5:11 3Tl)O nilhty D^l^ft 1»D121|7 T3 WlD HMO
On the other hand, 111)73 niiriW is much more specific:
the designation is defined as 'black' by lliy. The latter is
19
derived from a root meaning 'to be dark1 which, as such,
contains an indirect colour reference. The description of the
beloved's head focuses upon the ti*TN (+fl¥) ('complexion')
4
^ 20
(v. 10)// inty ('hair strands') contrast. Therefore, there
T
' ' 21 22
is no reason to delete 311)7D nilhW from this verse. T3 dflD
at the beginning of the verse, should be understood as a
98 BIBLICAL COLOUR TERMS

reference to the shine or gleam of the beloved's hair rather

23
than as a chromaticity reference.
3.2 Zech. 6:2, 6 D'-jne; tPOID
The colour opposition here is to tPaTK, .d'iat' (and 0^*1?
tPSON), all designating the colour of horses. The 'original1
colours of the horses in Zech. 1 and 6 are extensively discussed
24
by Fronzaroli.
3.3 Song. 1:5 niKn "OK rrnntu
(v. 6 umi '^nflTow niVnnu; 'OKU; •oiK-m bK)
The insistence that illlhu; - of complexion tone - is beautiful
stands in sharp contrast to the conventional idea of beauty -
DTK ,fi¥ ,Ht* - elaborated upon in Song. 5:10f. and Lam. 4:7f
Perhaps this should be understood as a juxtaposition of urban
beauty ideals vs. country ideals and realities. Be that as it
may, it would be difficult to translate JTTlhW here as 'black',
T .'
especially since JllfTlhB (in the next verse) functions as an
'•' '•"'
explanatory equivalent to the former. 25 The employment of
mmniy typifies the fact that to equate Ihty with 'black',
without an analysis of each particular context, is far from
accurate. Here the reference is specified: nihinu) ,fillHO
refer to the colour of sunburnt skin, that is, 'brown' in our
mode of speech.
26
3.4 Lam. 4:8 D*ltfn linum .1^
* * • ™* T ,
According to the ancient Versions *l1HU) is either 'soot'
(so the LXX), 'coal' (V), or 'blackness' (T).
Gradwohl sees 11 no = 'coal''•27 as the generative lexeme
from which IhO is derived, and hence defines "lino as 'coal-
T
28 '
black'. Against this hypothesis a few factors might be
mentioned. The diversity of the Targums points to a far from
certain understanding of TlhW. Unless we assume a loss of
the original meaning, caused by the existence of the more
frequent IHW whose denotation is not "coal-black1 only, we
must accept that TihUJ is an overloaded lexical item. While
Aramaic (like Syriac) does have K'nh'W = 'coal', this is not
used in the Targum here. Rather, the term NH031K = 'blackness'
is employed.
Finally, n^h seems too weak to denote 'blackness' that is
'blacker than black'. These considerations, in addition to
those noted in 3.0.4 above (pp. 96f.), seem to me much more
PRIMARY TERMS 99

weighty than those of Gradwohl. Tur-Sinai's proposal to define

1hiy as the 'colour of burnt things' -> 'dark' -> 'black' makes
T
sense.29 Consequently, TiriUJ would refer to 'blackness' or -
in this instance - to the more general 'darkness'.
3.5 N/^TFuy Qal
Job 30:30 yv\ >jn mn >B2)n f'yyn inty my
Similarly, an interpretation of IhW = 'was burnt' fits
both the word context (31h ''Jtt mil) and the extra-linguistic
reference much better than 'became black'. This is reflected
in the Aramaic Targum and in the LXX. The former uses OhW,
which means 'to be hot', 'dark' and 'black' in the Targums, in
30
Talmudic Aramaic, and in MH. The latter has eskototai, which
is also used to translate 1W1 in Lam. 4:8.
. -T
3.6 Summary
-lhe> in Biblical Hebrew signifies 'black1 (3.1.2, 3.2, and
TlCM - 'blackness' - 3.4), but also denotes the less specific
colour quality of 'dark' (especially when describing complexion
and skin colour, 3.3, 3.5; or hair colour, 3.1.1). This is
precisely the range of reference that should be expected in the
31
absence of terms referring to 'grey' and/or 'brown'. As for
its basic meaning, one should remember Berlin-Kay's empirically
proven conclusion: "... terms for 'black' and 'white' lack
known derivation". 32
100 BIBLICAL COLOUR TERMS

B.4 pTV and pV

Job 39:8. U>YP pIV to IhKI .iny*l» D^fl HuP.
4.0 Although pTT» as such appears only in this verse, its
inclusion within the primary category is supported by the
existence of related lexemes on other levels of the field.
4.1 The secondary formation plp/l* appears three times - twice
in Lev. (13:49 and 14:37, in descriptions of cloth/leather and
1
building material 'diseases'; and in opposition to OTfiTK) and
once in Ps. (68:14, in the syntagm V-11R p1pT>). Even if the
2
occurrences in Lev. may be considered relatively late, this
is not easily done in the case of Ps. 68. The latter contains
3
many points of contiguity with Ugaritic- Canaanite poetry;
4
hence, in substance it should be dated as pre-exilic. In
addition, the Ugaritic and Akkadian equivalents of Yllh PIpT -
5 6
yrq. hrs and gurasu ar-qu respectively - have the basic form,
not the derived duplicated one (as in Hebrew). The basic
sequence VPT*. or cognates thereof, function as colour terms in
many Semitic languages, while their range of reference
encompasses an area which stretches from 'pale silvery' to
7
'green' and 'yellow'. Therefore, Fronzaroli's suggestion
that p1pT> had existed in the language - after being formed
by analogy to d*rmK - before PIT filled the primary slot
o
superseding this section seems to me far-fetched. Rather,
I would assume that the fact that pll' is a hapax can be
attributed to accident. As Albeck says, although plV ,flp11*
and p^lin (Hif.) look like new MH creations they probably belong
' 9
to the stratum of ancient Hebrew.
4.2 The availability of other /PT derived terms - llpl?,
10
(1. 'mildew', 2. 'paleness' of face), the river name llpV,
11
and perhaps the place name DypV - points to the primariness
of some v/P1*-related lexeme at the head of the 'pale, yellow to
green' sector. The examination of the terms pi' and pi*, which
, V" TT
unlike pll^ appear in texts from the pre-exilic period onwards,
is of considerable importance for our problem.
Gradwohl summarizes the list of pi* occurrences by saying
that the lexeme means 'every green plant that grows quickly'.
The impression is that he interprets it simply as 'green
12
plants'. However, he does not comment explicitly upon the
PRIMARY TERMS 101

fact that in the majority of the cases (6 times) pV/pV appear

as the first component of a syntagm, be it 3W^ pV (Gen. 1:30,
9:3), !TTK>n p"P (Num. 22:4), or NBTT PT/PV (2 Ki. 19:26 = Isa.
13 ""!"•''.'
37:27; Ps. 37:2). In the two remaining passages - Ex. 10:15
and Isa. 15:6 - p_*l> appears on its own, but the notion of 'plant'
is supplied by the context (mwfl 3Wy ,YV in the former) and the
parallelism (T»^h ,NUTT in the latter). Logically (if not
syntactically) the nucleus of the syntagm in all cases but
Num. 22:4 is not pV, which is the first component, but 1U))> or
14
Ken, while pV functions as an attributive of qualifier. Thus
3V)y pT> can be translated as if it were a nominal phrase
15
containing an adjectival element + a nominal form. pV , then,
denotes a colour quality. Historically and etymologically it
might have denoted '(fresh) plants'. Nevertheless, in our text
this reference is preserved clearly only in Num. 22:4 - man pT>
- which is the equivalent of man 3tWy in other passages (Gen.
2:5; Ex. 9:22; Jer. 12:14; and more). As for Ex. 10:15 and
Isa. 15:6, there pV conveys the colour attribute, while the
notion of 'growing things' is supplied by other words within the
same contexts.
To summarize: although p^», from the pre-exilic period
(stage 2) onwards, denotes the colour qualitites of natural
growth, it is one step removed from its probable original
meaning. In that sense it demonstrates a semantic shift of
16
specialization, or narrowing. This might explain, for
instance, the consistency which is demonstrated by the Aramaic
Targums for the translation of pY>. = plV, Npll* in each and
every case, while the similar term Kplt. is reserved for the
translation of pT>. The latter can enter different syntagmatic
relations - p*l»(n) "U (1 Ki. 21:2; Dt. 11:10) or p"P nhllK (Prov.
15:17) - and is then understood as referring to 'herbs' or
'vegetables'. Hence, a 'pure' colour definition for pV seems
to be valid.
4.3 As for pIV (Job 39:8), the construction of this colour
term from the basic pV, possibly by analogy to DTK, is self-
explanatory. Even in its context - that of 'food for the wild
ass' - there is no necessity to define it as 'green things'
10 1Q
(= PV) or 'green plants'. If the LXX and V translate as
TT
20
they sometimes translate P"1?- which Gradwohl points out - they
102 BIBLICAL COLOUR TERMS

do so because both lexemes refer to the same colour quality,

while the employment of plV in connection with "natural colour1
(of growth) only may very well be arbitrary (see above). The
existence, on various levels, of derived terms - one of which
(P*lpV) not necessarily a reference to 'natural colour1 - and
of numerous cognates militate for the acceptance of p"P as a
basic colour lexeme in pre-exilic literature, to be later
supplanted by P1"l*0
4.4 The approximate extra-linguistic colour reference of pV
21
and pl"l* is not easy to determine. The cognates supply a
1
wide range: from 'pale', 'silvery' to 'yellow , 'gold', and
'green'. The area covered by the Hebrew term and supported by
the evidence of HPTJ, and VI 1h P1p*l> could be chromatically
defined as 'tan1, 'yellowish green1, 'green', 'greenish' - not
a highly saturated or strong chromatic entity,, Indeed, this
further supports the contention that pV. is used loosely: the
'green' it signifies includes areas later ascribed mostly to
3i1i? = 'yellow', 'pale'. In essence, this state of affairs is
in accordance with the Berlin-Kay evolutionary hypothesis. The
latter-have shown that after terms for 'white', 'black', and
'red' are established in a given language, the next lexeme to
22
appear is one signifying 'pale to green' or 'pale to yellow'.
Pines summarizes the matter neatly by saying that the name for
the 'green1 colour related to the name for plants originally
referred to all chromatic qualitites which can be observed within
the world of vegetation. 23
PRIMARY TERMS 103

B.5 in*
T
Lev. 13:30, 32, 36 Sru IJW
T T "
5.1 If SilS is here accepted as a primary term (of sorts), it
is for the following reasons:
a. Morphologically, it is built on the model of the other
chromatic value carriers such as DTK ,inu> ,plV<,
1 T T T
b. It is monolexemic.
c. Its signification is not included in that of any other
colour term of the same context and of the same chronological
period.
d. The related Hof. part. in;*)3 (Ezr. 8:27) and its
explanation - SilTD IVMtttl - are witness to the growing status,
albeit on a modest scale, of the lexeme during the exilic period
(stage 3), if not before that, depending on the chronological
placing of the P literature (Lev. 13).
e. In MH the term is narrowed further. It is applied as
2
modifier to lexemes other than "15W1, and evolves into a full-
fledged freely-used term whose reference is not open to various
interpretations. Therefore, we must admit that the process of
development which we first witness in Lev. 13 (and in Ezr. 8:27)
takes place largely outside the OT.
5.2 Gradwohl bases his definition of the reference of the term -
'reddish yellow' - on etymological and semantic links between
3
Hebrew 3M2 and 3MT, Arabic ^-^/~?' and _-—^> . In his
review of Gradwohl's book Blau agrees that both root sequences
exist in Arabic, but adds that a ci^s shift never occurs in that
language. Rather, he argues, it seems that the two roots are
historically unrelated, and that because of their similarity in
4
form they became similar in reference too.
5.3 When discussing the Versions for IlVi^ in Lev. 13 Gradwohl
compares the Aramaic Onkelos translation (ptt^O) to the render-
ings of LXX and V and concludes that while the Aramaic
understands 3CW as 'red' (Jastrow: '[dark] red' ), the other
two tend to translate into 'yellow', 'gleaming yellow', or
'yellowish-red'.6 He does not refer to the fact that the
evidence of Onkelos here is quite isolated, and that the other
major Aramaic translations agree, in effect, with the versions
of LXX and V. Thus,
Onkelos: pOID (1)TC>) "iyo7
104 BIBLICAL COLOUR TERMS

Pseudo-Jonathan: 3m I f h D nrtom "iyt>) (30, 32) and

3nt>2tt> 1)W (36) 8
9
Neophyti I: Ih^O lyty (30, 32), toil^sn my<!> (36)
Syriac: KV1QK) IJW10
Jastrow defines Uflt?^ - which features both in MH and in
the Aramaic Targums - as a Pal'el form of /ifii? and the meaning
11
he assigns to it is 'redden, glow, heat, gild 1 , etc. It
looks as if this definition is unduly influenced by the pn>10
of the Onkelos translation. Moresheth's definition - ,3h2in"
"p^l^O (= 'yellowed, gleaming') - seems to be much more
12
appropriate. Furthermore Moresheth, after examining various
contexts in which DM^ forms appear, states that in essence
Hhttt probably means 'burn in fire', and therefore is close in
13
meaning to in^UJ and in reference to 'yellow' = 3ft2.
I think that while Dh!?KJ is a Shaf'el form, 3nt?!tf is a case
of morphosemantic contamination: could it be a combination of
/HhS and /^n!? which refers both to chromatic qualities ('pale'
14
to 'tan' or 'reddish1, within the scope of our 'yellow1 ) and
the notion of 'brilliance'? Thus most of the Aramaic Targums
supply us with the referents 'gleaming yellow', 'like gold1 -
not necessarily 'red gold1 - that is, what we would call 'blond1
hair. As Pines points out, this is the signification of the
15
LXX and V terms as well.
Coming back to biblical Hebrew, the existence of roots
denoting 'gleam', 'brightness', or 'shine' which are identical
to j/UMif in their first two consonants but different in their
16
third consonant - II /5fTif ,i/"nfi¥ - strengthens this
interpretation, although the last two roots belong to the
•light, shine1 field and not to the colour field.
PRIMARY TERMS 105

B.6 SUMMARY
The primary, superordinate colour terms in biblical Hebrew
are (according to a descending scale of distribution):
1. DTK - from the pre-exilic period onwards; in all types of
OT literature; with the chromatic reference of 'brown-red-pink'.
2. ^"y - from the pre-exilic period onwards; again in various
types of literature; denoting 'brightness, brilliance' and the
neutral hue qualities of 'pale to white',
3. inty - from the exilic or the pre-exilic period (depending
on our dating of the literature underyling P : in juridical-
cultic (Lev.), poetic (Song.), and prophetic (Zech.) texts;
refers to 'black', but also pre-colour 'dark'.
4. pIV and pV - pV from the time of the monarchy, 171T>
only once (Job); here the occurrences supply us with an insight
as to the process of the development of a basic term. Like
their related cognates, the area covered is quite large: from
'pale' to 'yellowish* and 'green'.
5. unif - only in P and a derived term in Ezr. (post-exilic).
T
Reference - 'pale to (golden, or reddish) yellow, shiny'.
Hence, we conclude that until the 6th century B.C. Hebrew
colour lexicon - as transmitted to us - exhibits features of
stage III status according to Berlin-Kay ('white', 'black1, 'red1,
and 'green'), while after the exile the evolutionary process
reached stage IV of their classification (the addition of
'yellow'). Henceforth, the terms which are subordinate to the
above-mentioned five must be analyzed and organized structurally.
106 BIBLICAL COLOUR TERMS

C. SECONDARY TERMS IN THE OT

C.O The criteria employed for determining the subordinate

secondary status of colour terms are:
a. Although the term is monolexemic, it differs from
primary terms in its application and/or its specification; or
b. It is morphologically based on a primary term.
c. Its signification is included in that of the primary
term which governs it.
d. Its application is more restricted than that of a
primary term.
The various lexemes which belong to this stratum of the
field will be classified under the superordinate term of their
sector. Thus, C.I presents the discussion of tiTN governed
secondary terms (tTTttlK ,D^"
W'. ,VttN*
T
T
,p'"N!>*;
-T
for ih^-f see under 1 ^) ;
-,.-5

C.2 is headed by ^ (lh** ,n* and IT*); C.3 - by SlTO (,d<in

I'mnw*);
:
C.4 - by |711> (P'lpl'O ; and C.5 - by i?\X (!invn) .
•~ T T«. \
C.I SECONDARY TERMS WITHIN THE DTK SECTOR
T
l . 1 DtmiK
1.1.1 OWTK occurs in the text twice (Lev. 13:49, 14:37) on
its own and four times as a component of the compound DIOTK "p^
(Lev. 13:19, 24, 42, 43). dTnTN per se is directly opposed to
plpv and describes either cloth disease (13:49) or 'disease' of
building materials (14:37). tnOTN "pb, on the other hand, defines
the appearance of skin diseases - be they HK2; (19), JTif|3 (24),
y*l (42) , or yAJil JINK; (43) - and stands in opposition to ]&.
Therefore the two terms are discussed separately: DTttTN here,
and DTniN "p^ under Tertiary Terms.
1.1.2 Although dTOIK is genetically transparent, its actual
colour reference is far from clear. The ancient translations
are of very little use for this problem; one gets the impression
that no firm distinction was made between DTK and QTMTK ,i7lT)
1
and plpl*. The problem irked Jewish Commentators from early
times: is .d*TOTK a strong, bright 'red', or is it a pale, dull
'red'? In other words, the point of departure was: dTBTN
introduces a difference in the properties of brilliance and
saturation (rather than of chromaticity) , but which end of that
scale does it refer to?
2
R. Aqiba states, d'&n AlTnn T»D ... dTttTK. This means
that for him the lexeme denotes a diluted, less saturated form
SECONDARY TERMS 107

of DTK. R. Yishmael describes the same quality (llflan = 'the

mixed colour') as tfm JflTO 1»3 (for the tnttlN •}& of the
3
mra = Lev. 13:24) or D^nn ^ITnn tm (for the JttJ). They both
agree, then, that 0*T)3*TN is a diminutive form whose exact
designation can be conveyed by describing a mixture of a
typical chromatic base (blood, wine) with diluting, de-saturating
agents (water, snow, whitewash). However, we should bear in
mind that both R. Aqiba and R. Yishmael are discussingOTOTK "p^
4
in the context of skin diseases and not tntflK on its own. The
Talmud defines OTttTK itself as a strong, sharp DTK:
T
5
npiny irninT nr dimity trrcrm innx
6
Following this definition, Rashi explains,
7
D'p'n^B piT> ,pi|7T> .tpmiioiy tmx ,tnaiK
Q
Qimhi concurs, while Ibn Ezra (for plpl'') tends towards
the diminutive interpretation but still mentions the existence
9
of the opposite explanation.
1.1.3 The argument should not be restricted to the question
of the meaning of D'TO'TK ,|7'VT> or Ihlhty* in any given context.
GC C
Theoretically, it should be possible to reproduce a p aJ al/
pe°al ol lexeme from any pa°ol base of the same consonantal
sequence if a grammatical-semantic productive rule covering the
nature of the affiliation between the two is available (beyond
the notion that the partially duplicated form is genetically
secondary). Both denotations - the 'weaker' or, on the
contrary, a 'stronger' reference - appear at first glance to
be feasible. But when we consider similar morphological
constructions from outside the colour field - which one of the
solutions looks more suitable for our context and for the
greatest number of contexts in which such lexemes (like
n»0-ns'> ,!7pt>py ,cilO£JOK, etc,) occur? And are these two the
only possible explanations?
A survey of opinions found in modern research seems to be
in order at this point.
1.1.4 Modern biblical dictionaries are unanimous in
10
accepting that DTttTK is a diminutive form of D*TK: so BOB,
T
11 12
KB, and GB. The same view is presented by Moscati, who
gives tJlpT' and its Ethiopic parallel as an example for the
13
diminutive function of the model. This indeed is the meaning
14
of ti*WTK ,i?1p"P, and so on in Modern Hebrew. On the other
108 BIBLICAL COLOUR TERMS

hand, Guillaumont understands the form as one conveying

15 16
intensity, and so does Tur-Sinai. The latter view is in
accordance with the Talmud but does not explain the apparent
shift from 'intensive' to 'diminutive1 in later literature.
Gradwohl attempts to solve the problem by using tnmN "p!? as
a link; according to him, the weakening force of "p!? exerted a
17
influence on DTniK itself. Pines tends to agree with the
18
latter. This approach, though, does not explain the process
for plpv or imhiy*. Claiming that these were interpreted as
diminutive forms by analogy to d*Ttt*TK - which was influenced by
D1QTN 'pt? - seems improbable, especially since d*T01K and 13^
DTttTK are well differentiated both in the biblical text and in
the Mishna as referring to separate - albeit similar - colour
entities.
1.1.5 An interesting discussion of our problem was launched
by the activity of the Committee for Colour Terms, the fruit
of whose work appeared in 1331^5 VI. Har-Zahav opened the
19
debate. Citing the Mishna, medieval, and other sources he
describes the p al°al pattern of colour terms as a diminutive
model. He too interprets the compound dimK 1l!? as proof for
the 'diminutive' signification. Other lexemes of the same
pattern, such as ISDDn (Prov. 21:8), fllp^ptm (Jer. 23:12),
flippy y (Jud. 5:6), and n»0-n£P (Jer. 46:20) are classified by
him as 'intensive' forms.
The inconsistency of Har-Zahav's approach was, of course,
immediately noticed. Hence, an answer was not late in coming.
20
Zlotnik published his objections in the same volume of 1331E>5,
viz., it is inconceivable that the Sif. contradicts R. Aqiba.
Simply, the words of the Sages were not properly understood: no
intensive or diminutive force can be attached to the pattern.
Rather, a distributive force is meant. dTtiTK refers to 'red
spots'; plpl* - to 'green spots', and filh^hKJ (Song. 1:5) to
'sun-tanned patches'. Similarly lexemes like t>p^py, qiOflDN,
DJOJ j^n^Hfl ,1SD9n all represent the same 'distributive1 concept.
If we accept Zlotnik's arguments, then we must assign the
very same chromatic references which we attribute to the primary
terms also to the secondary lexemes of the same series. This is
generally acceptable for most occurrences, especially those of
Lev. and of Song. (1:5 mfnno, cf. 1:6 mitTO). Less convincing
SECONDARY TERMS 109

is the interpretation of YTin plpl* (Ps. 68:14) according to

this suggestion. In addition, the idea of 'speckles' and
'stains' in biblical literature is usually expressed otherwise -
by using the pa ol pattern (Tp3*, ^V*), although we have
JTlTlirori* for 'speckles/spots'. Finally, no support for this
analysis can be found in cognate languages. 21 This last point
22
was taken up by Shalem. Shalem quotes the unique report of
the Samaritan Targum about the twelve stones of the ,1^M - pOpfcb,
P1J30, 1700 (first row), P'np-P ,P"P ,P1-P (third row),' and so
on - and concludes that these are graded in order of intensity
from the weak to the strong. He applies the same principle to
Hebrew: he sees dtntK as a shortened form of DTKT tntf,T; similar
to duplicated superlative forms like S'OD 3">30 and TNtt TK)3. In
short, for him D*TOTK is a stronger d*TN ,p1pT> - a brighter and
23 •T
deeper p"l*T»; that is, he defines d"TN ,P1T> and so on as
references to the bottom - or the middle - of this sector,
while the morphologically secondary lexemes are understood as
references to the maximally saturated foci of the same colour
areas. This approach, again, is questionable both on
morphological and historical grounds.
1.1.6 I would like to propose an alternative approach, one
that would categorize the 'deep' (if not 'surface') semantic
significance underlying the various terms exhibiting the partially
duplicated pattern, be they colour-denoting lexemes or lexemes
from other fields.
Let us ignore, for a moment, the property of chromaticity
and instead try to define the difference between DIN and D^TK
in terms of brightness. Could DTttTK in the contexts in which
it appears refer to bright DTN? Let us remember that D*TN is
a blanket term which refers to 'pink' and 'brown' as well as
'red': any term derived from it might denote a section included
in it. On the other hand, the derived lexeme may be chiefly
24
motivated by dimensions other than hue. We even half-expect
hue determiners to follow the basic pa°ol pattern (cf.VOK*,
T
pIB*25), although the latter is by no means exclusive to colour
terms. DTttTN = 'bright red' or 'glaring pink', a glistening
and 'raw' appearance, describes skin diseases well, and this in
a context where DTK itself does not feature. The same interpret-
ation - not a change in hue, but rather in brightness/saturation -
110 BIBLICAL COLOUR TERMS

makes sense in the case of P*1p1* both in Lev. and in Ps. 68,
but not for nihinw in Song. 1:5 unless here too it is the sheen,
JS

not only the actual chroma, which is referred to.

If so, the expanded pattern represents a modification to
one or some of the qualities which the base it is derived from
signifies. In colour terms it is the 'brightness/gloss' or
'saturation' factor. In ty\)'y\)y this variable might be the
distributive factor, in ttflMlfl and fPfl-flfi* - intensity. Even
tPni-rtaiy DnK ^p'nmy pIV (1.1.3 above) might be explained
as referring to these dimensions of purity and gloss of the
colours described. This is possibly the reason why no firm
distinction is made by the ancient Versions between the primary
and the secondary terms. As for the modern interpretation and
usage of tiTnTK jp'ip'P ,Shinty as diminutive forms, these indeed
might have been influenced by the inaccurate interpretation of
M. Neg. 1:2 - which was understood to deal with d*Ttt"lK, while in
fact it deals with tn»TN 1^ - and not directly by the biblical
DtnTN Vlb.
1.2 D1KQ (Nah. 2:4), d^niKtt (P - Ex. 25:5; 26:14; 35: 7, 23;
TT; -TT:
27
36:19; 39:34). Commentators tend to agree that OTKtt (Pu. part.)
no
means 'dyed red1 or 'made red'. If so, the lexeme ought to be
considered a member of the secondary stratum: it denotes a state
of being an DTK of some sort. It modifies 'rams' skins' (flliy
D^^N - Ex.) and 'shield' (Nah.), the latter probably made of
' '* too. 29 Further, dWM is mostly understood as referring
hide
to a colour effect that is not a naturally reflected or
30
refracted DTK, but a man-made pigment-derived quality. The
identity of this colour reference, however, is not entirely
clear.
The NEB and the translation in the English edition of
31
Noth's commentary have 'tanned rams' skins'. This translation
is (deliberately?) ambiguous. Does it constitute a reference to
the colour ('tan', 'brownish-yeHow') of processed and dyed skin
which, in the absence of a more specific primary term, would
fall (in biblical Hebrew) under the heading DTK? Or else,
'tanned' may simply signify that the skins had now been
processed, with no overt colour reference included. Indeed,
the latter might be the case as far as d^TXti itself is concerned:
the possibility that it is not a colour term (etymologically
SECONDARY TERMS 111

speaking) should be considered.

In Arabic we find a root sequence 'adm which functions
within the field of leather-processing: .^l.j/ 'tanner',
i> 7 ^ X9 \
C -^ ' 'hide'. In the same language the dominant term for
'red' is not >derived from 'adm, but from a different root
(hmr > ^^rJV'). It is possible that the Arabic /'adm had the
same colour denotation as its Hebrew parallel. This significa-
tion, however, was somehow dropped, while the derived sense
remained. By the same token, it is conceivable that a similar
development occurred in biblical Hebrew. If so, D'Wn signifies
'processed' (leather) with no specific colour reference apart
from the indirect connotations of the usual colours of such
33
skins. In other words, it is possible that dnKQ/D^anKa
TT; TT '.
-
preserve a root - perhaps also the ultimate origin of DTK -
which at some point became distinct through polysemy and
developed into a homonym of the parent sequence. Even so, the
exact meaning of the less frequent homonym was probably forgotten
already in biblical times, and it became synchronically equated
with its more popular parallel. Finally, the various shades of
DTK evoked in Nah. 2:4-5 ('scarlet'; the colours of fire and
of blood) leave no choice but to accept a colour significance
for D^TKto, although a deliberate word play on 'tanned' + 'made/
dyed DTK* cannot be ruled out entirely. The ancient Versions
understood the lexeme O^mKtt in Ex. as a colour term too. One
is forced to conclude that despite its conjectured etymological
history, QTKO carries some kind of specification of hue or
brightness that is inclusive to d*TK.
The question remains: Why should the skins be processed in
this particular (although obscure for us) fashion? Gradwohl's
answer is, so that they become waterproof. 34 Was this why the
skins were apparently dyed? Haran argues that they had to be
dyed in order to match the 0'BhH Hilly - "sea-cows" or
"dolphins' skins"35 - which are 'T naturally of that colour, and
which are mentioned together with the tPltfTKtt
-T"]- • Qt»N
i i » niiy in all
36
six instances in Ex. This is borne out by the Aramaic Targum
to 'KmDD »!PJr (= 'scarlet')37 and fits in with the general
colour scheme of the tabernacle plan, which,is organized around
four main colour themes: 'white' (K7J3 ,13)5 types of 'red'
ClttJTiX ,f^ nytnn); 'blue purple' (ftbSrO; and the colour
112 BIBLICAL COLOUR TERMS

appearance of TlflO DflT - 'pure gold', gold being the most

dominant of the metals required.
1.3 Y»N* - Zech. 6:3, 7
1.3.1 Because Q^-^ON. occurs only twice, within the same vision,
and as a modifier of the same word (0>01d), something should be
said about the secondary status it is assigned here. This, at
least in part, depends on the basic lexical value that is
attributed to the personal name VlttN (Isa. 1:1, 2:1, and more),
although this is irrelevant for the identification of the bearer
of the name. Unlike the personal names >5tt)K (i. Neh. 11:12;
ii. 1 Ch. 6:31), (1)n»»3K (2 Ki. 12:22, 14:1, etc.), and rP»3K
(Am. 7:10, 12, 14), all clearly derived from I /YON = 'be strong,
TO ,
solid, hard', there is a possibility that V1QN is related to
the colour term YON*, which is etymologically different from
I i/YQK. This possibility is suggested - with a greater or
lesser degree of certainty - by Hare'ubeni,39 Shalem,40 and
Maisler-Mazar,41 contra BOB and KB 3 .42 As the context of -
Zech. 6 suggests a colour reference for tPJWaK, and as there are
other personal names or place names with colour connotations in
the OT (13^ j*^;7 >tJ^N, to name only a few), there is no reason
to reject this explanation out of hand. If it is accepted, it
supports the classification of Y>3K* as a secondary term; other-
wise, the latter-should be considered a tertiary term. There is
no justification, however, for excluding YQH* itself from the
colour field.43
1.3.2 The horses of the fourth chariot are referred to as
O^iflaN tP?Tia in 6:3. Further on the CPS^N tPO-10 of v.2 are not
mentioned; instead, D*T12 are named as an independent colour
group (v.6), separate from O'tfttK (v.7). Therefore it is proposed
in BH to omit d^-Tia from v.3 of the MT, as if it were a doublet
of D'JjmN.
Fronzaroli, on the other hand, assigns to the horses of
both visions (1:1-17, 6:1-8) the following colour terms: D^Tia,
D»a*TK ,d'lhl2> ,d»Jl!? - i.e., he views 0>-SttK as superfluous in'
••-.-: •: -T; ---.44
6:3, and as a corruption ofti»)3TKin 6:7. Mitchell holds the
:
same view for ch. 6,45 but says' that
''• the same opinion does not
necessarily apply to ch. 1.46 Charles, whose analysis is quite
similar to that of Fronzaroli, raises the matter of the
exceptional nature ofti'JTljlin this passage: the context in
SECONDARY TERMS 113

which O^Tia occurs elsewhere (Gen. 31:10, 12 - D^^pJ O^py

0»Ti;H) points towards a reference to a 'spotted' o'r 'speckled'
appearance, while all the other horses in ch. 6 and in ch. 1, be
47
their colours what they may, are presented as unicoloured.
Charles argues that there is no need to reconcile the two
passages to each other so that they are in perfect agreement,
and that it is possible that the Greek Versions rendered D'^IW
(1:8) as if it read D^niS in order to harmonize the two visions,
48
for there is no basis for such a rendering in the MT.
Furthermore, the element of unicolour is preserved in the
49
vision in Rev. 6, which draws on Zech. 1 and 6. Therefore, I
think that perhaps it would be preferable to
a. maintain the MT of Zech. 1:8, discounting the evidence
of the Versions;
b. omit D'Tp, and not tPJMN, from 6:3; and
c. assume an'error which excluded CPflTK from 6:6, with
the subsequent attempt to restore the four-colour element by
using both D^SttK and its (mistaken) explanatory gloss D'nia.
1.3.3 Although the Aramaic T translates tPtfttN as 'ash-coloured'
50
CpJttOR), most of the commentators who agree that VON* is a
colour term that should not be omitted from this passage tend
to view it as some type of 0"TK = 'red'. Of the traditional
T
Jewish commentators, Ibn Janah51 and Qimhi52 discount the
evidence of the Targum and draw an analogy between V»K* and V'lttfl
(Isa. 63:1). Ibn Ezra tries to reconcile the non-colour
interpretation with the 'colour' one by commenting (for 6:3),
,fmna^ JIKIJI wn ITU Him ?">»* mnn o>nnK
but adds (for 6 : 7 ) ,
53
.rme»an n^Din^ i^nw artf D^miKn on D^inNMi
Unless his second comment is to be understood as a proposed
(veiled) emendation, the two are clearly contradictory.
Among the moderns, both KB54 and Klptin TlU>5 ^SIK55 support
the 'gleaming', 'piebald' (the former) or 'flashing' OTN = V-l^n
(the latter) by an Arabic cognate root. Shalem too lists V)3K*
' ^6 "*"
under the blanket term dllK. None of these three specifies
what the status of VON* vis-a-vis dTK is. Nevertheless, this
T
specification is quite necessary because of the.occurrence of
tiTK within the same context as signifier for the colour of
another group of horses (6:2). Guillaume attempts to determine
114 BIBLICAL COLOUR TERMS

the relationship between d*TK and V^K* by suggesting an Arabic

57
parallel ( ^—^ ' ) which refers to 'reddish-white' dust
58 ^ "
colour. I think that this definition of VON* ('light reddish-
brown') solves the problems of both sense and reference for our
lexeme, although no great degree of certainty is possible
59
because of its rarity.
1.4 pViy*T - Zech. 1:8 - CPJltM D>p1K> d">ttTK d^OlO
.. • T: ''.'. ' '-r:
1.4.1 i7"l(1)K> (Isa. 5:2; Jer. 2:21), Mpl'ty (Gen. 49:11), and
" T "

D'p.Tny* (Isa. 16:8) refer to a 'type of vine' - this is clear

from the word context in each case. The classification of this
60
grape vine - a vine whose fruit and wine are red - is largely
dependent upon the etymologies and the cognates proposed for
plb*. On the other hand, the existence of these same terms,
T
together with the river name P1KJ ^hJ (Jud. 16:4), the place
name nple>n (Gen. 36:36; 1 Ch. 1:47), and the dubious form
1
• . T ". s-i ,.
nip'ntp (Isa. 19:9) strengthens the position of P10.* (though a
hapax ) as a secondary term. Cognates to the root have colour
denotations to an area whose focus is 'red'; hence it makes
sense to view plf* T
as the base for the series P1W ••
,nplB/
-r"
etc.,63
not vice versa. In addition, the distribution of the above
mentioned lexemes across literary genres and chronological
periods from early poetry (Gen. 49:11) through pre-exilic and
exilic prophecy (Isa., Jer.) to post-exilic historiography (Ch.),
together with the knowledge of the process of development
/P1W = /P"10 went through in MH (see below , and in part III of
this study), consolidate our position even further: it is
probably but an accident that p'"IW* itself is extant in Zech. 1:8
only.
1.4.2 d^plKf modifies d'DID, hence the problem of determining
its colour denotation is linked to the question of the realism
or lack of realism attributed to Zechariah's vision. The other
factors that should be considered are the mutually exclusive
existence of d*S*TN in the same vision; the etymological relation
to pity ,np"uy - 'vine of a certain type of appearance'; and the
•• T"
evidence of the cognates. The ancient Versions (LXX and V)
sidestep the issue by translating into 'many coloured' or
'speckled', as if the MT read d'Ttt, thus trying to overcome
both the problem of the reference of P*ny* and harmonizing the
T
vision in ch. 1 to the one in ch. 6.64 " Such, interestingly
SECONDARY TERMS 115

66
enough, is the interpretation of Ibn Janah,65 Ibn Ezra and
67
Qimhi. The Aramaic Targum has "PftlP = 'faint-coloured' or
• Aft "^ * AQ
'grey', but this is not accepted even by Rashi. Modern
scholars who deal with the lexeme draw an analogy between it
and between Arabic \Jsrq - 'be red', 'brown', or 'yellow' (for
70 71
sunshine) and its metathesis t/sgr - 'light red', 'blond';
</ 72
Akk. sarqu - 'light-red blood'; the various Aramaic formations
73 74
from V/1J7D and KIp^D; and the MH iriD.v'rno. Although all
concur that p'lW* refers to some kind of 'red', there is no
agreement as to its approximate denotation. As Pines points
out, it cannot mean 'bright red' in our context - this would be
taken up by DTK (especially if no realistic properties are
attached to the vision) - and therefore the analogies to K1|7>D
75
= 'red paint' or to the Akkadian are not very helpful. On the
other hand, the Arabic ^-£----1 = 'reddish-brown colour of
horses' is much more suitable. Thus the definition presented
7fi 77
in BOB, 'sorrel', is preferable to that of KB, 'bright red'.
Alternately we can follow the etymological thread extended from
•70
Arabic (and Ethiopic ) - where /srg defines the colour of the
rising sun - to ittp^l IfiO. of MH, where apparently the
references of plO and D313 (= 'crocus, saffron') are perceived
as so close that they come to be employed as a hendiadys
79
denoting the action of painting in general. If we take these
data into consideration we should accept Ben Yehuda's definition
of 'yellow-red'. 80 Finally, as Pines states, the vines denoted
by t7*1W ,fiplBare not necessarily of a 'red' colour - this
identification stems from that of \>*VD*, not the other way
81 ^°
round. The fruit of the vines thus named can actually be
yellow or tawny, not red, even though they are called pIB.
This seems well within the framework of our terminology:
it would still come under the heading DTK in most strata of
OT nomenclature.
116 BIBLICAL COLOUR TERMS

C.2 SECONDARY TERMS WITHIN THE 1^ SECTOR

2.1 in**,
T
jud. 5:10 - n'-nfry; n'lJiiN ^m
(Ezek. 27:18 - in*
TT
Itt^l lia^h
;V
1" ,
Ihif - I. Gen. 23:8, 25:9. II.'Gen. 46:10, Ex. 6:15.
III.? 1 Ch. 4:7 - "Ihin Q, in*1* K).
2.1.1 1h¥*, although a hapax, is classified as a secondary
term because of the existence of the proper name Ihi? already in
pre-exilic times, and the place name ini( in Ezek. 27:18.
Moreover, even when we accept - contra the traditional Jewish
4
interpretation - that IDS and IIS^p are both toponyms, the
5
colour connotations of the two lexemes should not be ignored,
ll^h is phonetically associated with Sbh = 'milk', and through
this with the colour of milk. Ittif is conventionally used for
the description of colour entities which are defined as 13^,
"rT 6
hence the translation of T, NJX^J tt^ti 1ti)>l = 'fine, clean wool'.
"T ' """ * * "~ "*"• ™
There is no need to assume (with Gradwohl)' that the T rendered
IMS as if it read 1h2*, that is, that the translation is derived
T
• 7
exclusively from the meaning of *lhif* in Jud. 5:10. On the
contrary: 1h2r in Jud. 5:10 is rendered 1*11 *^ 'JPtt bH Ip^om.
~ , g
*TP2 ,1»'i1^^ is used in the T also to render dpi: in this case,
it seems, the translator was aware of some kind of colour notion,
but not of the approximate denotation of this notion. The
translation is therefore a paraphrase which is hardly relevant
to the reference of the lexeme 1h^* or the related IfliJ (Ezek.
27:18).
2.1.2 Most scholars explain Ihif* mainly from its cognates
in Arabic and Syriac as referring to various 'white reddish' and
'brownish' shades. Thus in the BOB the term is defined as
'reddish-grey', 'tawny',9 and similarly by Moore 10 and by
Burney- 11 KB has 'yellowish red'. 12 Consequently, Gradwohl
describes the reference as 'reddish-grey' and includes it in
13
the section of his work superseded by "rot". On the other
hand, there is a very strong Jewish tradition that explains
IhS* as "p!?. In the Tal. Bab. Erub. 54b there is a Midrash
in which our verse is utilized for praising both students of
the Torah and the Torah itself - D»tt3h n^ft 1>N il"nrtt »^S
o»Knyi2> mini? .min ra TiaW n^'Dtf n^-mai vyb vyn v^nnw
.d'nnjn nniK
SECONDARY TERMS 117

This Midrash may have been inspired by the alliterative

connection, or the similarity/identity in pronunciation (in
post-biblical times) of i/lh!* and i/lfiif; nevertheless, it may
also have preserved a genuine tradition of the meaning of 1h$*
which is close to that of v/irw, that is, refers to a quality of
light and brightness rather than to that of chromatic!ty.
- - 14
Similarly, we find the lexeme sohar = 'white' in Mandaic.
The denotation of 1H^* is even more explicit in Tal. Bab.
Ber. 31b. The text of most printed editions has the antonymous
pairs V.lVtl'lIN; trtN/W; Ylfm/TihS. The last pair probably
presents the contrast of dTK/1^^, 'white' complexioned vs. 'red'
1r T T T" -I ,-
complexioned. However, the Munich Ms. and the Aruch Completum
have here Tlh^ Kt>l ti^K K1?, a ihw/'p^ contrast which is previous-
ly stated as such (O^^ rn!m - d>Tin\U) in R. Aqiba's words
quoted on the same page. Later Hebrew tradition, as represented
17 18
by Qimhi, Rashi, and Ibn Janah, has Ipb both for Jud. 5:10
TT
and for Ezek. 27:18. On the surface it seems that all these
are influenced by the Aramaic Targum to Ezek. 27:18 (and by the
Talmud), which they apply to the obscure hapax in Jud. 5:10.
19
The same interpretation is adhered to by Ben Yehuda,
Kaufmann, 20 and Gesenius. 21 Pines, after listing the cognate
evidence, tries to harmonize the two approaches and suggests
that IfiV* refers to 'white1, or 'white' tending to 'pink'.22
2.1.3 The common denominator of all the etymological links
proposed for IhS* is the reference to a strong element of
brightness, together with a weak quality of hue. These
attributes are supported by the apparent closeness of /Ihif to
ynh'Sf and /1M^, both referring to the 'light', although not
necessarily to the chromaticity aspect. This also explains
how various forms of /shr cover shades that range from 'tawny1
to 'yellow' to 'reddish-grey' in Arabic; in Syriac - 'red' (of
'face); in Mandaic - 'white'; and in Hebrew - 'white', or rather
'pale-coloured'. What is common to all these derived meanings
is the notion of luminosity or brightness which is specialized
or narrowed in each of the above-mentioned languages according
to the needs of its own lexical structure. The meaning
attributed to /shr in the Ugaritic lexeme shrr(t) - 'burn/shine
(of the sun)'23*"- fits in with this reconstructed history of
the root reference as well. The immediate implications of this
118 BIBLICAL COLOUR TERMS

approach for our problem are:

a. The significance of etymological evidence must be
restricted - if possible - to the notion underlying all or
most cognates, that is, to their common denominator.
b. There is no conceivable reason to reject the Jewish
tradition which equates 1H^* and l^; 1H2* is therefore
understood as a designation for 'light-coloured' or 'bright1,
which is an internal Hebrew development from the basic Semitic
signification of the root. This development occurred quite
early. It is supported by the distribution of the derived
personal name lip? and strengthened by the fact that filing
appears within the ancient poetic text of the Song of Deborah.
On the other hand, the term fell out of use later; significantly,
it is chiefly preserved in OT literature in derived names
(IflS , "intf), while in MH it features mainly in texts which relate
to Jud. 5:10. This eventual elision - the reason for which
cannot be determined unless we attribute it to the growing
status of *p^ itself - is what caused the confusion as to the
reference of the term already in the ancient Versions.
2.2 ITS ,.in* and IT* ,-13T,
Song. 5:10 - 01TK1 HS •'TIT
24
Lam. 4:7 - a!?nn ins tfva rpina IDT

(Ps. 73:7 - TOJ'y n!?hO W - BH has an emendation to

. T •• T T" TT
ia:ny
•« -; X>na
TT-- nx*
— •• on the basis of Lam. 4:7, and in order to
restore some sense to the MT).
2.2.1 Gradwohl does not discuss (l)h^ and "DT because he
considers them members of the 'light/darkness1, not the 'colour',
25
field. Fronzaroli points out that the exclusion is unjustified,
for it clearly depends on the structure of the colour field in
Gradwohl's own language, which is not necessarily identical
"7 f\
with that of OT Hebrew. The word contexts of Song. 5:10,
Lam. 4:7 (and perhaps Ps. 73:7) - the mentioning of dlTK and
•IttlN:
; T the comparison
c based on the QIK/I^
T T Tcontrast: the
specification of the colour referred to by A'^Ujtt and 3!?hO, which
are used elsewhere for defining *pb itself - all these require
the inclusion of MS ,-inv and -IDT within the semantic sector of
"~ ~ 27
which 1!l!7 is the superordinate term. Fronzaroli goes on to
say that it is possible that hi? and 41T_* are specifications of
"p^; although this is impossible to prove on internal Hebrew
SECONDARY TERMS 119

evidence, it is to be considered probable from the evidence of

28
supporting terms in other languages. Theoretically this
might be the case. However, because of the rarity of their
occurrence and their association with A^W. (Isa. 1:18; Ps. 51:9)
and 3!?h (Gen. 49:12), nothing can be said of them beyond a
general statement about their apparent synonymity with "p!?
TT
within the context of Lam. 4:7 and Ps. 73:7 (if the emendation
is to be accepted). For Song. 5:10, dlTKl US, cf. B.I.6 above
(pp. 71-75 ).
2.2.2 The secondary status assigned here to the nominal HS
and the verb -1h^ is dependent upon the above mentioned compar-
ative structures (Lam. 4:7) and not on the number of occurrences.
29
As for ,1T* (^-(/"pf), it usually means 'pure', 'unadulterated1
- thus in Ex. 27:20, Lev. 24:2 (oil), Ex. 30:34, and Lev. 24:7
(1*1313^), all attributed to P or H. A further development of the
term, peculiar to wisdom literature, makes it signify a moral
30
state, whereby it refers to qualities of 'purity' and
'righteousness'31 - so the relevant instances in Prov. and in
Job, and the verbal lexemes of the same root which appear in
Job only. Our instance (Lam. 4:7) is the only one in which
the word context and the literary context dictate the under-
standing of .13T^ as a colour term of sorts (//-Ihi? and the
association with 'snow'). Here we are probably concerned with
a shift of meaning that can be understood only if both stages
in the development of the lexeme are taken into account, *p^
serves as a symbol for purity, both concrete ('white' clothes -
Qoh. 9:8) and moral-spiritual (Isa. 1:18; Ps. 51:9; Dan. 11:35,
12:10). Thus the use of Mt_* as a colour term equivalent to "p!?
makes sense both poetically and linguistically. Strictly
speaking, though, because -13^-within such a word context - is
a hapax, it is not a secondary term, and is discussed here
merely for reasons of convenience. Finally, Fronzaroli says
that HL* probably specifies 'transparence' (cf. the nominal
tl*OlDT), which is one of the aspects of light/low chroma
32
phenomena indicated by "|*l!7. This may be so, but cannot be
TT
proven for AJNSDa..
~H<- .-13t,
- where the crux of the reference seems to
be the exaggerated poetic reference to the colour of the
complexion of Jerusalemite youths, that is, the 'dazzling
brightness', not the 'transparence' of their countenance.
120 BIBLICAL COLOUR TERMS

2.2.3 The basic denotation of (/IDT - related lexemes is

'clean->pure'. The same area of reference is attributed to
33
the direct cognates in Akk., Aram., and Arabic. This indeed
is the meaning carried by most occurrences in the OT, be they
nominal or verbal. The unmistakable 'brilliant/white'
reference of Lam. 4:7, though isolated, exhibits a transference
through contiguity of senses; thus 'clean' -» 'unblemished' —?
'pure1 -» 'luminous' -> 'dazzling/white'. In other words, the
characteristic attribute referred to by -13T_ here is not chroma,
but brilliance. This is precisely the quality which makes 1^
'white' seem the purest of colours, a symbol of cleanliness and
purity on the concrete as well as on the spiritual level.
2.2.4 Similarly, a common denominator for all /hhlf related
lexemes - >ins (Lam. 4:7), hS (Isa. 18:4; Jer. 4:11; Song. 5:10
and, by transference, Isa. 32:4 - flltlS "Dl^); I"l»n2* (Ezek. 24:7,
8; 26:4, 14; Neh. 4:7 Q), nrpn* (Ps. 68:7), nrun:** (Isa. 58:11)-
T •; TT; —
is the reference to a quality of light, shine, or dazzle and,
34
by transference 'clarity1 (the principal attribute of daylight)
35
or 'heat/burn'. Pines, following the witness of various
cognates, consequently doubts the ' 13^' meaning of MS (Lam. 4:7,
TT — 25
Song. 5:10) and restricts the reference to 'bright, glowing'.
However, as we have shown, 13^ does not serve as an exclusive
TT
marker of total lack of chromaticity: it denotes any light
colour, for the principal quality of what we call 'ID^/white'
is maximum illumination. Therefore, ('l)njJ must be considered a
synonym of sorts to "p^, especially since 3^n may serve
elsewhere (Gen. 49:12) to emphasize the attribute of 'dazzle',
'brightness1, even though there it appears in a syntagm with
i:rt itself.
TT
SECONDARY TERMS 121

C.3 SECONDARY TERMS WITHIN THE iriB SECTOR

T
3.1 nrnrw*
Song.' 1:5 - ...YTp »!tfttO.. .niK:n "ON MllhW
, T ;
1:6 - eraem ^jnarera mrnnw »JKB >:nKin !?K
As tnrnhKf is morphologically secondary, we would expect it
to modify the semantic range of its superordinate lexeme somewhat
1 2
(cf. tnjpTK). This is the view adopted by Gradwohl and Pines
on the basis of the ancient Versions.
However, mine; in v. 5 seems to be semantically equivalent
to the nimfTO of the next verse. If we disregard the difference
in form, there is no demarcation line between the reference of
the two terms. The T translates mmriKJ by NITnTp and has the
same \/VTP sequence in its free rendering of v. 5 (YTpflK). As
the difference between the two lexemes here is that of class-
verb vs. nominal form - no semantic conclusion can be drawn
3
from it. The Syriac has KQ51K, elsewhere used for translating
A
</irny - lexemes as well as the Hebrew /YTp. Hence, no great
use can be made of the Aramaic Targums to our passage.
On the other hand, although "1(1) HW itself is not utilized
for denoting the colour of 'sunburnt skin1 elsewhere in the OT,
two of its derived forms are. Thus we have '!?yQ 1ht!> >Tiy
--r" — f
(Job 30:30) and dlKfl Tlhtftt ."I^M (Lam. 4:8), both post-exilic
occurrences. Although in these two instances the skin-tone
alluded to is the result of a calamity, this poses no difficulty
for the interpretation of Song. 1:5-6, where the speaker insists
that her skin-colour - the result of outdoor living - has
nothing to do with her beauty (B.3.3 above.)
At the most, then, we can say that m'lTltna was probably
born out of the stylistic necessity for variety, and its
derivation from 1ht2> does not assist us in establishing a
structural relationship to the parent lexeme in this particular
context. Neither is MH of much use here: not only is JTlfTlhiy a
hapax in the OT, it has no real existence in MH either. n'»'1:»1h^
and its Talmudic Aramaic equivalent NJP'vllhKJ seem to be derived
directly from IhW, while fllltVlhiy - which appears in the Midrash
5
Rabba to our verse (Song. 1:6) - wholly depends on our fllJVlhB.
3.2 Din - Gen. 30(J).
Gen. 30:32 - D>3U)33 d.1h hW tttl
w. 33, 35 - tpnum dim
122 BIBLICAL COLOUR TERMS

v. 40 - vonQIn6
3.2.1 Fronzaroli suggests that tMfl possibly filled the slot
referring to 'black, dark coloured1 in pre-exilic times, the
7
same slot that was later filled by 1HK> derivatives when,
eventually, DJlfl dropped out of usage. This tentative assumption
is supported neither by the biblical text - the use of Q.lh is
limited to our chapter, and the lexeme modifies only words
referring to 'sheep' - nor by the evidence of the ancient
Versions, which is quite mixed. The most that can be said about
it is that it indeed fell out of favour, to the extent that its
approximate reference ceased to be clear to later generations.
3.2.2 Gradwohl 8 and other modern scholars9 rely on etymology:
as 0-1 h seems to be derived from /Otth, the meaning attributed to
it is that of 'dark', 'black'10 (-» 'burnt', 'blackened').11
This etymological explanation, although it is dictated by the
text, is not convincing. At any rate, the text of Gen. 30
presents a conflict between light coloured animals and dark
coloured ones on the one hand, and between unicoloured animals
and multicoloured ones on the other. 'Dark coloured' could be
considered an equivalent of inw whose reference, as we have seen,
T
is at times not limited to 'black' proper.12 However, there is
no indication - even in ancient texts - that Ih^. and 01h have the
same semantic value, and that they are diachronic synonyms which
can theoretically be substituted for one another.
The LXX has three times phaios ('greyish'?) and once
poikilos ('multicoloured'), the latter obviously not of any
relevance to the present discussion. The usual rendering for
inu) in the LXX is melas. The V has twice furvus ('dark',
'lustreless black') and twice niger, whereas the usual rendering
oflfuyis niger. 13 It looks as if there is a certain confusion as
to the denotation of 0.1 h vis-a-vis IhWj at any rate, there are
attempts to differentiate between the two and to assign a
specific value to each of them. This explains the two Aramaic
translations: although T has DlhB ('dark coloured', 'brown',
15
or even 'black' ), in Pseudo-Jonathan we find U)lh^ = 'reddish'.
Rashi apparently relies on the latter when he makes the comment,
•pjy!> ru:tf nNsnJi mrmnju ruena iiw!? .ty^z urn Di*TKt> nan oih»y oin
16
.nKiinn
On the other hand, the Midrash Ha-gadol comment, K^131
SECONDARY TERMS 123

NflMIN,17 makes use of the lexemeti^JlN,which is usually

T: s ,T
utilized for rendering the Hebrew HflW.
3.2.3 In view of this self-contradictory evidence, I think it
is pertinent to define D-ltl not as the earlier equivalent or
forerunner of 1MW, but as a specification of it. That IhB does
not appear in texts of the Monarchic period might be attributed
either to accidental reasons or to the possibility that this
same slot had been occupied by another lexeme, perhaps one
lg
generated by either i/Yp or j/"|KJh. Alternately, d-lh may be
considered a dialectal term restricted to the sphere of sheep-
husbandry and the technical language spoken by shepherds. At
any rate, d.lh is a secondary term within the IMW sector only
from a panchronistic and not a synchronistic aspect.
As for the approximate reference of the term, merely
'black' will not do. If we suppose that O-in became obsolete and
was dropped in favour of IhW, the least that can be said is that
we cannot detect the motivation for this semantic process. Even
if we attribute it to a growing preference for generating colour
lexemes of the Pa°ol pattern, we should recognize that the
tendency is not all-inclusive even on the primary level (cf.
"py) • Finally, the range seen in the Versions runs from 'dark'
TT
to 'black' to 'red', while Modern Hebrew uses the very same
lexeme to denote the colour 'brown', which occupies an inter-
mediary position between 'black' and 'red1 colour phenomena.
Therefore Pines is perhaps right in cautiously suggesting that
biblical d-lh has a semantic value approximately parallel to that
of Modern Hebrew Q.lh. 19 Within the scope of the OT, however,
it is isolated, underdeveloped, and structurally a subordinate
within the sector governed by IhW.
124 BIBLICAL COLOUR TERMS

C.4 SECONDARY TERMS WITHIN THE plT SECTOR

4.1
Lev. 13:49 - tnniK IK (npT yMt\ iThl
...Tun IK ">twz IK iiys IK TA^
Lev. 14:37 - man imp} yAjn nsm
...mo'Ttt'TK IK mpipT jfinypiy
PS. 68:14 - CIDM nam nil* *a:i3
.ynn pipTa rpmiaKi
4.1.1 viin pipT
The parallelism of Ps. 68:14 - 103 // V11H pIpV - defines
the colour term as denoting a 'pale', 'silvery' colour - although
the compound itself is a hendiadys referring to 'gold' (according
to at least some scholars ), there is no doubt that this partly
2
duplicated formation of /pT may function as a diminutive form
denoting pale or greenish 'yellow'.
3 If so, plpv is a specifi-
cation of the brightness/relative lack of saturation attribute
which belongs under the term plV and, by its existence in a
psalm that is nowadays considered quite ancient, strengthens the
4
position of the latter as a primary term. This explanation is
5
supported both by extra-linguistic considerations and by MH.
np.O)1pT which describes the colour of Esther's complexion
6
(Meg. 13a) can only mean 'pale', while the productive force of
7
/pV in the colour field of MH and Jewish Aramaic points to a
solid status of PIT and its derivatives already in biblical
times.
4.1.2 P3P/1? and tlWTK. (Lev. 13:49) are symptoms of cloth or
an
leather Dy*1X> d the same lexemes represent nyl^ symptoms of
T r
building materials in Lev. 14:37.
Although PIT and DTK are not present in these Priestly
diagnostic texts, and so cannot be directly compared to their
derived forms, it would seem reasonable that both denote not a
change in hue vis-a-vis the parental base, but rather a differ-
ence in the 'gloss' or saturation/luminosity aspects. Thus, if
OTBTK refers to 'glaring, bright pink' - a section of the t3TK
sector - then P1PV would denote a 'glossy' or 'light' 'yellow/
green', a colour quality appropriate to that of mildew, mould or
fungii and inclusive to the P1T> sector. The latter term should,
at least theoretically, encompass all hue/brightness/saturation
attributes not covered by inty .a'fts ,tHK , 1H5and their immediate
8 T T T TT
derivatives.
SECONDARY TERMS 125

C.5 SECONDARY TERMS WITHIN THE SM SECTOR

-r
5.1 Ezr. 8:27 - 3nT3
TT mitth. . .iniW lUShJ ^D1
~ T ; •-.
5.1.1 The existence of a Hof. part, is reason enough to pre-
suppose knowledge of the primary term SMtf and an active (Hif.)
I T
sequence underlying the Hof. form. Even so, the hapax UMiM
would not be classified as a secondary term if it did not appear
2
in MH in the same form - and that despite the fact that other
Sfil? derived lexemes feature there too. However, the added
T ,
explanation - DfiTS) ftllfih - makes it clear that the term is still
not self-explanatory. The explanation was necessary, perhaps,
because at this stage SMSftt had just been introduced into the
lexis. According to this argument SMStt belongs under the
tertiary rather than the secondary category.
5.1.2 Apparently, the copper vessels described here are com-
pared to gold not only from the general aspect of their beauty,
but also from the more specific colour aspect as well. Thus
Rashi draws an analogy from our verse to Shi? in Lev. 13 and says,
3
ninr iaD mns snrn n'jnnt> non nns ne^i
^4 ^ TT
Similarly also Qinhi. While it is true that gold varies in
colour according to its relative purity - archaeological as well
as linguistic evidence of this abounds - the typical appearance
of gold has always been considered gleaming and yellowish in
colour. These two attributes are those denoted by Shifla, and I
can see no reason for translating "aus rotglanzendem Kupfer", as
5
Gradwohl does. Rather, it seems that from the exilic period
onwards 3M2 and its subordinates are understood as separate from
, T
tHN, although the range of the latter might have previously
T
covered at least some sectors of the former (the other parts
being previously included in 1^ and (71V) . 3^^? expresses the
aspect of 'be made into a yellowish, bright colour', and by
specification - a colour whose visual attributes are similar to
those of gold. Interestingly enough, 'gold1 is also referred to
by means of using a lexeme from another sector of the colour
field, cf. y.lin i^pV (Ps. 68:14). 6 This can be accounted for
either by the referential similarity of the visible attributes
connoted by either lexeme, or by the fact that each one belongs
to a different synchronic layer of the language.
126 BIBLICAL COLOUR TERMS

D. TERTIARY TERMS IN THE OT

D.O The DTK sector is the best defined within the colour field.
Therefore, not surprisingly, it is the most prolific as far as
tertiary terms are concerned. These include 'OIQ'TK. (3 times);
the syntagm mmK "]& (4 times); *^ton (once) and rtt^pn (once);
V'Dtfl (once, and perhaps one verb formation through an emendation
of ynnn, Ps. 68:24, to [QTO] ynhfl); and >i£) llttltth in Job 16:16
(+ Ps. 75:9?). Apart from sharing QTttTK 1^ with the DTK segment,
the only other iat> dominated lexeme is .llltl (Isa. 29:22), a
TT "T
single occurrence to which we might add the proposed ^lih^ .11111
3 , T ">T
(BH to Isa. 19:9). ^Bh (linwty) of Lam. 4:8 is perhaps a
tertiary subordinated to ihB), and the same might be said of
ftina in the syntagm rij^ flirt!) n'ina (Lev. 13:39) - excluding the
idiomatic usage of Qal formations of /nfO, usually + "py - where
/fill!) means 'dull' or 'matt'. No such terms are to be found in
the PIT* or unsf sectors - unless we delegate Hfl^tt to a tertiary
position - which is to be expected from the evolutionary point
of view. The referents of most terms are relatively clear,
either because of their etymological ancestry (':nttTK,tna*TK p!>)
or their word context (^bnn ,Vlfcn). Finally, only D^TK related
terms are found in all strata of the language, from pre-monarchial
poetry onwards - another proof for the supreme status of DTK in
T
the colour field.
TERTIARY TERMS 127

D.I TERTIARY TERMS WITHIN THE tHN SECTOR

1.1 •>}(1)»7N
Gen. 25:25 - IXIp1'1! 1)N>> IlYWD 1^3 'ilttlK IWin NiPl
.iwy im>
1 Sam. 16:12 - . >N1 S101 Q^J^y ?l£P 2dy ^1»7N K1H1
1 Sam. 17:42 -
3 "Witt nfl> dy »J*mKl 1)tt fpn >D...

1.1.1 1 Two preliminary problems related to the occurrences of

'31J3TK in the passages quoted are:
a. Does the lexeme describe the appearance of hair, as
seems to be indicated by Gen. 25:25, or of skin/complexion?
b. Whereas in Gen. 25:25 the term is neutral, it is used
as praise in Sam. An analogy between the d*TK (+ "p^) ideal of
T -rT
beauty in Lam. 4:7-8 and the Sam. verses immediately springs to
mind - how can we reconcile the instances to each other?
The ancient Hebrew Midrashim do not agree among themselves
' 4
as to the object whose name is modified by 'Jlft'TK. Of the
modern scholars, Skinner seems to prefer the 'skin/complexion'
5 6
approach; Gradwohl leaves the question open; and Gunkel is
7
quite straightforward: he talks about "tfautfarJbe". This last
view is justified by the text of Gen. 25. While lyw IWttO 1t>3
supplies the etiology to the name Vyw ,'OIQ'TK supplies the
allusion to d'lTK ,d'*TK ('pottage'), and Dn("T»S") . The story
is enhanced by assuming that two separate physical attributes,
not only one, are described. Further, this fits in with the
description of David in Sam. which, in turn, is strengthened by
the utilization of j'dlK for the same purpose in Lam., a text
that belongs to a period of the language in which 'Oltt'TK does
not feature. As for question b, one must assume that the term
is neutral, and thus can be used both in an ameliorative and in
a pejorative sense.
1.1.2 Pines says that 'OIKHK neither corresponds to nor is
contrasted with any other colour. Moreover, the evidence of the
ancient Versions here is rather dubious: the LXX is late, the V
Q

translates rufus, and the Aramaic translations use various

9
forms of /plO, especially those used for rendering d*ltt*TK.
These two points should be modified somewhat by the following
data.
First, there are two diachronic equivalents to the pre-exilic
128 BIBLICAL COLOUR TERMS

*31)3TK - the primary verb form '10TK of the exilic period

(Lam. 4:7) and perhaps the secondary term DTQTK (Lev. 13),
which defines skin-tone as well and which is considered either
post exilic or of the late monarchial era (depending on our view
as to the chronological context of P). The various translations
of the Aramaic Versions for Gen. 25:25 may have preserved the
notion that OTQTK was understood as at least a partial
substitute for 'OinTN. The chief Aramaic traditions for our
verse are -
10
T°: iyon i*tt pino
ll
Pseudo-Jonathan: *\ym "I^D n^lD "Ipnb
12
Syriac: myOT KflpTy KtOI "TpflO
13
Neophyti I: TJfl>n l!?D [3 rPt>] ID pbpttO
The Pseudo-Jonathan rendering - which is recommended as
14
an emendation to the text of the Neophyti I - appears again
215
as the rendering of the same Targum to OTttTK in Lev. 13:4
16
elsewhere translated KpQD (v. 19) or Npnio (14:37). The
isolated deviation to ^Tpino (for DTOTN) in Lev. 13 cannot
readily be explained, and this will not be attempted here. I
would just like to point out that DTQTK might have been
comprehended as a synonym, or substitute, for ^JlttTN - which
is perhaps the reason why no occurrences of ^JIQTK are recorded
after the exile. While this is no conclusive proof by itself,
the non-existence of the term in post-biblical Hebrew literature -
apart from passages which discuss the verses here dealt with -
is indicative of this trend. Finally, the form poptto (Neophyti
I) belongs, of course, to the same morphological pattern as
DTnTK; it is rather tempting to assume a dependence of the
Aramaic form (also in Samaritan Aramaic 17 ) on the Hebrew one.
However, the usual rendering of DT01K in the Neophyti I for
' 18
Lev. 13-14 is not papQO, but plBD or KpntD,.
1.1.3 As for the extra-linguistic reference of *31DTK, this
probably covers the "rofcbraun" 19 section superseded by DTK.
This interpretation is explained by Gunkel in the following
manner: "Mann amusiert sich ilber die rotbraune Hautfarbe der
Edomiter; die leute, die sich diese Sagen erzahlen, werden
selber gelblich ausgesehen haben vgl. die Farbe der Kanaanaer
auf den agyptischen Bildern"; therefore, he concludes, 'JlttTX
20 21
is "rotbraun, braunlich, bronzefarben"; so also Skinner,
TERTIARY TERMS 129

22 23
Caspar Levias, BDB, and others.
An alternative view is to understand ^in*TN as covering the
same sector as d*Tfl*TK itself, that is, 'glowing1 (see also •IB'TK,
24 1T
Lam. 4:7), 'red(dish)'. The second approach - still well
25
Within the area of dTK - is denoted by the Aramaic "IpinO.
In view of the scarcity of the lexeme I find it impossible to
decide which of the two is approximately correct. It seems,
nevertheless, that the 'reddish', or 'glowing red' reference
will suit the 1 Sam. occurrences better than the 'reddish-brow
one.
1.2 dlOTK "]&
Lev. 13:19 mywrK ?u:tf nina IN run!? mw
•:v:—'"• T T < •-'•••" TT; «•:
:24 rmt> IK ntt-miN Mtf fliro
:42 &:P?™. *pt> y^
:43 n»imtf run!? y>an ww
As stated above, DTOTK p^ should be differentiated from
D*rn*TK - the latter refers to the appearance of cloth and
building material 'diseases', but the former - to skin diseases.
In addition, Q*m*TK "\^ occurs in apposition to "p^, while d"f)3*TN
itself is in apposition to plpv.
Although treatments of tnmx tend to confuse it with
tnmK ID!? - no firm distinction is made between OTttlK as a
component of the syntagm and tTTttTK as a single term - I think
that the two should be dealt with separately, both on grounds
of structure and of reference. Otherwise, no sense can be made
of the apparent contradiction between the discussion in Neg. 1:2
76
and in Tal. Yer. Succ. 53d COconcerning the diminutive or
intensive force of D*rmK. The compound naturally refers to
both elements, lit? and dlttTK; hence they modify each other, and
the syntagm properly belongs under the superordinate "p!? as
much as under QTK. The combination of primary and secondary
terms refers to an entity which is distinct from that of each
component, but lies on the boundary between them. This seems
to be sufficient reason for categorizing dlttlX 1^ as an
independent unit. Morphological considerations, together with
the fact that the term is restricted to the description of skin
lesions, dictate its classification as a tertiary term. It must
be noted, however, that succeeding generations did not make this
distinction; and that Pines is perhaps correct in speculating
130 BIBLICAL COLOUR TERMS

that tnnTK later became invested with a diminutive force

27
through its identification with tTTOTN pX
1.3 »^3|3, wWan
Gen. *49:12 ' n^hn d^e> p^-1 1?2» tp^y ^ton
Prov. 23:29-30 1»2Q ^ O'lhKlrt tP3»y M^?G ^
The inclusion of »^»^!)h and fil^Ofl under the DTK heading
• • i— •: - -r
is not at all certain. Although historically it has been
28
understood as a 'red' of sorts, modern scholars tend to
29
define it as 'dark, dull', which is especially suitable for
Prov. 23:29 but also quite appropriate for Gen. 49:12. If the
second interpretation is adopted - as has been done here - then
the crux of the PV/*^*5OH juxtaposition is the brightness vs.
dullness contrast rather than a 'white'//'red' chromaticity
contrast. In this case, the inclusion of »!;»tOh here is
justified only by the general notion which dTK originally
represents, that is, a chromatic/saturated quality that is felt
to be different from IhB or pt>, but otherwise is only loosely
f TT
defined.
Another solution, also referring to the brightness dimension
rather than to chroma, is offered by Gunkel. His translation to
Gen. 49:12 reads "Seine Augen funkeln von Wein"; he continues by
emphasizing "sicher nicht Trube" and cites Groessmann, who
compares ^'^Dh to the similar /"5h5 (Ezek. 23:40), "wodurch sie
30
feurig und glanzend werden". A similar approach is taken up
by Caspar Levias, who defines '^!?Dh as a reference to the
'lustre of dark red wine' and IVlWUh as an 'inflammation of the
31
eye' (from drinking). This suggestion is, then, a compromise
containing both elements of brightness and hue. Further, the
possible 'sparkle, gleam' explanation is questioned by Toy,
although he too seems to favour the "dull red effect produced
by excessive drinking" for Prov. 23:29. 32
To summarize: the Versions and early Jewish commentators
understand ^'totl (and fl-l^tOM) as referring to DTK - type hues.
In this case it is interpreted as an idiomatic equivalent of
DTK, as when it is translated by /i^BO-derived lexemes; or, on
T
the other hand, as a partial synonym of the latter - as in
33
Syriac, where we have Kiah 1)3 Tn^Py 1A1T - his eyes are
'dark red'.34
On the other hand, modern scholars who do not uphold this
TERTIARY TERMS 131

explanation interpret ''^ton as either 'dull, dark1 ( < i/ttflD) or,

31
by contrast, 'sparkle'. C. Levias's interpretation brings the
two strands together. Finally, as far as the colour value is
concerned, the extra-linguistic entity referred to is 'dark',
only loosely controlled by d'*TK. Alternately it is the
saturation or brightness element which is emphasized, in which
case the discussion of ">'?'>'?3ft here is highly questionable. Even
35
so, it must still be considered a colour term, together with
such lexemes as ,1B?h and fiMD.
. -T
1.4 y.inn , v^YJtfi
isa. 63:1 - m^n dn^ vi»n DITKB >o m •>»
Ps. 68:24 - d"D T>n ynon IJfltf? - BH3: a proposed
emendation to 1113)3 dTKQ Vtt llKft din I^Vi ynnn lyn!?.36'37 The
hapax YtQft is defined by its context as referring to some visible
phenomenon linguistically classified under dTK. Etymologically
the lexeme belongs to the field of 'taste' or 'heat1, not that
of colour. 38 The Jewish Aramaic meaning of 'scarlet, red'
39
attributed to XJl*'m'»ri,KSntt1>n and recorded by Jastrow is
probably dependent upon our passage, and of little value for
the identification of Y-lQfV itself.
3
As for Ps. 68:24, the second proposal recorded in BH
dTa yrnn - is corroborated by some of the Versions (LXX, Syriac,
40 oN
Aramaic T) and by another biblical verse (Ps. 68:11). On the
other hand, the substitution of dWQ for d'D">Kti solves the problem
TT ;
of d'lPNn and Jifi-in, while employing a full imagery revolving
round a 'blood' + 'red1 picture. However, this reconstructed
picture, although a possibility, relies too heavily on Isa. 63
ff. for its sense and its meaning. I think that Gradwohl is
probably right in defining V-lttH as a 'violent, glaring red' =
'blood colour',41 for this is indicated by the context of the
passage.
i. 5 "iKnnn*
"~''"""''
Job 16:16 - ftt •'iio42 (.-ntt-uan : Q) rn&nan. »js
1.5.1 The word context of the verse points to an interpretation
of IIBinn as a colour term denoting some kind of 'redness' of
the face, the result of excessive crying. However, the etymolo-
gies suggested for this /IBM derived lexeme, as for other /Itth
formations, are far from satisfactory. Thus Gradwohl, after
considering biblical evidence together with the cognates,
132 BIBLICAL COLOUR TERMS

actually admits defeat.43 Biblical dictionaries present quite a

confused picture as to the differentiation and the organization
of various lah roots and the text occurrences that belong to
44
each. Because there is no agreement and because this line of
research - although interesting - proves to be fruitless for
our problem, another direction of questioning is to be pursued
here.
1.5.2 naian of Job 16:16 differs from the liaian in Lam. 1:20,
2:11 in its syntagmatic structure. The latter occurrences -
liaian >ya and ">yn naian respectively - seem to be an idiom,
one that is still vital in MH. 45 Therefore, the difference in
usage is much more important than that of genetic origin, and
perhaps much more illuminating. As it is, the various taar and
hmr roots preserved in OT language have become well nigh
indistinguishable through phonetic shifts and/or semantic
contiguity. This may have occurred through the common derivation
46
of both naian forms from a /"iah denoting 'boil, heat1, or
47
through a possible current allusion to inn (= ")>'), of which
both qualities of fermentation and colour can be considered a
typical attribute.
To come back to the evaluation of linian *y» vs. liaian >J£) :
even though Lam. 2:Ha - 'J^y flljwn^ l^D - is reminiscent of
Job. 16:16, it presents a stereotyped picture of grief that is
different from that of '•ya liaian. The latter is well defined
by the analogy in Lam. 1:20, - 'UlpS *3!> IflflJ (= 'my hearts
disturbed, agitated1) - although whether this liaiah is related
to /*ian = 'ferment' or /iafl = 'burn, parch' is again not very
48
clear. At any rate, the 'disturbed, agitated' explanation is
hardly adequate for the image conjured up by the words of Job.
16:16 - the 'red(denned)' element here alluded to is dictated by
common sense, if not by any other considerations. Thus the
49
rendering, "My face is flushed with weeping" looks like the
-, - 5 50
best solution.
TERTIARY TERMS 133

D.2 TERTIARY TERMS WITHIN THE ^ SECTOR

TT
2.1 Tin
-T
isa. 29:22 - iTin> •pis nny NVi spy* im* nny Nt>
(Isa. 19:9 - ^TJn d»A*iK1 miJ»"N!> d^JlBfl nsy lEDl. C£.
T
3 a
BH , and the text in IQIsa , where we find ITin for ">Tin. As
a result of the proposed emendation the text in both instances
will have the same parallel pair - BinX/Yin) . /Tin (in the OT
in Dan. 7:9) is the regular Aramaic equivalent of the Hebrew
V/155, and through Aramaic it probably infiltrated post-biblical
2
Hebrew as well. Further, the concept of 'paleness of face' as
indicating shame - which is the subject matter of both verses
cited above - is expressed in MH by an idiom containing a "p^-
derived form, d>:m ITjn *33 "p^Ofi to.3 Therefore, conclude
4 5
Melamed and Gradwohl, the verb form in biblical Hebrew is a
loan from Aramaic.
On the other hand, the occurrences of l^h (LXX: jbyssus) in
6
Esther 1:6, 8:15, although late and unique to this book; the
7
proper names 1.1h ,'^lh and IWfl, whose etymologies are obscure;
• ~ "r ' 8
the early ^1h of Gen. 40:16 - 'white bread and cake' or 'white
9 • . IQ
flour'; according to Qimhi, even tP*llh = 'nobles, freemen';
and finally, the existence of the root hwr, alongside Vbyd, as a
colour denoting base in classical Arabic, not only in Aramaic
j]^
and in Syriac - all these point to the possibility that /Tin
had originally been, alongside /"pV, a member of the basic stock
of Ancient Semitic, available for utilization in Hebrew as well
as in Aramaic; that it was later discarded - not without leaving
some related traces in the lexis - in favour of the more popular
^155; and that only later still it was reintroduced into Hebrew,
12
this time through the secondary influence of Aramaic.
134 BIBLICAL COLOUR TERMS

D.3 TERTIARY TERMS WITHIN THE IflW SECTOR

T
iu>h
Lam. 4:8 DIKtl YlhUW "HUH1T
: ' J -~
Bib. IhU) denotes 'black' but also the less specific 'dark1,
although not the 'darkness' of night but rather that of complexion
2
or hair. The absence of normal natural light, i.e., daylight,
is habitually referred to by M^flK ,^fl (Jo. 2:2), or i/YTJ7.
~r" -V
Hence, /l^fl belongs to the semantic field of 'darkness/light' -
within the structure of the vocabulary it lies on the boundary
of the colour field; but, although the two overlap in places,
representing related stages of cognitive and linguistic
development, they are basically distinct from each other,
3
limyn ,1H?h> however, has a unique flavour: its sense would best
be transmitted as 'darker than dark', 'blacker than black1. As
such, the poetic usage of "^Wn in our verse attributes more
colour intensity to ^K>h than to IThB. Therefore, within this
idiolectic and emotive framework, ^tyfl does function as a proper
colour term whose denotation is self-defined, in contradistinc-
tion to the occurrences of other lexemes derived from the same
base.
3.2 nna*
.... nns
T,. 4
Lev. 13:6 yMti HMD ...-/•/- "p!? (v.4)
T TT
v. 21 O nro
r
(mna, JINK;) -1+ n&rm-TK
'.".— ••;
nj^...iK ?mt>
v;:--: T T: TT;
(v.is).
v. 26 .nnD (mro) -//- run!? IK nana-ix nant? (v.24).
v. 28 .HMD (as in previous instance).
v. 39 ,Ktt
^*
nina*' (tihna)//njn^
TV T*
cn'inaTVrhna)
X *.*
(v.38).
5
v. 56 nro (y>^n) -MDTO'TK IK jripv (v.49),
1J1*)) nKoo.lOn K^7 (v.55) .
3.2.1 To begin with, one must classify the difference between
the above-listed occurrences and between the other occurrences of
/HMD-derived forms in biblical language. The other occurrences
can be roughly divided into verb forms and nominals (functioning
as noun modifiers). The verb category includes both Qal and Pi.
formations. The Qal formations - barring one instance,
Isa. 42:4 (cf. below) - always appear in a tight syntagmatic
construction (noun phrase + verb phrase, or subject and
predicate) with "py or D^J^V, thus -
Gen. 27:1 ^OK'ia "PJ'y •P'*'??1?'1 (Isaac).
Dt. 34:7 Oin!> 03 Ktn i:py nnns K!? (Moses).
,. T-.T
TERTIARY TERMS 135

Zech. 11:17 .firisn nna 1JW Vyi (the shepherd)

"' ' ~*
Job 17:7 -t^-jy BjDa ttttp6
Hence, we have an idiom which features in various types of
OT literature, and of different chronological ages. The message
is clear - the idiom describes a state of lack of sight, or
dimness of sight, due to old age or to excessive crying (for
grief). It is similar to other expressions employing a verb
form + T>y (tnipy) and signifying the same idea, viz. D>:py -IDKin
(Ps. 69:24, Lam. 5:17; cf. section 3.1 above), nttp "\">^y (1 Sam.
4:15), and ">^y oyon rmy (Ps. 6:8). Therefore, although it
seems etymologically identical to the /rtfl5 - derived term in
Lev. 13, its actual usage within the language marks ^fifO Qal as
an only partly transparent component of an idiom meaning 'x
cannot see'. As such it lies outside the sphere of the present
investigation. The same argument excludes the nominal Mfl3* in
the syntagm nins l!?nn "P^yi, 1 Sam. 3:2 (cf. 4:15).
The last Qal form,7 yW K^l n?O» K!? (Isa. 42:4, together
with the other occurrence in the same chapter - fUID* Nt? MfD tlJlWD,
v. 3), albeit late, probably represents the basic sense from
which, by internal polysemic shifts, both usages of 'become
unseeing1 and 'dark or matt' (Lev. 13) were derived.
Although these two instances supply an important link, they
are not considered members of any stratum of the colour field,
for they clearly refer to 'dim1 or 'dimming'. Finally, the Pi.
forms (1 Sam. 3:13, Ezek. 21:12) are so far removed in sense and
reference - to 'rebuke' or 'lessen1 - from the Qal forms as to
o
warrant a separate genealogy, although Ezek. 21:12 may express
9
a metaphorical development of the basic sense. As for the
nominal fliTj>(fVn, Isa. 61:3), this belongs with the verb form of
Ezek. 21:12. The nrp of Nah. 3:19, though considered a nominal
exhibiting the same basic sense of 'dimming, lessening —^
10 11
relief, is better amended to MMA = 'healing', and is
rightly treated as an independent entry whose connection with
the colour term nro, if any, is opaque.
3.2.2 On the other hand, the i/HMD forms of Lev. 13 are
defined by their general context and their oppositions as a
colour term, and this should be discussed independently of the
12
other forms reviewed above.
The clue to understanding the reference of the term is to
136 BIBLICAL COLOUR TERMS

be found outside the OT. M. Neg. 2 deals with the difficulty of

perceiving a JlfU) fnn% against a dark skin or when examined under
unfavourable conditions of light (Neg. 2:1 ff.). In the same
source, the antithetical term defining flili) by contrast is nina
fiTy. Applied to the 'appearance of skin lesions' described in
Lev. 13, it means that MfD denotes 'darkening', or perhaps 'lack
of gloss' - although the last interpretation is questionable in
the case of the {Tito, (v. 39) alluded to by tt>^ fi'injj jfina and
Jmt?
T: flina
••ifina
TV (vv.
*• 38,' 39).
' If so,' then nJT3
T" refers rspecifically
to the dimension of intensity, or (lack of) luminosity, not to
any chromatic quality. Thus it defines only one attribute of the
phenomena which creates a sensation of colour. Hence, its
inclusion under 1Hiy is only partly justified. Alternately, it
should be classified - together with the D*y of MH - under a
T~
separate heading presenting terms which refer to secondary
13
attributes.
MAN-MADE COLOURS 137

E. MM MADE 'COLOURS': PAINTS, DYES, PIGMENTS, DYED MATERIALS,

MAKE-UP, WRITINGMATERIALS

0. CLASSIFICATION AND GENERAL REMARKS

0.1 Gradwohl divides the section into four main parts:
I. Terms for colouring.
II. Animal-derived dyes.
III. Vegetable dyes.
1
IV. Mineral dyes.

The guiding principle of this arrangement - the origin of

the pigment in the 'real' (= extra-linguistic) world - is valid
as such, but not linguistically meaningful. There are no
linguistic indications - morphological or lexical - that this
is the principle of organization according to which the speakers
of a given language classify the lexical signs referring to the
relevant colour experience. On the level of daily life and
therefore, on that of linguistic collocation, the sphere of
application is probably more important than that of derivation,
usage more definitive than etymology.
Further, this way of organization does not explain certain
features of the linguistic structure, such as the dual function of
most names for textiles. These usually designate both a colour
property and the type of material dyed by the specific agent
(and see below).
Our classification is organized under the following headings:
1) Colour and paint, colouring and painting.
2) Textiles (2.1 Crimson, scarlet - 'JB ,^)yo ,y^1H ,d'^JlJ3,
"OB ny!nn; 2.2 Purple-pink to very dark - imiK ,n^:m;
2.3 "p^- KM ,V12 ,03*13 ,Tlh; 2.4 - Multi-coloured textiles).
3) Cosmetics (>19 ,/5fi5 ,/1J7W?) .
4) Dyeing Agents, pigments (4.1 dTN ,$nx - IBM ,nN-19;
T T -T T
133? 031.3?).
5) Writing'Materials (5.1 "IhB - m? ;5.2 IKM ,Y1ty) .
T" ~" T ',''.'
6) Whitewash (TW ,1>) „
Hypothetically an informant, unless specialized in textile
dyeing or pigment making, would classify the terms enumerated
above according to the visual attributes designated by them
within the double-stranded frame of reference, that of materia
and that of visible colour. A priori it would seem that in a
non-specialized context the colour would be recognized as a more
138 BIBLICAL COLOUR TERMS

important element than the type of cloth. This is reflected in

the language. Thus 10A1N ,">%%) , ^tilD are the names of specific
hues within the CHN sector (with f1^3fl bordering on it), and can
function both as names for pigments and as proper colour
notations (especially in poetic contexts - Isa. 1:18; Song. 4:3,
7:6; Nah. 2:4), as well as designations for the actual fabrics -
most commonly wool - that are dyed with the specific pigment.
This too would support our categorization contra Gradwohl's. On
the other hand, fabrics which are known to us to have been
'white' - bleached, but not dyed - require no specific chromatic
definition. Linguistically they are not represented by names
for colour or for colouring and do not function in comparative
structures or in metaphors of the type cited above. Their
colour significance, although clear, has to be discovered
etymologically (2.3 below), because it is only one of the
typical qualities designated by them. Strictly speaking, such
lexemes should be placed under Indirect Colour Allusions; they
are discussed here merely in order to present a more complete
picture of the textile field.
The situation concerning materials for make-up is different -
here the actual colour signification is all-important. Theoreti-
cally the referents should be clear enough. Unfortunately, the
number of occurrences does not warrant a straight-forward
identification (3 below). The few names for pigments that we
know, mainly from extra-biblical sources, 'are not colour names
(4 below). These terms are often simply the lexemes referring
to the source from which the dyeing agent is extracted (mostly
from plants, in our case). They are different from group 2
of our list in that they do not lend their names to certain kinds
of fabric. Materially they are dissimilar to IftATN ,'JW, and
TT-." ' T
D^Jl also from another aspect: they were not used on their own
but were mixed, in varying proportions, to achieve different
chromatic results. (When variation is required for the
manufacture of items denoted by group 2 terms, methods of
dilution and processing are resorted to, rather than mixing).
Thus, chemical analysis of dyed wool fabrics Yadin discovered in
a Judean Desert cave revealed that only four colouring agents -
indigo, alizarin (HK13), carminic acid, and saffron, together
with alum-iron mordants - were utilized for the production of a
MAN-MADE COLOURS 139

3
total of 33 (!) hues. There is no reason to believe that the
methods used by the dyeing industry during the late Mishnaic
period were not similar, even if more modern, to those practised
hundreds of years earlier. On the other hand, this makes the
'hue' classification of the terms designating the dyes in
accordance with Primary Term headings impossible, which forms
another distinction between the members of groups 2 and 4.
Perhaps this is the reason why group 4 terms are never used to
convey a colour notion through a metaphor or simile. Writing
materials: "PT is neutral, and extra-biblical sources are to
be appealed to for discovering its colour denotations, for no
connotations of the lexeme itself point in this direction.
Finally, whitewashing: again, no direct allusion can be
found in biblical literature for an equation *p^ - V> ,*T'W or
the practice of using T> and T>U) for achieving white looking
surfaces. The lexemes, their meaning, and their status have to
be explained by using the occurrences in MH, which is invaluable
for the investigation of the whole field.
140 BIBLICAL COLOUR TERMS

E.I COLOUR AND PAINT, COLOURING AND PAINTING - GENERAL TERMS

1.1 Archaeological evidence as well as literary sources make it

clear that not only were coloured textiles (and ceramics) known
to the inhabitants of Palestine, but that they were also locally
1
produced. The detailed information transmitted in post-biblical
sources about the arts of painting and dyeing seems to be the
2
result of centuries-old practice, and Yadin's excavations
3
corroborate this, albeit for the second century A.D. Apart
from this extra-biblical evidence, the bible itself supplies us
with some data, although indirectly.
Even the passages dealing with the making of the Tabernacle
4
and its contents by Bezal'el, Oholiab, and their colleagues do
not credit the ancient craftsmen with the knowledge of preparing
purple dye and applying it to cloth. On the other hand, cheaper
pigments were probably used by local industry for pre-monarchial
days onwards. The transmitted naming of two Issachar descendants
(Gen. 46:13; 1 Ch. 7:1,2), fathers of tribal branches (Num. 26:
23 ), after the two pigments sources yblfl ('kermes') and nx-19
('madder') - the former an animal extraction, the latter a
vegetable dye - probably indicates the existence of dyers' guilds
6
in ancient Israel. Undoubtedly it is no coincidence that these
individuals and the minor Judge7 TTlT "p nK53 13 yt>i fl (Jud. 10:
1,2) belong to a northern tribe whose connections with the
89
Phoenicians may have been particularly strong. ' A later
tradition (1 Ch. 4:21) tells of a byssus workers' guild within
the tribe of Judah. While the lexeme V'13 itself is late - the
10
post-exilic equivalent of earlier W2) - the tradition can be
accepted as valid at least for the late biblical period. As
Herszberg points out, the art of making fine white linen,
although involving no usage of pigments or dyes, is quite
11
complicated, and therefore is included under the present
heading.
1.2 This is as far as our extra-linguistic evidence will go.
The linguistic evidence is even more sparse. There is no
general term for 'colour' - the ySX* of Jud. 5:30 (D»)n* t^B) is
probably a homonym of the MH yys - 'dye', 'pigment' -^ 'colour'
12 ""'*'
and its derivatives. Etymologically, tify^ ,ym ,&»^j}. n,
and perhaps lljtttf of biblical Hebrew are distinct from the \/ym =
MAN-MADE COLOURS 141

'dip', 'wet' - of biblical Aramaic 3 (Dan. 4:12, 20, 22,

5:21, Pa. and Hithp.), the latter being the source of the new
14
term which fills the obvious gap in MH. No term for the
activity of 'painting' or of 'drawing' has come down to us
15
either. The conjectured /dttO, allegedly to be found in
2 Ki. 9:30 (fPJ'y .>133 dE|?n), Job 13:27 O^Al T05 dWfll), and
Job 33:11 (*raa dUP)16 is far from satisfactory. The MT is
better retained in Job (^>1//'»ilini1K ;TO = 'stocks'), while
the usage of the general verb /ti'ty is, for lack of any
specialized term, perfectly in order for 2 Ki. 9:30. Further,
d'JOttD/dO ,d>»0 in biblical Hebrew or in MH refer to 'drugs'
and 'chemical agents', loosely defined and/or narrowed down by
each of their contexts, not necessarily to 'dyes/pigments' and
17
certainly not to 'colour' alone. Hence, the existence of the
1R
term in this particular slot in OT Hebrew is unsubstantiated.
n^n (Jer. 22:14), although syntagmatically connected to 1BW,
presents a hapax in such a context and nothing can be deduced
from it for OT language. Similarly, the semantic gap might be
the reason for enlisting a lexeme from a close, partly parallel
field to serve where no specialized term exists in Gen. 49:lib -
nniO d'njy dTSI Itzna!? Vl W!3. The verb form t>±) undoubtedly
signifies not 'washing' or 'cleansing' but 'dyeing'. 19 The
usage makes sense not only because washing is one of the steps
of the cleaning/dyeing process, but also because /D3D, which is
related to /KDD, perhaps still retains the allusion to 'tread,
press'. 'Tread'is the basic signification from which both
v/033 and /unD developed in different directions: 'wash'; and
'oppress', 'subdue'. 20 Similarly, we remember, the superordinate
lexeme for 'dyeing' in MH is derived from \/ym, originally (as in
Dan.) meaning 'dip, wet'. In both cases the area of reference is
21
expanded and then narrowed. /DID, however, remained in the
field of 'washing and cleaning'.
1.3 The absence of specific superordinate terms for either
'colour1 or 'pigment/dye' of any kind dictates utilization of
other general terms when these topical notions have to be
communicated. The latter are taken from the field of visual
perception (of which the colour field is a natural component).
These are22 1) Vy , pya - thus n^an -pys IJ'JH (Num. 11:7);
most occurrences of names of metals and precious stones in Ezek.
142 BIBLICAL COLOUR TERMS

(ch. 1, 8, 10); one in Dan. (10:6); the colour appearance of

skin lesions in Lev. 13; and perhaps also htf* (BH ; from Lam.
4:7; MT - W) lJM>y 3^n» = 'their appearance1 (='complexion')
is more dazzling than milk, Ps. 73:7. From the point of view
of distribution this lexical development of 1>J> is almost unique
to Ezekiel, P, and a few post-exilic sources. 2) MK"in, of the
formula "pya.. .rwntt, sometimes serves in specific 'colour'
contexts - n^wm'"pyS) 1MT flK^tti) (Ezek. 8:2), JlE|hi Npttp *"*?}»
(Ezek. 40:3), riylif ilN'inS>...lltt'm*m rmt> (Lev. 13:43). The
*•*'"' 23
distribution here is exclusive to Ezekiel and P.
In both cases, the material is too localized to support
polysemic development within the colour field of either term.
At the most, they serve as substitutes where there is a need to
fill a semantic void or to side-step it. On the other hand, MH
possesses quite an array of 'blanket terms', well differentiated
into 'colour'/'dye'/'pigment1 categories, either borrowed from
other fields or original to ours.24
MAN-MADE COLOURS 143

E.2 TEXTILES

2.1 SCARLET, CRIMSON (types of d'tK):

^KTD .tpy^nn ,ytnn ,»JB> ,nyt7in »:n>> ^JB ny^in
2.1.1 Occurrences and distribution
"> 3B ny^lft = Ex., Num., within the description of the
materials for the Tabernacle.
nyt>lfl ">W - Lev. 14:4, 6, 49, 51f.; Num. 19:6. The
context is that of a ritual cleansing/expiation
ceremony in which cedar wood, hyssop, and
two birds are used as well.
>iB - Gen. 38:28, 30; 2 Sam. 1:24; Jer. 4:30;
josh. 2:18, 21 (>a»n mpfi; uwn Din niflfi);
Song. 4:3 (VnHlflKJ ">wn OHn*)); Conjectured
>T
~" ' 1
from D*:>K;
•T
- Isa. 1:18. Prov. 31:21.
ytnn - Isa. 1:18 (y!ntt3 l»nN 1 >); Lam. 4:5 (+ n.pr.m.,
Jud. 10:1, Gen. 46:13, Num. 26:23, 1 Ch. 7:1,2;
and »ytnnn, Num. 26:23).
ory>>rin - Nah. 2:4
^)J13 - 2 Ch. 2:6, 13; 3:14. Conjectured - Song. 7:6
(MT %na//1tt>1K)
2
2.1.2 The kermes worm (Coccus ilicis L.f the shield louse) lends
its name, or part thereof, both to the colour appearance 'crimson1
(Song. 4:3, Isa. 1:18) and to the material/cloth of the same
colour - HV nytnn (Ex., Num.), t^ttlD (2 Ch.), and >3W (Gen.,
2 Sam., Jud., - 'crimson thread').
The abundance of designations referring to the same items,
as can be gleaned from the list above, may be explained as
belonging to sources that are diachronically separate and/or
stylistically distinct.
The 'coloured material' is designated in P by the tautology
3
^3W fly^lfl (whereas the same material, elsewhere termed >3B> Olfl,
when used for ritualistic purposes is arranged in the reverse
4
order - ny!?in ^3KJ ). The other sources, narrative and prophetic
alike, usually have >:iK> only. In Isa. 1:18 the occurrence of
(•WIN1*) 3ft1)rO is probably dictated by the need to parallel
(D)>:iBO, at the beginning of the verse. The hapax tpystfltt (Nah.
2:4), paralleling d^lKQ of the first stich, may be the result of
the influence of the form of the latter. Lam. 4:5, on the other
144 BIBLICAL COLOUR TERMS

hand, contains a straightforward substitution of J^ltl 'for >JB. In

2 Sam. 1:24 we read Q»3*ry ay 1*V tOp3>B}lJ whew we compare this
phrase to Lam. 4:5a, D>3TJ»5
T t)*!OKil, and especially if we accept
"-* j
one of the emendations forwarded in BH - to &*JTy Ofl£ - then
Lam. 4:5 seems to be a literary borrowing from 2 Sam. 1:24. Seen
in this light, the substitution of Jftlft fotr »J» (v. 5b) is
meaningful; moreover, although MH uses njainly the Aramaic term
11*11MT, the archaizing Temple Scroll from Qumran has y^llt, not
6
*JU>. The former is also attested as a personal name from the
period of the Judges. Finally b»»*tt, referring to the material as
7
well as to the colour (Song. 7:6), is a rare post-exilic synonym
which - in the last analysis - was not accepted: in MH the Aramaic
tl'TlMT features largely, while ^n*O seems to have been dropped.
2.1.3 Gradwohl derives >3U> both from Ugaritic tn and Arabic
8 ' "*"
sana' - 'gloss, light'; according to him, the passage in Gen.
38 (where a crimson thread is tied onto Zerah's hand) and the
g
Aramaic IPIIflT strengthen this interpretation. However, Blau
comments that one should choose either the Arabic or the Ugaritic
10
cognate as a base; and Landsberger proves - I think
satisfactorily - that tt*17MT is derived from a culture word
whose origin is opaque, and that Gradwohl's attempt to define
both ">M1 and JI'llhT as originally designations of 'gloss1,
T
'light', is - to quote - 'eine wilde Etymologic'.11 He
suggests a connection between *JK> and Akk. sanu, sinltu -
tolourful',12 'dyed' (textile). 13~ In Hebrew, then, the wider
sense has been narrowed to denote a 'dyed textile' par excellence,
second in value only to 113A1K1 fl^f) ('purple'), while the
original range was retained in the source language. The borrowed
term, with its narrowed sense, then functions as the equivalent
of yt>1fi, a native term referring to 'worms' in general and, by
14
specialization, to the kermes worm. It is clearly understood
in narrative and prophetic sources that the two distinct terms
relate to the same extra-linguistic entity: the tautological
usage of P is secondary, perhaps the result of an overloading of
y^in ,nyt7lft and a subsequent requirement for a modifier when the
term is employed as a denotation for coloured/dyed textile of
a certain type.
2.1.4 The actual reference to a 'scarlet, crimson' appearance is
included within the scope of biblical QTK. Thus both yt>in and
MAN-MADE COLOURS 145

>3BJ serve for conveying the sin/purity, blood (of sacrifice)/

cleanliness, l^/OTK
Tr T
contrasts of Isa. 1:18. As for Song. 7:6,.
5
if the mention of (^3"1|> =) ^tflS and l^tt IttAlK has nothing to
do with a colour reference but is a roundabout allusion to
expensive, rare stuff, then there is no problem. Otherwise
some emotive meaning must be assumed - perhaps Feliks is right
16
in referring to the weaving process, although henna-dyed hair
seems to be out of the question here. Finally, in Song. 4:3
(VniflflB *3Bii OlhD), 'JW again functions as an emotive equivalent
of the superordinate DTK, while in *|H|71 llttlM n^£O the pomegranate
has no colour connotation, although it was used for dyeing in
i 17 10
post-biblical times (cf. MH). The beloved's cheek is compared
19
to a rounded pomegranate slice for fullness and symmetry.
2.2 PURPLE: IttAlK ,H^n
TTi~
2.2.1 in>1K - Occurrences and distribution.
1»A1K occurs in the OT 39 times, together with one 11A1K
(2 Ch. 2:6) and 3 times, KJU1K (Dan. 5:7, 16, 29). In most
TT ;
passages it appears as one component of a sequence naming rich
and expensive materials, two or more in number. Some of the
syntagms are recurrent, as if they were regular formulae. Thus,
a. inWO n^Dtl - Jer. 10:9, Ezek. (5 times), and mainly
within the P sections dealing with the Tabernacle. 20 In many of
these verses we have the quartet ITUttJ W)"\ '3W nytnm imiKI nton
T T
21 ' "
(Ex. 25:4 and more), the supposed colour symbolism of which
22
later generations repeatedly pondered.
b. im-lNl WU - Prov. 31:22, 23 or
innNl V-13 - Esth. 1:6; 8:15.
c. (Vim) najni inyiN - Ezek. 27:16.
T ' • '
d. t>^Q1D^l Yl^l rtfona in^K2 - 2 Ch. 2:13 (and cf.
2 Ch. 3:14 for the same list, but in different order).
e. In parallelism: 1ttA1K//^MT - Song. 3:10.
•|)3nN//^fl"-O - Song. 7:6.
f. inriK as noun modifier: imiK "m - Num. 4:13;
Jud. 8:26; 1»>1K. „ .''bin - Esth. 1:6; 1»nN.. .VDn - Esth. 8:15.
In MH this practice of using in>1K as a noun modifier is carried
further: there we have Din .ITmBft ,mn^W3 ,VDJt of IKJA'IK (and
of rrnihT, 'crimson') .
2 . 2 . 2 Jl!on - Occurrences and distribution.
a. As the first component in the syntagm imiKl n!?3n - cf.
146 BIBLICAL COLOUR TERMS

2.2.1, a. supra.
b. 1.1 HI n*M1 - Esth. 8:15; ntttJll (DfllD) 11 h - Esth. 1:6;
cf. 2.2.1, b. above for the parallel 1BA1K1 VIS. 24
c. finpll n^DH (>ni5>:i) - Ezek. 27:24; cf. v. 7 -
imiKi n^n...nnpin m
d. n!>Dt1 as modifier for: *m - Num. 4:7 (and v. 12); in
the previous verse - ntOfl t>^3 1>S, as in Ex. 28:31 - itfon ^3,
pure fltOH; ^H£> - Ex. 28:37; flitf!^ - 26:4; 1»>1K1 JltOfl) 'OBI
C^JK> ny!nni 26:28, 33.
2.2.3 In fact IttAlK and fl^fi appear more often as a pair than
as separate terms. They also interchange as signifiers for royal
attire, that is, as a symbol of power and government. Thus
Jud. 8:26, 1*1)3 O^n 'jyv IBAIKil »1Xil, is analogous to Esth. 8:15
- (1-ini) ntOfl niStttl W1^!?1. H!7Dn is usually designated in
English by 'blue purple'. However, it must be borne in mind
that extra-linguistically darker 'purple' wools differed from
lighter ones (1QA1K) more in saturation than in hue; further,
the terms n^tfl and 1&A1K are generic, not specific. Although
25
cognates for Jl^On abound, tl^Jfi itself is a rather vague colour
marker. Parallel terms, for example the Ugaritic phm ('dark1,
almost 'black') and the Akk. -iqni, uqnatu ('lapiz-lazuli' colour),
conveye the colour diversity referred to by tl!?Dfl much better
J£\
than our term. Landsberger supplies a synoptic list of
27
lJ3A1K/tl^H terms in the Near East and in the early Versions:
the various terms probably reflect the local popularity of one or
the other 'royal purple' shades, all of them the produce of
28
different articulations of the same basic dyes.
One last point: n^3fi, when in conjunction with IttAIN,
29
always appears as the first member of the pair. It is usually
30
thought that 10>1K was the more expensive of the two. However,
one should ask whether the rigid word order does not reflect
(subjective) relative importance attributed to the product cited
31
first, at least for the user of that idiom. This importance
can be the result of price, or - which cannot be ascertained -
connected to a symbolical value attached to n!on shades that
were considered typical (sea? sky, with a taboo attached, at
least in OT times? and cf. Tal. Bab. Sot. 17a and Hull. 89a,
among others).
MAN-MADE COLOURS 147

2.2.4 IflAIN is usually translated as 'red purple'. Plinius

states that the typical, most expensive shade was a dark rich
purple, the colour of coagulated blood,32 quite similar to
crimson. Hence, the less expensive kermes was used for
imitating this kind of 'purple'. 33 Other IttAlK shades ran from
'pink' to other types of 'red', bordering on 'violet'/'dark
purple' (= Ilton). "Only in the last 150 years has purple meant
34
a definite shade with two permissible variations". Again,
the lexeme IttAlK itself signifies mainly wool dyed in a certain
way (by a pigment derived chiefly from Murex brandaris and Murex
35
trunculus*). Insofar as it carries a colour reference, this
depends on the most popular/typical shades current, in our case,
among OT people. This wide scope, although still under &TK,
is perhaps one of the reasons why IttAIN is relatively unsuitable
36
for metaphoric - emotive expression.
2.2.5 Of the actual extracting of 1»A1K and nWn dyes, their
processing, and the extended monopoly of the Phoenicians much
37
has been written, and the process has been defined as well as
can be expected. Etymologically, the attempt to derive 1tt>1K
38
from Sanskrit rag = 'colour', 'red' (Benari, before 1858, ) is
not convincing: rather, as Loewenstamm sums up, we should accept
Albright's opinion that this culture word can be traced to
Hittite arkantma(n), with the sense of 'tribute' —> "expensive
dyed cloth', and that the colour notion is carried by the lexeme
39 40
only in the last stage of its development. ' The same usage
can be found in Ugaritic as well: so, for instance, in the
Ba'al cycle (Gordon, 137:37), where we find argrnn//mnh (=
^offering', 'tribute').41 For knowledge of Near Eastern trade
in purple and the various terms designating it the reader is
referred to the relevant works by Goetze, Landsberger,
42
Thureau-Dangin, Veenhof, Gradwohl, and recently M. Elat.
43
2.2.6 Tur-Sinai equates /DAI* with v/OPX and IttAIN with nnpl.
Traces of a common origin might be found in 'IfifJ'iS tifly (Ezek. 27:7a);
nnplM lt> 1B)K...le;jn (17:3); and especially in n£a.!> ^in flint?"!!?
(Ps. 45:15), WB...n»»t)»j?1...nBir» (Jud. 5:30), and DfTOAl (Ps.'
TT • -T! • Ts• TT; •
68:28). Nevertheless, even if the two lexemes share an ultimate
common origin, the actual slot each occupies within biblical
Hebrew is distinct from the other's: while IttAlK has come to
denote (by a narrowing shift) a certain type of dyed wool, flflpl
148 BIBLICAL COLOUR TERMS

44
still refers to multi-coloured cloth.
2.2.7 Jo conclude: from the colour point of view, IttAlK falls
• 45
under the superordination of QTN although it denotes more than
one specific shade. Thus, the author of the Temple Scroll writes
DITtf IttjnN (sic!), so as to exclude those shades that are not
46
typically 'red'. ffojn encompasses a colour scope running from
heliotrope to deep-sea blue to violet or even green, 47 which is
better defined by parallel non-Hebrew terms than by the Hebrew
terms or contexts themselves.
2.3 'WHITE' TEXTILES - 03^3 ,1111 ,\M2l ,m
2.3.0 The lexemes grouped under this heading refer to certain
types of cloth of various natural fibres, which have two
qualities in common - they are expensive and they are cleaned/
bleached, although no colouring additives are applied to them.
Hence, they fall under the blanket designation d'33^ d»*TA3
48 *"r; 'T;
(Qoh. 9:8), but their other aspect, that of 'type of cloth',
is as typical as the typical colour appearance.
2.3.1 The diachronic difference between the earlier OT and the
later \vi3, both denoting 'fine undyed linen1, has been well
established by A. Hurvitz.49 The terms differ in etymology,
that is, should be seen against separate cultural - geographical
circumstances as well as chronological factors, although the
identity of the product referred to seems not to have changed.
50
WO is of Egyptian provenance, while V-l^ is of properly Semitic
origin and continues to be vital in MH and in Aramaic, through
51
which it passed into Greek and other European languages.
Recent attempts to establish a Sanskrit or Egyptian provenance
for y.13 are far from convincing. 52 Moreover, a 'white1 denotation
- even if secondary - is apparent. Cognates are to be found in
53
Arabic 'aJbyad' = 'white'; Heb. flSPl*; and Akk. pisu, pasu = 'be
54
white', 'colourless1, and 'pesu1 = 'white', 'colourless', the
55
latter as modifier for 'wool' and 'clothes', as well as lexemes
referring to other entities. Therefore V'13, or more properly
/i>yd, originally evoked a colour quality identical or similar to
that covered by "p!? in biblical Hebrew.
Another equivalent to WD - this time a synchronous one - is
"T3, which is understood as a 'white linen material1 too. Such is
the employment of the lexeme in the P sources (Ex.) which describe
MAN-MADE COLOURS 149

the priestly garments.56

2.3.2 "Hfl is peculiar to the book of Esther. In 1:6 we have
IttAIKl VlX..rtom DOID lltl; and in 8:15, ...l-lhl fttttn ITO 5tt ensttt
. innKl VIS VIDltt If we disregard 0£)13 for a moment (see below,
2.3.3) - for it seems to denote roughly the same entity as 1-lh,
or at least the same colour; is lacking in 8:15; and disturbs
the inner balance of the double barrelled pairs - we are left
with the formula "iltl + VIS + Jlton + IfcAlK, which refers to
expensive 'white'/'purple' cloth symbolizing royal splendour,
power, and authority. 1!)h, then, is a synonym - at least from
57
the aspect of its colour - of V3S, probably used here for
literary/stylistic reasons. Further, the personal name 11h,
perhaps also >*nn and ">13n, were quite popular throughout the
' co T
biblical period. If "rtn - 'cloth', 'lit! (Isa. 19:9), and
these personal names are all /Tin-derived, then we must assume
that v/llfl originally featured in Hebrew stock but was later
rejected as an uneconomic doublet of "p^, leaving traces only
Tr
• 59 The renewed
in nomenclature and technical terms (ISh ,''Ifi).
popularity of v"Tlf through a growing Aramaic influence
facilitated the return of 1J|H as an allosemanteme of VIS, but
only in this isolated case. Meanwhile VIS has supplanted W2),
and TIM is of no real significance outside our formula.
2.3.3 0313 (Esther. 1:6 - ntoni 03"D Tin unsttt) came to Hebrew
60
as a loan from Sanskrit through Persian together with the
product ('cotton') carrying the same name, which was first
61
brought to the Near East from India by Sennacherib. Within
our context it functions as a gloss or synonym to Tin (cf. above) -
s-*\

in Sanskrit .karpas signifies 'natural', 'undyed cotton', and

this seems to be its meaning within the present context. No
wonder, then, that it was not accepted as a general designation
/• -7

for 'cotton' in MH, but was dropped in favour of IflA 1)3^.

2.4 MULTI-COLOURED TEXTILES
2.4.1 D^nna
Ezek. 27:24 - t^JTia ^^S-l ntti?*!! t\"?3t\ 'oi^AS
d^ttia is a loan from Assyrian birmu - 'multicoloured,
' : cloth, probably woven of threads differing in hue,
variegated' 64
65
not dyed afterwards. For general considerations concerning
6
6
66
birmu cf. Landsberger's definitive article. Appropriately,
67
Q^QIiJ occurs together with n^rtl and MQJ71; if an analogy is to
150 BIBLICAL COLOUR TERMS

be drawn, it should be deduced that d>ttia, like the other two

items, was quite valuable.
2.4.2 j/dpl ,n»p1 and related lexemes.
. T 5 •
dpi (Qal part, functioning as an epithet noun) and •$%?[ are
widely distributed, appearing Throughout the OT from the age of
the Judges onwards. From the cognates supplied by the
f-Q
dictionaries - in Arabic, Ethiopic, Aramaic, and Syriac - it
seems that originally the reference was to a 'many coloured
object", or to 'make a thing many coloured by engraving, drawing,
69
writing, or broidering1. Thus we have not only Mttpl ^TAS -
and that especially in Ezek. (16:10, 13, 18; 26:16, for garments
worn by high officials of Tyre; 27:7 - sailcloth; 16, 24; Ps.
45:15) and dpi ilKJytt (Ex. 26:36, 27:16, etc.) = 'coloured weave1 -
but also nnpl = 'coloured feathers' (Ezek. 17:3) and 'coloured' =
T
'' 70
mosaic pebbles (1 Ch. 29:2).
The basic 'multi-coloured' notion occurs in post-biblical
language; and in Tal. Bab. Shabb. 107b. (in a discussion of
rmnn and -prm-Dri of Jer. 13:23) we read, Ktt^N 'PmilTntl
. ..n->5> -»yaa -pam laa -pniiisiah ia:n ->Kfi »ttp»i fapi ->NPT
niimsn ,^np*1 and V^IU are understood to refer to
1
(multi)coloured surfaces' - thus in Jewish Aramaic but probably
applicable to Hebrew, as the latter seems to carry the basic
notion common to the cognates.
The well-established status of /dpi in the language may be
inferred from the existence of the name dpi, which serves as a
71 72
personal name as well as a toponym.
73
2.4.3 To conclude: iiap.1 and dpi ilwytt(P), the regular Hebrew
terms for denoting 'multi-coloured cloth', stem from a root
signifying 'multi-coloured appearance'. As for the hapax d'ttia,
since it occurs in a context which contains MQp.1. as well, and
since the two lexemes function as synonyms within their context,
we conclude that they are equivalent in reference, and that the
74
motivation for loaning d^ttlS could have been merely stylistic.
MAN-MADE COLOURS 151

E.3 COSMETIC PREPARATIONS

3.0 Biblical information concerning cosmetic preparations is
1
scanty. H-IS is mentioned twice (2 Ki. 9:30, Jer. 4:30). A
7 3
verbal /!tfO occurrence - .VJp? . ??D (Ezek. 23:40) - again refers
to eye make-up. A problematic third instance is d^JPy nn^tW?
(Isa. 3:16).
The questions concerning the activities referred to and/or
the materials used can be answered by utilizing the cognates -
especially in MH and Aramaic - and knowledge of ancient Near
East customs. The answers, at least in the cases of .1-13 and |/"li7K>,
are not unambiguous. The colour effect, however, is not inherent
in the term but implied - the end product, so to speak, of the
process.
3.1 n-ia, K"5n5
Gradwohl rightly rejects a 'red, reddish' reference for ti-19
2 3
on grounds of common sense (vs. etymology). Rabin goes one
step further towards an actual identification by stating that the
components of eye-paint in ancient Egypt were apparently
different from those used in Palestine - hence Lucas's objection
4
to the identification of 1*10 as 'stibium' is not applicable to
our case.
Whether antimony (in Arabia), stibium, or arsenic
(Babylonia), bhb is a pulverized powder whose effect is to
7
blacken the area around the eyes to which it is applied. As
such it is used even today. Rabin claims that 110 should
therefore be recognized as the semantic equivalent of ^itlS,
t>Q13 (<-/5fi5),term whose etymology and extra-linguistic identity
"" a
is much more transparent, and whose distribution is wider (albeit
in post-biblical texts). The equation was already made by Ibn
g
Janah and Qimhi.
3.2 /W Pi.
Within its context D»i»y niliJWtt (Isa. 3:16) implies 'ogle,
10 11
look around'. Gradwohl, following the Midrash, suggests that
_ 12
we relate our instance to MH NIp'O + i/1t?0 - to paint with Siqra'.
Unless we assume that Siqra is not necessarily 'red paint', or
that already in bib. Hebrew v^i7W/"lt70 was expanded to denote
'paint(face)', 'make-up' in general, Gradwohl's suggestion is
not acceptable. The chemical composition of lOp'O itself is far
152 BIBLICAL COLOUR TERMS

from certain: fucus, minium, or ochre - all used for achieving

13
a 'red1 effect - have been suggested.
Within MH itself, it seems, there are two homophonous /Ipb
sequences: the one from which ftlilpb - 'looking, ogling' —?
'curious' - is derived; and a second one, which looks like a
denominative of the Aramaic KIJ^O. Thus we might assume that
0*3*3? mi|?wn belongs to the first sequence. This does not
necessarily imply that the two 1pd/CJ roots are etymological
homonyms. Ochre or iron oxides were used for make-up purposes
14
in Egypt, and were probably known outside Egypt too; bib.
Hebrew, though, lacks a name for that preparation. Later on,
through the Aramaic Klf'O, the sequence was developed along
different lines, that is, j/lpD Qal and Pi., meaning 'to paint
with Siqra1. The latter, however, is diachronically secondary,
reborrowed, and semantically distinct from bib. Hebrew /1pi2>,
which has its continuation in MH too.
MAN-MADE COLOURS 153

E.4 DYEING AGENTS AND PIGMENT SOURCES

4.1 ilK-19 (madder), lOS (henna), and oVlS (saffron) are known as
1
multipurpose dyeing agents from post-biblical Hebrew as well as
2
from other extra-biblical sources. Although the information
supplied by those sources may be applicable to biblical times,
there are only small hints in the OT itself for such dyeing
practices. HK-ia (KuMa Tinctorum, 'madder') is registered only
T
3 ' 4
as a personal name. 133 (henna) is mentioned twice (Song. 1:14,
4:13) or perhaps three times (7:12, but cf. BH ). As Zohary
points out, it is the fragrance of 133 that counts here: we
cannot infer that the use of henna as a colouring agent was
5
known to the Israelites at this stage. &313 ('saffron', Crocus
6
Sativus) occurs only in Song. 4:14, within the 'fragrance'
context of the previous term. It is a loan from Sanskrit,
7
denoting a product imported from India. Again, there is no
evidence of using 0313 for dyeing in biblical times. In post-
biblical times, however, it was locally cultivated and used for
Q

that purpose.
4.2 IKW (Jer. 22:14; Ezek. 23:14.
-T ,
1BW3 nW) (Jer. 22:14) - the context defines *\W>) as a paint of
some sort. The Akk. cognate points to a 'red' signification.
g
Gradwohl and the Dictionaries define it as 'minium, red lead'.
However, in Egypt red pigment was produced of ochre from the
10
pre-dynastic period. The Aramaic T 1*3ttttO - from inao,
originally perhaps 'arsenic, orpiment1 - was expanded to denote
'painting agent" in general, hence is of no value for our "TBfly.
It seems possible that "HJNJf is tl\e earliest equivalent of the
11
later attested 1710, used for make-up, painting and writing.
154 BIBLICAL COLOUR TERMS

E.5 WRITING MATERIALS

As in the preceding sections, the information is poor and

the colour reference obscure, especially when compared with the
material available in MH sources - for instance Wish. Gitt. 2:3,
oKQ>i7t> Nina i±i ^:m oiniwi Kip'os titn via •psnia *m Painting
materials of all colours of the rainbow have been found in Egypt
of the pre-dynastic period (5000-3000 B.C.) onwards. The
1
pigments are mainly mineral based, but some are vegetable based.
Egyptian ink, according to Lucas, was either red (iron oxide) or
black (carbon, charcoal, or soot).2
Bib. 1»*j (only Jer. 36:18) was apparently black, at any rate
dark.3 IKnu, once mentioned as used in monumental writing
(Ezek. 23:14) and once in painting (Jer. 22:14), is 'minium1,
'red lead', 3 or 'red ochre'; for the identification with |710
cf. 4.2 above.
The third possible member of the series - 11^ (Isa. 44:13) -
is interpreted by Koehler as 'reddish paint' (of iron oxide
base). Gradwohl concludes that the context is not definitive,
and that therefore the meaning should be left undecided. 6 I
think that his conclusion is valid.
MAN-MADE COLOURS 155

E.6 WHITEWASH, PLASTER

1> = 'chalk', 'lime' - Isa. 27:9.
Bib. Aramaic: Dan. 5:5
*PW = 'lime, whitewash' - Isa. 33:12; Am. 2:1.
Dt. 27:2, 4.

6.2 VA* The occurrence in Isa. 27:9 - *JinK3 rQTtt ^K to iniBn

Dl^SJn 1A - stresses the fragile quality of chalk, not its colour.
On the other hand, Dan. 5:5 - ttfttT KT> ty Kliem:i ^p!? infOl
ND^tt **r K^»n - can be translated "on the plaster of the palace
wall" (NEB), or even 'on the whitewashed wall of the King's
1
Palace'. The colour connotation here is acceptable. It seems
2
that TA* is an Aramaism in bib. Hebrew; as MH shows, it
remained in the language as a designation for the mineral, but
3
not much more than that.
6.3 *PU) . In MH *T*o has come to denote 'whitewash' - hence the
derived Qal and Pi. verbs, I^D, and the idiom tO*M?l T>0, one
T~ *
of the criteria for defining degrees of 'whiteness1. Further, we
have a clear IhlU/I^O contrast in M. Sukk. 4:9 - the two cups (for
libation) V>n >:iOtt TphBItt 1'nw K!?K "Pil VajW' As for bib.
Hebrew, no direct colour connotation is contained in Isa. 33:12
or in Am. 2:1. The allusion, if any, seems to come about by
free association. The crux of the picture is excessive cruelty
(Am.) and quick destruction (Isa.). On the other hand, Dt. 27:2,
4 imply a process of plastering/whitewashing the altar stones so
that they become uniform in exterior colour and suitable for
writing upon (v.3). Still, the evenness of texture might be the
more important factor, which means that the (inherent) colour
property is not evoked by the biblical term.
156 BIBLICAL COLOUR TERMS

F. INDIRECT COLOUR CONNOTATIONS

F.O This section lists lexemes which contain etymological,

phonetic, or semantic links to known and/or otherwise supposedly
lost Hebrew colour terms, but are not colour terms per se - their
chief function within the language system is not to signify or
evoke a colour sensation (sections F.I - F.3 below). In
addition, a short summary will present names of items/objects
which are characteristically associated with a certain colour
appearance (section F.4 below). The internal arrangement
1
follows the method adhered to throughout this study. I would
like to stress that in most cases of the lexemes cited below -
and especially in the case of personal names and toponyms - it is
extremely difficult to establish a satisfactory etymology.
Consequently, their phonetic links with known colour terms are
many a time dubious or conjectured. On the other hand, the fact
that generations upon generations of speakers and readers looked
for or saw the 'colour' link, even elaborated it for various
purposes, is quite meaningful, and certainly easier to detect
than elusive etymology of the regular phonetic kind. To put it
differently: the data supplied by folk etymology is more readily
available than any other. Hence, unless new material concerning
each lexeme can be added to existing literature, no discussion
will be attempted. The reader can find references to represent-
ative or sample opinions and prevalent views in the notes.

F.I PROPER NAMES

1.1 Personal and Gentilic Names

1.1.1 Under DTK
2
tnTK - Gen. 25:30, (DITK !»» NIP ID t>y...nTM OIKM DTNn) ,
and more = Esau; also >nTK (Dt. 24:8), Q'OnK (2 Ch. 25:14),
0"n*"TK (1 Ki. 11:17), rTPn'TN (11:1). For the wealth of biblical
3
and post biblical word play upon tilTK ,0"TK, and t)T cf. Zakovitz.
T
4"
Was it originally the name of a God? and cf. dlTN *niy (2 Sam.
5
6:10). It is also the name of the nation and of the land of
Edom, conceived of as named after the Great Father. Alternatively,
is there a connection with d*TN through the 'red rock' appearance
f
of Petra and its environs? Otherwise, the name may be derived
from /'dm = 'be pleasant', as is the meaning of the root in Ge'ez.
INDIRECT COLOUR CONNOTATIONS 157

While the etymology is uncertain, the literary-phonetic

similarity is well established.
VlttN - Isaiah's father, Isa. 1:1, etc. Otherwise, - from
V/VttN = 'be strong', or a shortened theophoric name (cf. (l)rPlttJN) „
Q

fP^Dh
T - Neh. 1:1, 10:2. Otherwise, a theophoric element
'~ "• 9
preceded by /M^n + *y
Tmn - Gen. 33:19, ch. 34 (9 times), and more; the father
of Shechem, after the term for the common ass - from 'reddish',
10 11
'greyish-red'. Alternatively, from |/10h -"weight, load'?
hit (Gen. 38:28-30) is linked by folk etymology and word
~V • 12
play to "OB and dlTK, and to the concept of 'bright colour'.
Gradwohl accepts the word play as a reflection of an authentic
13
etymological link; this, however, is shown by Landsberger to
14 15
be incorrect. Other \/r\*it derived names are hit, son of Re'uel;
16
hit, father of Yobab (both in Edomite genealogies); HIT son of
~~ ** 17 18 ~ *' 19
Shimeon; son of Gershon the Levite; hit. the negro C>aiD!i);
90 2l 22
the gentilic ^m-TCi; the theophoric n»tm and rpmn.
• •.-- * T'.-: T'.—.' 23
DMB - a Levite (1 Ch. 24:27), also designating a stone.
Probably a loan from Akkadian, where samu is 'dark red', samtu -
24
Carnelian.
ybifl, son of r\N>13 (kermes, madder); ^iftfl. 2 5 lilp^ the
26 27 28
Hittite, also as part of the toponym 111GJJ Ifl, and 10^.
« '29
Together with 10y ('gazelle') and filfly (also a toponym), TllpX.
is considered by some to have originated in an y"l3y sequence which
30
in Arabic has a 'reddish-white' reference and is perhaps also
31
the base of ID)?.
TT
1.1.2 Under l!l£
•p!? - Jacob's father-in-law, 52 times in Gen. Originally
TT
32
the old Assyrian moon god. Note the covert ironic word play
33
upon the name in Gen. 30:34-37.
n/KJJ^ - n.pr.m.
r Ezr. 2:45; Neh. 7:48,
TT; 34 4
"Oll^ - Son of Gershon, also family name. A reference
35
to the colour of the body?
• 7/r 7-7
1h^ - Father of Efron; a son of Shimeon; of Judah. '
Interestingly, both names of father and son - Illfly ,1h2
contain colour allusions.
Tltl, the name of five different persons, 38 one of them a
39 40 41
Midianite (Num. 31:8, Josh. 13:21); 71-iti; n«in; anand fil^n?
• — T'
all seem to be based on \/^1^> the interdialectal Aramaic
158 BIBLICAL COLOUR TERMS

equivalent to /"pX If so, they confirm the view that /llh,

originally of Hebrew stock as well, was dropped in favour of
42
/155, and later partially restored through Aramaic influence.
tO'y - Gent, and n.pr.m., see ^y in in the next section
(1.2.2 below).
1.1.3 Under it\v
43T
31(1)y - from lily = 'raven', 'crow'; with Arabic, Akk.,
" 44
Aram., and Syr. cognates of a 'be black' connotation.
45
Up - gentilic and ancestral name of Arabian nomads.
A comparison with IflB proper is to be found in Song. 1:5, nilHW
lip '!?i1KX..mK:i1 'OK. See also |/11p in verb formations; and
Mllp, maYll?, limp, (wady), and MH ni*1p = 'pot' (blackened by
'-" ' "* '' 4h "*" "\
fire), bib. Heb. 1-113.
1.1.4 Others
47
dpi - a Midianite ruler, two Israelite individuals, and
48
a town within the Benjamin area - cf. dpi ,r\0p1(o A possible
explanation for the name may perhaps be surmised by assuming a
shift from 'multi-coloured appearance' to 'splendour', 'beauty' -
which cannot be ascertained from existing lexical sources.49
1.2 PLACE NAMES
1.2.1 Under DTK
, T ,
DTK related toponyms are probably tillK (1.1.1) and d>»1K ilt>yn
T
?() on ' ''"'•
(Josh. 15:7, 18:17), Arabic Tal'at ed-dam. On the other han
nnlK (Gen. 14:2), n»1K (Josh. 19:13) and Sp-lil *01K (Josh. 19:3
T;— TT-: ••'.*" • T~;
are directly related to i101K4 = 'ground, land, earth'; for the
possibility of a link between nni{<. and dTK see below, 2.1.1,
51 : 'T52
p. 161. plE; t>in (Jud. 16:4) fr from ple>, chchoice vine of 'red'
" 53 "
= 'dark' grapes. Eshel suggests that the place was named after
54 • 55 '
the vines grown in it. np12W_in Edom - Like dlTK itself,
with connotations of 'redness' (cf. dlTK^ 1.1.1 above).
fl^Dhn rum (1 Sam. 23:19, 26:1, 3)56 -"a 'dark' hill? Cf.
^»tOn ,n-l^h, and n^DH.57 Illfly in;58 may = et-rayyiieh.59
1.2.2 Under 1!lt>
•\& (Dt. 1.1) - see flJ^ (Num. 33:20, 21). ,1^ - two
TT T' '
cities of this name are known: one in South-West Judah, in the
60 61
plain, and the other - a station in the Wilderness = *pt> above.
T r
' 62 '
n:n^ - Modern Lubban, near Shiloh; no connection with roi^ =
'frankincense' apart from the homophony. "jlJl!? - 70 times.63
INDIRECT COLOUR CONNOTATIONS 159

n^ lin^U - in Asher (Josh. 19:26). While 1irPU> usually

designates the border of Egypt as demarcated by the delta or one
64
of the branches of the Nile, here it denotes a river. Aharoni
identifies it with the Kishon river and tlil^ itself with Tel Abu
65
Huwam, while Mazar - with Salmona. TltVW itself is usually not
considered an offshoot of IhB = 'dark, black', but the appearance
of both 1HW and 13^ sequences in the same compound name is
striking.
In general, the question arises: is there a transparent
connection between the colour term "p!? and the toponyms that seem
to belong with it etymologically? A. Cohen shows that there is,
at least in the cases of flJISt? (Lubban) - where the colour of
T '
the local earth that is used for pottery as well is whitish-
yellow, and a neighbouring village is called Huwwara - and of
66
M33!?. The same is probably true for limb, from its snowy
T:
67 "r*
appearance, cf. 'Mont Blanc'.
IDS (Ezek. 27:18). If it is not a direct colour denotation
but a place-name (1!"l¥ IBS//113^(1 V>, in a context of toponyms
and ethnic groups and their commercial specialities), and if the
68
place-name is not to be amended, then the location has to be
69
identified. The 'colour1 associations, especially with "ittS,
are unmistakeable. "y^y "in.70 ttt'y, also a gentilic and proper
T
71 "
name, is according to Fronzaroli of a North-Western Semitic
72
root signifying 'bare, white'.
1.2.3 Under iVlB
T
Ilin (Ezek. 47:16, 18) - the Eastern part of the Bashan,
73
modern Jebl-ed-Druz. From a Yemenite cognate meaning 'black',
after the colour of basalt?74 11' lift bfl3, 75 east of Jerusalem;
76
cf. /"HP- According to Gesenius, from the 'turbid' appearance
77
of the water.
1.2.4 Under pil>
11i?"v»il >» (Josh. 19:46) - Perhaps through the vegetation of
* 78
the marshes around the river. dyftl*, although usually not
T:-,T 79
considered part of this sequence, might also belong here.
1.2.5 Others
80
Dpi, a toponym and the name of a Midianite ruler, is the
"*" 81
old name of Petra. In Josh. 18:27 it is designated as a
1
OO Q ?
Benjaminite locality. Mini rP!l and ti>im "m - Opinions cited
160 BIBLICAL COLOUR TERMS

84
in BOB attribute iTUM, at least, to our 1J3J, after leopards (?)
"*"* * "f OC
found in the place, or the striped or spotted appearance of it.
The etymologies suggested for tPIQJ HJ are different.
INDIRECT COLOUR CONNOTATIONS 161

F.2 COVERT COLOUR ALLUSIONS - POSSIBILITIES

This section contains material relating to lexemes denoting

objects or concepts that are often linked to and/or derived
from root sequences used - or once used - for signifying colour,
whether in Hebrew or in cognate languages. Lexemes referring to
precious stones and metals are here excluded and will be
discussed below, section F.3.

2.1 Under DTK

T
2.1.1 OIK and DT ,DTK, and MttTK.are so similar phonetically
that they absolutely demand paranomasia and speculative
etymologization, in addition to the DTK fromfifcTK.and T3JJ.
(Gen. 2:7, nOTKn in 13y DTXn 11K D^K 'n 12" 1; 2:19; 3:19;
Ezek. 28:20; Zeph. 1:2-3) conceptual link which the Hebrew
creation story shares with other Eastern myths. The DTK-lay
TT T*
MttTK (= 'red earth1) - DTK series is strung together by Josephus,
TT-: 1
Antiquities. DTK and DT are notably linked in the chiastic
formula (Gen. 9:6) T.Dtsn inT DTK!! DTK?! DT T.pe>, while DTK and DT
are joined together in 2 Ki. 3:22, DTD D>»TK D>feiTl<, Gradwohl
correctly asserts that apart from the basic religious significance
attributed to blood and to D*1K as the colour of blood, no genetic
3 ^
link can be established. DT appears to be an original bi-literal.
We might speculate that DTK is a morphological expansion of the
basic stem to which first a specific colour value is attached
and then the chromatic/saturation/brilliance aspects are widened,
so that the original 'blood colour' sense remains only as a
focus. If so, the parallel evidence of the cognates shows that
this expansion occurred very early in the parent language.
However, one cannot go beyond the realm of cautious speculation
in this matter.
• 4
Similarly, a generation fittTK—} DTK - even if the synonym
10y and lexemes related to it, like T3y = 'gazelle1 and various
5 **
proper names, might carry indirect colour allusions close to
DTK - is impossible to establish beyond the literary-conceptual
link attributed to the two lexemes. DTK and DTK perhaps present
a different case. DTK can be related to Arabic adan?atun = 'skin',
TT 7
and alternately to the Ethiopic /'dm = 'be pleasant, good looking1,
which could represent a non-colour polysemic development within
Hebrew itself.
162 BIBLICAL COLOUR TERMS

2.1.2 Other lexemes linked to DTK subordinated terms are listed

below.
piu; (Isa. 6:2; Jer. 2:21), np"lB? (Gen. 49:11), and n^-lio
T
"*" v • • 99
(Isa. 16:8)8 are all named after the "
colour of the grapes = pit!/.
10 1
/inn related terms are inn 'wine (Dt. 32:14, Isa. 27:2;
Ps. 75:9 - cf. BH ), 'blood of grapes'. Although it is usually
considered an offshoot of i/inn = 'ferment', 11 this is by no means
certain.
It is equally satisfactory to relate inn to /inn = 'be red'
12 '*'*•' 13
(of some sort), together with Ibn Janah and Qimhi, and as
14
mentioned by Levias. Less doubtful is the 'red' element of
11*tt.r\= 'ass1 and 1-inni = 'roe-buck' (Dt. 14:5; 1 Ki. 5:3),
Ullendorff relates Infill to \/inh = 'carry weight' .15 This,
however, does not explain 1-1 J3h^, which seems to derive from the
same root. Furthermore, Gradwohl 16 mentions the Spanish burro =
'ass'. When we consider that burro is probably derived from
purros = 'red', and that other animals also have 'colourful'
names (103? ,inJ), the comparison is quite relevant.
2.2 Under p!?
TT ._
2.2.1 t\*tf> C//i1»n) = 'moon'; .Ul^ = 'frankincense'; iTi^ =
Storax officinalis (Gen. 30:37, in a passage playing on '"pV);
and even MJl^ = 'brick' and its derivatives - /*p5 Qal and
perhaps ISl^n - all contain a transparent phonetic/semantic link
to the basic lack of chroma/luminosity (-> purity) inherent in
Itf.18
TT
2.2.2 Others
WIl*, pi. D^ito19 - of fowl or snakes - is perhaps
"70
etymologically related to V-li. *»*?13f (Job 28:17) = 'glass',
21
from i/TDf, see HT* ,131. iwio' ,nju;iw, 22 CPpnyn WlUIB) 23 - from
* *" *yA ^ *y c T"T
Egyptian sssn, ssn, cf. WW = 'fine white linen'. The IKJIty
is identified as Lilium Candidum, the Q'pnyn lUWlia as the
9^- ,
Narcissus. Both are predominantly white flowers, yo ,i13^B
—i—2 7 T T"
/5*W = 'be hoary, grow white or grey with age' —> 'grow old'
(thus !lb etc. in MH). The colour notion is especially pronounced
in passages like 13 np^lT n3>U; (Hos. 7:9); il^W niKflh n*Wy (Prov.
16:31); and n^^e; D^JpT 11H (Prov. 20:29). On the other hand,
the colour allusion is insignificant in the idioms il^lt)
T
ra»U»
~r ••
=
'great old age' (Gen. 15:15, 25:8; Jud. 8:32; 1 Ch. 29:28), of
INDIRECT COLOUR CONNOTATIONS 163

nt>1KU; tn^n ira»b UN T>11h = 'let a person die a natural death'

(Gen. 42:38; 44:29, 31; 1 Ki. 2:6, 9 - dna instead of OlbKO in
the last instance = 'cause a person to die by violence'0
2.3 Under iVlB
T
2.3.1 JVilhW (Qoh. 11:10) = 'young maturity', from dark colour
~f~ 28 '
of hair; ino (Joel 2:2) = 'darkness'; and 1int!> (Lam. 4:8) =
29 *
'darkness/blackness'.
2.3.2 Some ."lK>nr usages, notably llhwn .IWh (Lam. 4:8) and perhaps
' 30 :•.T
also VIKfi lentil (Ex. 10:15), are a so intense that they seem to
signify blackness, which is an intensified form of darkness.
1/ITp = 'be dull, dark', when not in conjunction with /55tt
= 'mourn' (Jer. 4:28) or in the derived sense of 'mourn', ls a
conveys a sensation that recalls the one referred to by inu) (so
also n.1YTp, Isa. 50:3; mpYTi?, Mai. 3:14). In the OT almost
half of the v/'VP occurrences relate to loss of light from the
sky (Jer. 4:28), sun (Job 30:2832),or stars (Ezek. 32:7). inw
is never used in such a context. Perhaps we may surmise that
y"VTp and \/irny are partially synonymous in reference, although
different in regard to application and context."
1*1»3 - Job 3;5, with H^H and M^y as parallels - the
context and the Syriac cognate point towards a 'blackness',
34
'darkness' denotation.
35 36
nitt^S - mainly in Job (9 times) and Ps. (4 times) but
37
also in Isa., Am., and Jer. The traditional pointing reads
nin^S, 'shadow of death'; this meta-analysis was accepted by the
ancient Versions and older commentators. Modern research,
however, has shown that the correct form is Mtt^if - from /D^ -
38
with cognates in Akk., Ethiopic, Arabic, and possibly Ugaritic,
all with a 'be black/dark' reference. For a survey of the
material the reader is referred to S.M. Paul's article. 39
llTKfl (Nah. 2:11; Jo. 2:6} and 1*1$ (Num. 11:8; Ju
1 Sam. 2:14) = 'pot'. I agree with Gradwohl's conclusion that
40
the two are basically identical. On the other hand, his
interpretation of 111X3 liDi? = 'were scared' (in a context of
war) is not adequate, perhaps because he tries to establish a
transparent relationship between the components of the idiom and
the idiom itself. This can hardly be done; a better result is
achieved, I think, by comparing the idiom to others which
contain similar elements, that is, OIKtl IIHBM ^WH (Lam. 4:8
164 BIBLICAL COLOUR TERMS

perhaps also fyyn irn>> my y (Job. 30:30), and certainly to the MH

—r" ~T
JVinpn ^IKOoo.'OD IMnj (T.Bab.Shabb. 30a) and ^IBO VPJfl 11>nu>n
•rmpn (T.Yer.Hag. II, 77d).
2.4 Under pV
T
* 41
2.4.1 pTt ,\)^">
TT
are discussed under P11*, Primary Terms.
'."•' 42 43
lip*!".,- 5 times in the pair llpVI llfl*TC/ and once with D">:i3
T
44 '
features regularly in MH too. Usually llpl* is defined as a
wheat disease ('mildew', 'rust') caused by bacteria. Hare'ubeni
45
discusses the possibility that by equating 11£J;Jlp and HP*!?, one
can discard the 'disease' notion, and interpret: too much rain
(Am. 4:7, *HM3fl fiDN npt?h) causes TlpV, yellowing of the ears;
while little or no rain (UD^Il,, „ .VOmi N!> 1WK rtp^m), together
46
with a drying wind, causes llflW
T
(v. 9).
47 '
2.4.2 "pjn - 19 times, in Dt., Ki., Jer., Ezek., Isa.II,
Hos., Ps., Job, Song., and Ch. Gradwohl accepts majority ruling
on this matter, which considers 13VI to be 'fresh, wet, luxuriant',
48
not a colour term. So lately Morag, who connects the root
' ran' 49
with Arabic CsJ d suggests a basic sense of 'high, lofty'.
Nevertheless, the fact that most occurrences are in syntagms
containing either 'tree' or other references to vegetation should
leave the question open.
INDIRECT COLOUR CONNOTATIONS 165

F.3 METALS, GEMS, STONES1

3.0 The Ancients did not assign names to gems according to
their mineral combination but according to external properties,
mainly brilliance and colour or the strength and toughness of
the stone.2 Hence confusion in identification and translation
of biblical terms for precious and semi-precious stones was
inevitable. Add to that the tendency to transfer names of such
rare commodities through countries and languages together with
attempts at etymologizing within the target language, and
identification is made even more difficult. Thus the Versions
3
and sources are not trustworthy. The same applies to Num.Rabb.2,
which describes the stones and typical colours of the Twelve
4
Tribes. The comparative table of the jewels prepared by
Shalem illustrates these points succinctly, and is therefore
reproduced here.6
LXX Mid.Rabb. Syr. T Yer.II T Yer.I T Onk. OT
TPYIO msYw Npn»o Knpao Knp'fe'o ipno DTK
TPTOIO "pnomo NP*P NnpT NnpT> Ipv rnoa
DITAIOQ iinjppn Kpin Kjnpin Kiipis pin np*o
opitrtN TUPTD Kins KrrmD rrmT^K inr-imfm IDJ
01-pflOO li:P1»QKO KJ^flD NIPblflKPb Tl^-pOO fT^W 1"»£JO

o^flON* T>n>nnt'K KnypJ rtw vy T^TD o!?nao ot>,T

IT'IU1'!? V^DID pi'JDJp miT 11J>T>DJp K^IUp DW^

O'UKDK D^DKaK KJT3n3 V^ll^^ VPI^ NVU) "\Z\D

oioo^onK n>o>?a>n x!>Ay "py vnAiar K^y vy ti.'y^y w nat>nK

D10'>t>n'>1D VODKlIp WBin KM* D113 Ktt1? 0113 KO* D113 WWin
K31

li>!?»ii VPI^KIS Ntn-D Kh^na ni^i^n K^m one;

K!?h
iT'DiaiK Do*^>ia naw K!>AIO rpann ^IOJQ naw^
VTlOi£3K

The lexemes which belong to our subject are those defining

the gem itself by its colour, whether by using original Hebrew/
Semitic stock or through a loan from the language spoken where the
gem was imported from or through. Similarly, names for metals
will be mentioned only if there are indications within the text
itself of conscious usage of the term for conveying colour
properties per se, or where a reference to colour is etymologi-
cal ly inherent to the term.
166 BIBLICAL COLOUR TERMS

3.1 Gems and Stones

* 7 8
3.1.1 DIN, the most transparently constructed, is identified
V 1 9
either as 'Carnelian or as 'red ruby'. However, Carnelian is
10 11
probably DflO, a loan from Assyrian Sama, Samtu, unless the
latter designates another 'red' gem in Hebrew; that is, the
reference within the target language does not necessarily have
to correspond to that in the original language. In most
Versions DMKJ is identified as 'beryl'. 1SUJ, another 'red' gem
(Versions: beryl, agate) is perhaps a loan from Assyrian too
(so BDB and others) - cf. aban sabi, of the colour of uncooked
12 ' 13
ox's blood. ^75* tentatively identified as 'ruby' or another
14
red stone, is variously understood on the basis of Hebrew |/YT3
= 'strike fire',14 Arabic ^YV'-V o-VJ_>J 'extreme redness', and
16
Ethiopic. On the other hand, Rabin raises the possibility of
17
a derivation from a place name. If this is accepted no colour
• 18
significance can be attributed to T31!). Finally, nn^DK is
!"" TT .' —
usually identified as 'amethyst' of 'reddish/brown' or 'wine'
19
colour; the derivation from Egyptian hnmt seems superior to
the Hebrew-based etymologies from /ti~5Ff = 'dream' or 1int?n (=
20
colour of egg-yolk).
3.1.2 Opaque colour references can be traced for the following:
2i
?y.TV)3, yellow 'topaz' or 'chrysolite' (Aram.T from/P*^) -
perhaps from pita, in Sanskrit 'yellow', 'golden',22 although
Job 28:1 - KM3 tlTDO - assigns the gem an Ethiopian origin.23
24
np^|3, although previously linked to Hebrew |/p13 -
'flashing', 'sparkling1 gems (Aram. T) - is currently considered
a loan from Sanskrit markatu and identified with 'green' stones
25
such as 'emerald', 'malachite', or 'green beryl'.
' 26
133. Following the Versions, it is variously identified
27
as 'emerald', 'turquoise1, 'ruby', or 'carbuncle'. However, if
28
it is the same as 1-13 'JiK (Isa. 54:11; 1 Ch. 29:2) then, as
:
29 " '~
Qimhi says, a 'black' stone is indicated, and this is supported
30
by the Syriac KT^TiJ. One of the dark gems - 'onyx' or a 'dark
31
jasper', for instance - is a plausible identification.
3.1.3 3 Two stones were recognized as 'blue', 'turquoise', or
'green-blue'. These are 1*9032 and K>»Enfl. 33 Passages like
Ex. 24:10, DSIO! *paOfi M^ flKOfaD 1^>1 nnm ^OI0> »r#X UK 1NT1
•liTH#a»nE>n the VSO-made celestial footstool/chair in Ezek.
(1:26, 10:1); the celestial wheels which are made of W^enn
INDIRECT COLOUR CONNOTATIONS 167

(l:16ff., 10:9ff.); and the Aramaic rendering of 0»nn as D11D

M3', 'colour of the sea' 34 - all these point away from the usual
'sapphire', 'topaz', or 'jasper' renderings. Qimhi rightly
defines KPenft as Jtonn VyD MlPyl miD PK,35 Thus *PDO could
36
be identified as 'lapiz lazuli', and KPUHJl as 'turquoise' or
another 'blue1 stone.37
3.1.4 W>V^ and the contracted form VK), which appears in post-
exilic sources only,38 were interpreted " as 'marble' and understood
to be light coloured, 'white' or 'yellowish' (cf. EW = y.13). 39
On the other hand, they do not serve as colour indicators or
substitutes. OTja (a hapax - Esth. 1:6, KJBM Ditt), of Egyptian
provenance, designates 'alabaster': both Egyptian and local
40
alabaster ware were of white-yellowish colour,
3.2 Metals
Although gold can be processed to achieve various degrees
of purity and colour,41 a typical 'gold' colour was apparently
thought of as 'yellow' or 'yellow' tending to 'red'. Ezr. 8:27,
2
}iVrj) nVinn.o.^n^n nwnj »^D1. is indicative. The etymology of
TT : r;
~ " " '-- ^^ 43
V-11h inUg., Akk., and Phoenician; Arabic <^Jr_s > 'be yellow';
post-biblical Hebrew, Aramaic and Syriac y'lh (= 'safflower',
44
from its colour) - are all connected to 'yellow' too, and so
45
is biblical V11H P1PT» (Ps. 68:14). dH3 - a synonym for 3nT
"•"•' 46
(Ps. 119:127), Ylin (Prov. 8:19), and t£J (Song. 5:11; Isa. 13:12)
- becomes a designation for 'blood stain' in MH and in Jewish
Aramaic, thus preserving an indirect link with the most popular
colour of the metal. Finally, 103 - 'silver', perhaps the 'pale
47 "•' '•'
metal' - is not used in any 'colour' sense in bib. Hebrew, but
is the basis for derived terms denoting 'pale' and 'become pale'
48
in MH and in Jewish Aramaic. However, from the etymological
viewpoint, ^t)OD cognates and/or inter-language equivalents
sometimes have other denotations: thus Akk. Kasapu, Arabic .
<* '• = 'break into pieces, tear', ^?*&
^jr2-p «LJ!*
-"ci 'silver', cL_55^-~^
49 ' —
= 'of silver'. Cf., nevertheless, the Arabic etymological
equivalent to Hebrew v^DD, e.g. {_>»-—-^ and related terms. The
basic meaning of the latter - 'discover, uncover, open up' - is
perhaps connected to a basic sense of 'light', 'bring to light',
'throw light upon', which is common to the cognate sequences in
both languages.
168 BIBLICAL COLOUR TERMS

F.4 COMPARISONS AND COLLATIONS: OBJECTS/ITEMS EVOKED IN

ORDER TO CONVEY TYPICAL COLOUR PROPERTIES.

4.1 For D'TK

T
DT = 'blood' - 2 Ki. 3:22 (water).
"P> = 'wine' - Gen. 49:12, Prov. 23:29 (eyes).
D^j->ja = 'corals' - Lam. 4:7 (complexion; or lips? cf. BH
•OK* ,y!nn = 'crimson', 'scarlet cloth1 - Isa. 1:18; Nah.
(D'yblttS; - sins, warriors).
4.2 For pb
TT
3t?n = 'milk' - Gen. 49:12 (teeth); Lam. 4:7 (complexion);
Song. 5:12 (eyes?).
nyD = 'gum resin' - Num. 11:7 (Manna).
T> = 'coriander' - Ex. 16:11; Num. 11:7 (Manna).
IBS (natural or bleached) = 'wool', Isa. 1:18; Dan. 7:9
(Aram.; hair).
At>K> = 'snow' - Ex. 4:6; Num. 12:10; 2 Ki. 5:27 (all for
nylS consistency/texture, not colour); Isa. 1:18 (sins); Ps. 51:9
(sinner); Dan. 7:9 (Aram.; clothing).
4.3 For iflB
Tina = 'blackness', 'darkness' - Lam. 4:8 (skin, appearance).
*B-13 = 'African', 'dark-skinned person' - Jer. 13:23
(appearance of human skin).
Yip ^flN = 'tents of Tip' - Song 1:5 (complexion).
T"
:niy = 'raven' - Song. 5:11 (hair).
4.4 For nYl*
T
SMT = 'gold' - Ezr. 8:27 (copper); Zech. 4:12-14 (Silt as a
poetical equivalent of oil, 1HX*).
T£3 dill) = Song. 5:11 (hair; referring to sheen and gloss,
not to chromaticity).
4.5 For 'blue', the colour of the sky (no primary term in
evidence).
V90 = 'lapis lazuli'? - Ex. 24:10; Ezek. 1:26, 10:1
(heavenly footstool/chair).
KPunn = 'turquoise'? (= 'sea colour') or another blue-green
stone - Ezek. 1:16, 10:9 (heavenly wheels).
MULTI-COLOUR PHENOMENA 169

G. TERMS FOK SPECKLES, SPOTS, AND OTHER MULTI-COLOURED

1
PHENOMENA.

G.I This category includes:

a. A group of lexemes distinguishable by a morphological
feature, the pa°ol pattern. All the lexemes occur within the
stories about Jacob and Laban's herds (Gen. 30:29-43; 31:4-18):
*TpJ and *Tpy* here only, Ylil* also in Zech. 6:3, 6 for horses).
.T . *r r 22
*Tp.J and Ipy* serve as bases for derivatives as well.
T -r
b. Various other terms, each denoting a visible entity
whose characteristic is an uneven colour appearance (Sections
2.1, 2.2, and 2.3 below).

G.2 Miscellany.
2.1 Terms appearing in Gen. 30 (in addition to *Tp3 ,*Tpy* ,Yl!l*).
K.lt>0 - of 'spotted1, 'variegated' sheep3 (Gen. 30:33, 35) -
occurs in opposition to TpD ,1py* ,0^; no colour signification
can be detected, though - Qimhi is probably right in saying that
the difference between *Tp3 and K-l!?0 is in the size of the stain
4 *
referred to. The related niK^Ott = 'patched' ('sandals',
Josh. 9:5) point to the same conclusion. niK^D Ulna (Ezek. 16:16)
is usually interpreted according to the sense in Gen. 30.
D^Wp/d'^TO and tPOOy^ (Gen. 30:41-42) are usually understood
5
as carrying a 'strong'//'feeble' contrast. At first glance this
interpretation seems satisfactory: thus, the story concludes,
Jacob managed to breed strong stock. On the other hand, the
semantic part-equivalence of /iwp // /Tpy ('knot', 'bind'),
together with the occurrence of 0**Tpy earlier in the same
chapter, facilitate another interpretation. Similarly, although
0>30y ,tpD3?M (v. 42) are assigned to \/t\vy = 'be weak, faint',
6
they' may belong with a homonymic /tlOy sequence which otherwise
occurs only in Ps. 65:14 and 73:6. This sequence - denoting
'envelope, cover' - has an Ethiopic cognate which means '(a)
cover1, but also 'a web', 'texture', the latter being much more
suitable for our context at least as far as O^spy is concerned -
cpuynu is better understood as 'when (they) grow weak'. All in
all, most of the terms relevant to our discussion are almost
unique to our narrative (cf. 0-1 h). Therefore, one could perhaps
hazard the conjecture that TlBp* and IBfiJtt* here, or in the
source-material underlying the story, originally signified
170 BIBLICAL COLOUR TERMS

approximately the same as Ipy*, while H^oy* referred to the

opposite notion of 'uniform texture1. This approach would mean
that Gen. 30:41-2 is to be considered a distinctive substratum
of the present narrative. And to conclude, the usage of D'QOy
and tpOJtfD, albeit with different senses, could be viewed as a
deliberate case of paranomasia.
2.2 Lexemes referring to skin areas which are distinct from
the surrounding skin. Both pita (Lev. 13:39) andftlft.3.(Lev. 13,
""" 8 '*
8 times) denotes 'bright', 'shiny spots', set off by a
differently coloured background. In the OT both appear in this
context only; however, in MH and in Aramaic both are vital and
productive.
mi:mn*- a hapax, Jer. 13:23,
l»jnai:in in:n Illy W]5 lorpn - refers to marks, stripes, or
T \\
spots. The exact shape of these delineated skin areas, and even
their colour, is dependent upon the identification of 103 (see
below).
2.3 10J (Isa. 11:6; Jer. 5:6, 13:23; Hos. 13:7; Hab. 1:8;
Song. 4:8) and related lexemes. There is no agreement about
whether "l»!l is a 'tiger1 or a 'leopard', although the name of
r
the animal*" refers to the speckles on its skin. 9 In bib. Hebrew
we find toponyms which look as if they were related to the same,
consonantal sequence. These are M1QJ (Num. 32:3), Mini n^3 (Num.
32:36; Josh. 13:27), ?tP1»3 >E (Isa. 16:6). The latter, however,
do not help in solving the problem. On the other hand, MH usage
10
(Pi., Pu.) points to a 'spotted1 reference. While this seems
to exclude 'striped' or 'checkered' appearance, no further
specification can be made on linguistic grounds beyond the
notion that i/IQJ refers to some kind of delineated shape.
Alternately we might assume that the origin of 1)33 is obscure,
and that the signification /1J3J = 'of spotted/variegated
appearance' is denominative and secondary.
11
2.4 General considerations.
Only two of the above mentioned lexemes can be assigned to
superordinate terms with any certainty. These are pfl'a and flltl^
(2T2 supra.), which belong to the "\^*y dominated sector - from
'bright', 'shining', 'glossy' to 'white'. Otherwise, attempts
to classify the actual colour or even the shape of the spots,
speckles and stripes, from the ancient Versions onwards, are
MULTI-COLOUR PHENOMENA 171

less than convincing. 12 This evident confusion is the result

of the lack of clarity concerning the references the terms
supposedly carry. And no wonder: for the most part the latter
are so rare and/or idiolectical as to be almost opaque. Never-
theless, one conclusion emerges: the various etymologies point
more to the field of 'space/spatial arrangements' than to
colour. Gradwohl's attempt at differentiating grades or sizes
that "TpJ ,N.l!?0
T
,Yl3* denote seems to be the best solution
-r T 13 .
available for these lexemes. TPV* has no clear colour
T
signification, and the same applies to Million* and 1>3i. To
sum up: mostly one can only discern the colour contrast vis-a-vis
its environment. Still, this is a part of the colour sensation.
As such it belongs to the periphery of the colour field, as well
14
as to the spatial relationship field.
172 BIBLICAL COLOUR TERMS

H. APPENDIX: THE pa°ol PATTERN

H.I The Proto-Semitic pa°ul — Heb. pa°ol 1 pattern is utilized
quite extensively for the construction of colour terms, especially
those of the first three categories (primary, secondary, tertiary)
but also for the terms denoting colour differences of spatial
arrangements (*Ipy* ,VpJ ,*T13*) • The pattern is not an
exclusively 'colour' one - it is used for generating lexemes
which relate to phenomena from other areas of experience. The
common denominator is, supposedly, that they all denote an
2
'indefinite external quality'. Further, lexemes modelled upon
it cannot always be defined as so-called 'adjectives' - thus
n-m
•:— oWl
TT D'lKh
TT (Gen. 25:30) and pivT (//Witt, Job 39:8) definitely
function as substantives. As in many other cases in Hebrew, the
demarcation line between adjectival and nominal, and between
syntactical surface structures (vs. deep structures) and the
semantic significance applicable to any morphological pattern,
is fairly fluid. Already the Medieval Jewish commentators and
3
grammarians noticed that the basically Aristotelian division
into 'parts of speech1 or 'word classes' was not exactly suitable
4
for the description of Hebrew (and of other languages as well),
which created problems of definition. It follows, then, that if
a member of the series DTK ,1fH2> ,Hilif and so on is to be defined
as farbadjectiv (Gradwohl), it is to be done with a qualifying
statement, viz. that the ultimate criterion for classification
is functional (= syntactical), not morphological. Hence, the
pattern cannot be named D>y3^ *1Kn !?i?Kff3 unless we bear these
comments in mind. The fact that it was utilized almost
exclusively for generating new 'colour' lexemes in Modern Hebrew
5
merely reflects an artificial conscious choice that is irrelevant
for our data. On the other hand, in bib. Hebrew - as in MH - it
is not exclusive. Other patterns can be resorted to, be they
nominal or else verbal, even for the signification of a 'state'
or 'quality' - cf. -inTK vs. DTK , "pa^il vs. "p^, where no
',""" T •t ' TT
difference in deep structure can be discerned. Therefore, a
rigid description of alternative surface structures as separate
6
notional classes is not adequate.
The lists below contain (in alphabetical order) biblical
Hebrew data concerning the following categories of items:
APPENDIX 173

a. Colour lexemes constructed after the pa°ol pattern.

b. Other nominal/adjectival lexemes built upon the same
model.
7
c. Verb forms of the same pattern (the Qal theme).
H.2 a. Colour terms
."iriBiT ,|Hi!7*
T
,ins*
T
,in^T ,i\>y*
"t
,"r'pJT ,PivT ,*na* ,V»N*
T
T
,D*TK
T
Total: 10 terms.
b. Other nominals
i. Lexemes usually functioning as adjectivals (= noun
.modifiers).
,i 8 spy.&yT ,\iny
T
,&y ,iby ,\>tm ,rmT ,^H
T
,D'»K
T
"^ 11 ,•lopi(f ,D13>
.am - '9 ,n•spy
T T T T
Total: 13 lexemes.
ii. Lexemes which usually function as nominals.
,nann ,n!m ,rmN14 ,maK 13,na&K 12,q'oK* ,nnJ?K ,m>K
.naiy 16,?n'iy 15,m.iiS
T '. T T
Total: 11, one uncertain.
c. Verb forms
.fc»',itti> ,?Q^ .C^ia1') to> ,IA»
1
Total: 5 ground forms.
H.3 Interpretation of Data
Even after granting that distribution of formal-morphological
features - as much as that of lexical items - might be accidental,
the evidence indicates that the pattern signifies a state or
quality which is expressed either adjectivally or verbally. The
purely substantive forms, on the other hand, are mostly f.; at
least in some of the cases the lexemes seem to signify a
variation of reference in relation to the basic m. form wherever
this exists: cf. lias — miD flON* — nflDK ,TlflK — iVTflK. These
T -r\ T T\ ' •• -r'.,
then represent a secondary convergence from more than one basic
pattern. Broadly speaking the pattern cannot be considered an
exclusively adjectival one. Although four out of the five
primary colour terms belong to this group, ^together with three
subsidiary terms and three terms denoting speckles and spots
(list a), the pattern itself, as we have seen, is not unique
to this field.
H.4 The *pa al Pattern = Heb. pa al occurs in the colour field
in three lexemes only - -p^ (primary), ;?v (pre-primary or
TT TT
174 BIBLICAL COLOUR TERMS

indirect), and ftH 1 ? ( f . , 'moon'; indirect).

There is another difference between "p!? and the other two
TT
better-established terras (DTK and *lhB>): no Qal occurrences of
18
the former are attested. Excluding the possibility that both
features are a matter of random chance - and the non-existence
of a Qal theme in MH is an indication that this should be done -
the following conjectures are submitted. First, Berlin and Kay
have shown, and to my mind conclusively so, that terms denoting
'black1/'white1 usually evolve from 'dark'/'light' and are
19
created in most languages as a synchronous contrasted pair,
that is, the genesis of a colour notation involves at least a
primary dichotomy.
The basic morpho-semantic contrast in biblical Hebrew is
enhanced by being derived both from the differing root sequences
employed and from the different patterns. Secondly, while the
findings concerning the pa•"*O~ol pattern point to its relatively
wide employment as a nominal/adjectival/verb base, most lexemes
constructed upon the pa al pattern function as substantives, or
20
at least 'epithet nouns'. A cursory glance at the material
21
presented in Bauer-Leander will ascertain this. Cf. the
statement, 'Ofters ist die Adjective Bedeutung sekundar, so
22 23
in -p^22 ... pi1*'.23 Hence, especially with "tf> (and pV.p-p),
TT TT TT TT •••;
the borderline between noun and adjective is so fluid that the
only criteria for defining an actual occurrence is functional-
syntactical rather than formal.
Ill COLOUR TERMS IN MH
This page intentionally left blank
GENERAL 177

A. GENERAL

A.I As we have seen, the density of terms in biblical Hebrew is

far from satisfactory. So many gaps are left within the field
that existing terms are extended over a very wide range of
reference, to the extent that they may become semantically
overloaded and thus not coherent or specific enough. Other
terms are too sparsely documented to allow any definite
conclusions. In short, the field does not appear as a well-
developed segment of OT vocabulary.
1 2
The situation is different in regards to MH. ' Already in
the Mishna/Tosefta' and early Midrashim, and - to a greater
extent - in the Talmudim, we find a field that is enlarged in
scope and application (if not equally on all levels); much more
discriminating and detailed in reference; quantitatively much
more extensive; and in short, a far better reflection of realia
and of technological practices than the picture gleaned from OT
material. Undoubtedly this development is partly no development
per se, but the result of the preservation of terms dictated by
the types of material preserved mainly in Halachic literature.
On the other hand, the quantitative and qualitative condition of
the field as a whole must reflect a society which is technologi-
cally much more advanced than its counterpart in biblical times,
a state of affairs that must influence the language used by the
members of that society. Characteristically, it is the sector
of painting, dyeing, and so on which contains the greater amount
of new lexemes.
2. Our purpose in the present section is to demonstrate how
existing lexical resources, when retained, were adapted and/or
expanded by MH speakers in order to cope with changing conditions/
periods. 'The vocabulary of MH is composed of the following
elements: (1) Hebrew; (2) loan-words from Persian, Akkadian,
3
Greek, Latin, and Aramaic1. Here we are mainly interested in
the first category, namely, the application of various morpho-
semantic processes to established OT consonantal sequences and
the generation of new terms from that stock (whether belonging to
the colour field in biblical Hebrew or to another field).
Instances of the second category, although of great importance
for a study of the colour field whose focus is MH, will be
178 COLOUR TERMS IN MH

presented only insofar as they fill a basic gap for which

biblical Hebrew native stock was not suited, or where they tend
to supplant native terms.
The method adopted corresponds to the one employed in Part
II for the classification of the relevant lexemes according to
the structure of the field. At the beginning of each section
biblical Hebrew terms will be stated, followed by a list of
innovated MH lexemes related to them etymologically and/or
semantically. The lists are intended to be representative
rather than exhaustive, descriptive rather than analytical.
Greater emphasis has been accorded to the primary and secondary -
tertiary levels and to the section dealing with painting/dyeing
than to the sector of indirect colour allusions, for the former
categories constitute the core of the field, while the latter
4
lies on its boundaries. Therefore, no special treatment will
be accorded to these indirect significations, but they will be
mentioned together with related terms wherever this seems
appropriate. Finally, the chief semantically equivalent Jewish
Aramaic terms are supplied in most cases.
PRIMARY TERMS 179

B. PRIMARY TERMS

B.I DIN terms

T
1.1 Biblical Hebrew: DTK; verb forms: Qal, Hif., Hithp.
Range of reference: from 'brown* through 'red' to 'pink'.
Focus: blood colour?
Items whose colour is referred to: animals' hides, lentils,
blood, wine, human complexion.
1.2 MH. Qal, Hithp. are not attested, but Pi. and Hif. usages
take over the relevant references. In addition we find DTK
(biblical name of precious stone) = 'redness', as in the syntagm
'Q-TK tttKajV (Yer. Shabb. VIII, lOc) . Thus the Pi. formation -
in biblical language on the secondary level (ttfTKO = 'dyed red') -
has been expanded to signify 'cause redness', with a certain
loss of the 'dye' connotation. Otherwise, the Pi. form belongs
to the secondary level in MH as well.
Another new nominal is MttTK = 'red' colour (as in Num. R. 27).
The two patterns utilized for constructing DTK and MOTK are also
present in the "p^ field of the same level - 13(1)^ and M3^,
cf. below (2.2). Finally, the Hif. form ('become DTK') assumes
T
the specific signification of 'blush'.
The area of reference is now somewhat limited by the
appearance of new subordinate terms which are ultimately related
to DIN, but entered the lexis through an intermediate stage
(like the lexemes directly derived from dlfiTK - cf. infra, C.I);
by other terms on all levels which can be used to discriminate a
shade/degree of saturation formerly included in tiTK (Mp-iny H^YinT);
and by the growing prominence of 3h^. Cf., for instance, M. Nidd.
6:7 where, apart from D'TK and IhW, various comparative structures
are resorted to when specific definitions of shades on the axis
between those two are required.
1.3 In order to complete the picture - we are dealing with an
increasingly bilingual community - let us note the Aramaic
equivalents. These are i/P^b-derived: pn-lD; Qal, Pa., Af. verb
forms, and Kimpmo (parallel to MJTTK).
180 COLOUR TERMS IN MH

B.2 "}& terms

2.1 Biblical Hebrew; 1^; verb forms - Hif., Pi. (Dan. 11:35),
TT
Hithp.
Area of reference: 'bright', 'clear', 'hueless', 'light',
'pale', 'white'.
Focus: the colour of snow? wool?
Items whose colour is referred to: animals' fleece, woodbark,
teeth, TA-seed, hair, clothes, the appearance of diseased skin.
Transferred to denote 'purification', '(moral) purity',
'cleanliness' (verb forms).
2.2 MH. "p!? still dominates the field. Pi./Pu. are quite
TT
extensively employed, signifying 'whiten, bleach (cloth)' as well
as 'clean', 'polish', 'smelt (^.heat)'. This polysemic develop-
ment, although hardly attested in the OT apart from in Dan.,
should be inferred as the link to the transferred sense of
(spiritual and moral) 'purity' (Hif., Isa. 1:18, Ps. 51:9).
Hithp., Nithp., and Hif. (for the special reference of the Hif.
form + p'JD in the metaphorical sense of 'cause shame' - T. Bab.
B.M. 59a., etc. ) are present as well. Two new nominals -
1U1^ jtl-lin!? - refer directly to the quality of being "p^>.
'.* '. ~* T^"
Inevitably, it seems that a competition for this slot
between 1/155 and y"T>n derived lexemes (in biblical Hebrew in
Isa. 29:22, perhaps also in Isa. 19:9, cf. II., D.2.,
above; and in biblical Aramaic, Dan. 7:9) should have taken
place. However, the Aramaic term did not supplant the native
one. Instead, it entered the language, but underwent a shift: —}•
'clarify', 'explain' (Pi., Hithp.).
Scope and application: these are as wide as in biblical
language, which can be gleaned from a few idiomatic usages such
as in&M mt!> ('field of wheat'), D^Jit? (= 'white' garments, also
TT- ..• . T:
fruit), p^ - 'white' part of the eye; or from the definitions
of mmand lTNt!> in the Mishna (Neg. l:2f. - 'snow', 'whitewash'
of the temple, 'egg-shell', 'bleached wool') and in the Tosefta'
(Neg. 1:5 - 'wax'; ^Qllp/l^lttlP = 'yellow-green'). On the other
hand, Itflif and 1^!? make each other mutually exclusive on the
T TT
brilliance scale, inasmuch as 3fiS and Q'TK display the same
relationship in respect to hue.
2.3 Jewish Aramaic: as noted above, the superordinates are ^lin
derived: thus 11»n ,1(1)in; and Qal, Pa., Af. verb formations.
T
PRIMARY TERMS 181

B.3 intJJ terms

T
3.1 Biblical Hebrew: 1(l)nu; ;1imu? Verb formations - Qal.
Area of reference: 'dark-coloured', 'black'.
Items whose colour is referred to: human hair, human skin
(complexion), animals' skin.
3.2 MH. l'nt!> ; \7Tnw Qal not attested; Hif. (Hof.), causative
('blacken') as well as intrans. ('become black1). The contrast
parallel relationship with "p^ is probably reflected in the
idiom mnpn ^IBD 1^30 IVriBn = 'was (made) ashamed1; cf.
(D*3S) "pS^n (trans., intrans.) above (2.2), the difference in
deep structure between the two idioms notwithstanding.
"inty (and its related terms) is well established as a
superordinate of this sector in MH - its productive force is
even more apparent within the subordinate strata. In addition,
the many formations based on it in Jewish Aramaic - especially
verb forms (Qal, Pa., Af.) and secondary formations alongside
the native Aramaic DD1K and its derivatives - point to the
•
vitality of IhW in MH, so much so that although not unknown in
Aramaic - cf. Syriac Klin^U) ('coal') - its growing popularity
could be attributed to MH provenance. On the other hand, Aramaic
usage infiltrates MH; cf. MH 031K ('very dark', see below, C.3.2)
and Nithp. - dDKnj (+ D^S), 'be black(ened)', 'sunburnt' (cf.
Song. 1:5-6; Job 30:30), whereas no v'lHW Hithp. form is attested.
In addition, the scope of the term is segmented, and perhaps also
delineated, by the introduction of lexemes such as OlhB = 'dark,
black' - a case of morpho-semantic contamination of IhUJ and 0-in? -
which refers to the 'dark' signification of IhW (and is the
Aramaic rendering of the subordinate din in Gen. 30).
The 'dark' signification, as in biblical Hebrew, is denoted
also by \/1^n and ^YTp-derived lexemes. Especially instructive
are the syntagm 1»Jfl IVTpn (Gen. Rabb. 12), which is analogical/
interchangeable with 1>3Q IVflWfl (= 'become black"), and the MH
m^Tp = 'pot', apparently from its blackened, charred appearance.
On the other hand, there are plenty of instances where a 13^ - 1hU>
TT T
contrast denotes 'light colcured1/'dark coloured' antithesis,
not 'black/white'; so in the case of 'hair1 (M. Ned. 3:8), 'figs'
(Teru. 4:8), and 'fruit' in general (Maas. 1:3), to name but a
few.
182 COLOUR TERMS IN MH

3.3 Jewish Aramaic: the relevant y'tON derivatives are D31K;

NKD1K ;NJttO1N jxninDIX ;NBi:3*K = 'blackness', 'darkness'. The
verb series, however, is t/^f^-based: thus Qal, Pa., Af., in
addition to other, morphologically secondary and semantically
1
subordinate lexemes. /1WM and v'TTp are utilized as well.
PRIMARY TERMS 183

B.4 p"P terms.

4.1 Biblical Hebrew: pIV (hapax); ?iry>,
Area of reference: 'pale1, 'silvery', 'yellow-green'.
Items whose colour is referred to: vegetation (whether fresh
or deteriorating).
4.2 MH. pV is further established. If in biblical Hebrew it
is linked to vegetation only - although there are indications
that even in OT language it was better established (Vllfl p1pT>,
Ps. 68:14) - in MH its application is much more flexible (for
example Mish. Ed. 5:6, P1£i] 0*1 = 'yellowish blood'). An
additional lexeme on this level is the Hif. p^lfl, which appears
with D*^X2l = 'onions' and other types of vegetation and also
with d^S (M. Sot. 3:3, 4). From the occurrences in MH together
with those in the cognates (Aramaic Targums, Syriac and Mandaic)
it is clear that pll^. and its derivatives still denote the same
wide area as in biblical Hebrew, not necessarily 'green' in
every instance. Therefore, the 'pale', 'colourless' aspect is
common to plT and to 1D^. Thus we find not only a 1^!? - inw
T TT TT. T.
contrast concerning vegetation (3.2 above), but also a pi*. - in^.
contrast (M. Sheb. 6:3). In addition, a new term now occupies
the intermediate position which is partly covered by "p^ and
partly by pi*. The Hif. (and Nif.) denominative of q&3 means
'be(come) pale', 'colourless', while the Qal formation signifies
'feel ashamed'; cf. "PJS IJ^^M and Aramaic \/^03 with 'face' and
•hair'.
4.3 Jewish Aramaic: PIT, KplT is, among other things, the
equivalent of Heb. pT>. Verb forms - Af.
184 COLOUR TERMS IN MH

B.5 a?12 terms

T •
5.1 Biblical Hebrew: nnif
Area of reference: 'bright', 'gleaming', 'yellow'?
Items referred to: human hair.
5.2 MH. di\X is in the process of developing into a proper
primary term which is considerably wider in distribution, scope,
and flexibility than in biblical Hebrew. It is applied to 'hair',
but also to the 'colour of pigeons' feathers'. On the other
hand - as would be expected - its focus shifts from 'reddish-
yellow' (cf. Ezr. 8:27) to 'yellow' itself: hence the new term
in^ ,3nt>»3 both in MH and in Jewish Aramaic (cf. II, B. 5.3
above, pp. 103-4 ).
The process is exemplified by the productive force the
lexeme now exhibits: there are a Qal verb form (+ 0*33, cf.
T. Bab. Hull. 7b), and a Hif. one (+ nwru, M. Midd. 2:3) which
has already been attested through the Hof. form in biblical
Hebrew. These verb forms should not be confused with their
homonym, a derivative of /Ihif.1 A nominal formation, again
with 'feathers', is ^rPS.
PRIMARY TERMS 185

B.6 Others: a term for 'Blue'.

MH lacks a primary term for 'blue1, as does biblical
Hebrew. j/tTD related lexemes, including ^IfO, are still
transparently connected to their base - they signify a 'dark
colour' notion which lies within the IhW sector.
T
This interpretation is in contrast to Rashi's comment on
K^hlDD (T. Bab. Hull. 47b), which he explains as 'sky-blue'.
On the other hand, it appears that the shade of fl^Dfl more
popular in MH times and later was quite pale; this can be gleaned
from a passage about fi^fl which reads, iimi D''! tP^ finn nt>3Jin
1'DO in*!? finn ypll JPpl!? (T. Bab. Sot. 17a; Hull. 89a, and
elsewhere).
Therefore, in the absence of an explicit term referring
to 'sky colour, blue', tlton is used as a substitute of sorts;
cf. the contrast ntOH - 1^^T T
and Jl^Dn - »frlD
*
(= 'blue/green',
'dark') in the following passage: ?lPintyi WAV M "PUP TWNn
•>niDt> nbDn im mix iry»!?K 'i .i^W nton vn i»D>wn (M. Ber. 1:2
186 OUR TERMS IN MH

C. SECONDARY AND TERTIARY TERMS

C. 1 Under DTK
T
1.1 Biblical Hebrew: VOK* ,pie;* ,DrrK»*
TT!
•»jimK ,DimK •}& ,mWDn ,^»>Dn ,vinn ,inian*
Scope: various aspects of chromaticity/saturation
governed by OIK.
1.2 MH. OiniK (cf. II, C.I.I, p. 106 above) is interpreted
either as 'reddish1, 'pink1 (Mishnah) or 'deep red1 (Talmud); the
point of contention, then, is the saturation, not the fact that
it is an DTK-governed (Q^ttTTK^u; OiniK) and not only an DTK-derived
lexeme. Another partly reduplicated formation is flin^nTK,,
signifying the quality of being D*TK or DTK-like. dJTKtt retains
T
the same reference ('made' or 'dyed into DTK'), although the
scope of the parent-formation (Pi.) is expanded (cf. above, III,
B.I, p. 179). plto* does not feature although it generates MH
T 1
VpIO (Qal, Pi.) = 'paint into bright red'; later, though, we
can see how the meaning is expanded to denote 'paint1 in general,
and so also the nominal piD (cf. below, D.3.2). It seems that
r r
~
Albeck is right in dissociating pit) from the Aramaic KIp'O
2
TT -r x •
the latter and its denominative MH verb forms (again Qal, Pi.)
retain the sense of 'paint with Siqra' or 'paint eyes' (utilizing
the hapax in Isa. 3:16, with a shift in meaning). This could
perhaps be a case of a merger of two distinct root-sequences, or
at least a convergence of polysemic elements in Aramaic, and
therefore in MH as well.
Of the next level "OlttlK survives, as is attested by the
personal name TD31K too. QTQTK "\tf falls out of use - M. Neg.
1:2 speaks about "POaw/A^eOW .1-1119,1, that is, the D*TO1K element
mixed with 1^, exemplified by 'snow' and 'white-wash'. f'yf^yn
disappears; the same fate befalls y.ion (but cf. Jewish Aram.
4 '
KnsinT), 'red' garments ) and VttK* of the secondary level (but
cf. Kifl3-1K = 'steak') - an indication in, each case of the
Tt
supportive/esoteric nature of the biblical lexemes. IBinn* is
elided completely - /IBM forms serve for denoting "burn, heat',
but not in a re-duplicated formation. A trace of colour
reference is opaquely inherent in littfl and its derivatives, and
in 1£1K. (= 'ruby', cf. the Arabic cognate). Otherwise, the
lack of distinctive terms referring to segments of the DTK sector
SECONDARY AND TERTIARY TERMS 187

is sometimes sourly felt. Then equations with substances of

typical 'red' shades are evoked. Particularly instructive is
a passage in the Mishna which deals with the appearance of
menstrual blood (Nidd. 2:6-7):
nmx wn3i &3-13 111731 line/ni tnxn - niwo tPNno tmn.fronn
...^s iv»:a »n»n3i in!?n *n»a3 qN dnniK >K»B IPS .A-irn3i
T
• T TT . - ' " '' > i ~t"
nntr .HBO - *pn pay .mns - 11 nty .nsnn ona - DTK inn*
•'«• T •.".•; T T T T

nypio - nmx W031 .inw TI-DD - 03-13 11J731 .lino - pn

in v THKI o'n D^p^n •JJK; - >irn3i 0 Q»a q^imi Qi3h rr>3
.•>Jiiu;n i"M
(and see the Talmudic discussion of the same passage, Bab. Nidd.
19a, ff.). An additional simile within the same sphere is
MmiM nJmiBO (Lev. Rabb. 12). Another term is dip: in biblical
T ' V
Hebrew it denotes a type of gold (II, F.3.2 above, p. 167J, in
5
MH '(red) stain' (of menstrual blood) —> 'stain1 (of any kind).
In a series relating to human skin colour we find 11(T>>
(M. Bekh. 7:6; T. Bab. Bekh. 45b.) = 'flame-coloured', of face
or skin, in conjunction with ">W13 = 'dark-skinned' and
1|7^VP\l5 = 'white-skinned', in a list cataloguing firstborn
priests who cannot officiate because of a physical defect, in
this case the extreme colour of the skin. Finally, an indirect
colour allusion: the planet Mars is named tPTNtt (that is, the
Hif. part, form is utilized for constructing a" nominal lexeme).
1.3 Jewish Aramaic. Kpi»(>)0 is the equivalent of diaiK.
Other terms are ^IplfiD and pnpOO (Sam. Aramaic). For renderings
of these and other biblical terms, see II, C and D above, p.106ff.
Additional \/pnO and (/dTK related lexemes are well documented as
well.7
188 COLOUR TERMS IN MH

C.2 Under -p*

TT
2.1 Biblical Hebrew: 1112* ,1T* ,h£
/Wf
Scope: 'light coloured', 'pale', 'bright' - mainly luminosity
and brightness (not chromaticity) aspects.
2.2 MH. Both n* and ,1T* denote aspects of light ('clearness',
'luminosity'). *Vn^* features as well. Nevertheless, all three
remain on the fringe of the sector, where the latter merges with
the light/darkness field.
An additional MH lexeme is fl^-lia!? ('whiteness'), another
member in a series that includes JPttlOTK and JPl.l'Tnw as well.
Another lexeme - I-IS^ - occupies a slot in this sector as
well as in the field of cleanliness and purification (cf. the Pi.
and Hithp. usage), /lin, on the other hand, is only productive
on the level of names containing colour allusions (1-11in =
'leucoma', 'cataract').
Let us return to the dimension of brightness and gloss,
which is included in "p!?. A further development is demonstrated
TT
by the utilization of i/\>t\5 ,^"litt (biblical Hebrew: Lev. 13-14,
pita jfllita; and once ITia, Job 37:21) in various formations,
especially verbs - thus piTi:i ('albino'?), and the Hif. forms of
both roots.
M. Neg. 1:1-2, already quoted above, contains elements
1
similar to biblical ones - ' "p^ TT
as 'snow1, 'wool', 'milk' -
but also two new specifications: 'whitewash', and 'egg-shell'.
Another passage reads Jl^>1ttD "p^ = 'the colour of pearls'
(T. Bab. Yoma 75a, in a discussion of Ex. 16:31, 1^ TA jrm).
For skin-colour of humans, apart from the above mentioned
IpaVlPII^ (a loan-word from Greek, C.I.2 above), we have 'J(1)»1Ji
(= 'German') as a general denotation for 'whiteskinned man' (op
>K;l!), M. Neg. 2:1). The same passage explicitly refers to "pt?
and Itliy as the two extreme poles of the colour spectrum between
which the other colours should be placed:
bNyoup "n ,niy ^Bi^n nrorn nnD 'Jini>a IPN-I:J nry rnna
"i .Q^ju 1 '^ N^K d>j^ Ntn tp-nnt2> NX..!«au;» "on imK
Jinihw-jn'm Q»is ono d-^ttao &>-i»^ w -IQIK Ki>i?y
nKin vm^n is^pm »aia»n do K^^a .imjii^i niiit>i
...•>313>13
SECONDARY AND TERTIARY TERMS 189

A comparison of the observations quoted in this Mishna -

and in Neg. in general - to Lev. 13 shows how much more articulate,
discriminating and colour conscious is the former (Cf. further
the discussion in the Talmud, Sheb. 6a f.)
190 COLOUR TERMS IN MH

C.3 Under ine>

3.1 Biblical Hebrew: irnrw* ,tMh
VWf ,nro
Scope: 'dull', 'dark' to 'black'.
3.2 MH. din falls out of use. I'tnrw* is used as a base
generating the nominal IMininw ('blackness', 'darkness' =
'sinfulness1, Song. Rabb. to Song. 1:6). tl'Tlinw completes the
basic series n>:im£ ,JV1KP»TK. lin*W, following its Aramaic
equivalent, denotes 'coal' (Cf. biblical Hebrew 11ITO, according
to Gradwohl, II, B.3 above, p. 181 ff.) in opposition to
ri^Aia = Vrt (Ex. Rabb., 42). njl2> = 'dull', 'dim', 'dark' is
defined by its opposition to i\ty mfO (cf. M. Neg. 2:1 quoted
above). A parallel for ... ^fl IVtltfiUT. Yer. Hag. II. 77b),
using v/IWH, is to be found in the Qumran Hymns, 1QH 5:32,
JltJflK!? *i£) TIW3 1U>rPl. Further discrimination of shades of
'blackness' and their differentiation are presented through new
terms: ^QTm - mnnriB ,dinK> jDDlK in opposition to 11(12 (Tal. Ber.
36b); Oman (<n*D, Gen. Rabb. 36); and t>ih3 (cf. Ezek. 23:40).
Intermediate positions ('grey'?) between *p!> and inu> are
TT f

apparently denoted by q-lHO or q-1hK> (opp. "p^ - M. Hull. 11:2

Tal. Hull. ISSb1) for 'sheep', n^d ,Ma»0 (Tos. Neg. 1:4, cf.
biblical language) for 'hair'. A dark person is a >EJ13 (M. Neg
2:1), and the lexeme functions as a substitute for llflW in
syntagms such as '>K;i3ri ATlIlK (M. Sukk. 3:6) and •>BM3i] T»», and
is a member in the series which includes llfPA ; If^VlPll^
(M. Bekh. 7:6); and rPQO, Tal. Bekh. 45b (from 'pitcher', whose
~ 'T 2
characteristic colour was dark ). However, most distinctions are
defined by means of a simile which evokes a typical colour
quality. „Thus limyT is compared to rQjy
T-I ;,nn
• — ,n3T
,• •• ,Tliy
•• ,MYTP.
-j-,. ',
Hit! ,1'^. This practice is an indication that existing
terms, although greater in number and more specific than those
of biblical Hebrew, are still inadequate, and hence the profusion
of comparisons.
3.3 Jewish Aramaic. To the terms mentioned above we should add
/Dfi; -} nine (biblical Hebrew Qilll), d^hW, K»n(»)K; - all
referring to 'dark', 'black', perhaps a shade lighter than
lilTO proper. Km'Vnnw (Ntnnw) looks like a Hebraism. Other
\/iniy related lexemes contain colour allusions ('NlirPiy', 'coal'),
T
but not 'colour' notions per se.
SECONDARY AND TERTIARY TERMS 191

C.4 Under plT

T
4.1 Biblical Hebrew: pip*)1*
Scope: 'pale', 'yellowish', 'greenish'.
4.2 MH. p1p*T> is used for defining the same 'diseases' as in
Lev. 14, but also for complexion - nfi'fl npllpV 1Ht)K (Tal. Meg.
13a). Otherwise no new lexemes are generated from y'PV on the
directly subordinate level. There are further developments in
other parts of the field: flp)<lT> - nipT>. (name of a vegetable and
a fish) and nipT> ('vegetables') are added to the old pT» and
-rj TT
np-p. T"
An unrelated term is ItltTK^ - 'greenish', 'dark green' - of
what looks like Arabic provenance (cf. "TttllN) . Other descriptions
are those utilizing similes or derived from analogies: TfOD PIT
- and in M. Ber. 1:2 TlID is put in juxtaposition to iftSft ;
AliriKD PIT? (Tal. Nidd. 19b, a denominative H i f . ) ; fllJTCO plV
(T. Yer. Sukk. Ill, 53d); and .more.1
2 .
Herszberg lists 02n3-derived terms within this section.
This is not accurate: the idiom 03*13 IK/1 p-jO (M. Kel. 15:2)
seems to indicate 'paint', 'varnish' in general, not only with
pID or 0313. The denominative Pi. and Nithp. formations + d'3D
either signify 'become pale' - an equivalent of ,cpODM + O^O
TO^M + Q'ia - or figuratively 'was sorry, angry', without any
colour connotation. An occurrence in Song. Rabb., TO£) IKG'OriJ
(to Song. 1:6), discusses sun-tanned complexion - that is, bronze
colour, elsewhere referred to by Hflif. D313 "HP is a specification
T V*" V V
of DTK (C.I.2 above, p. 187). Therefore, I think that D313,
belongs to the section dealing with paints and dyes. Admittedly,
though, some of its usages belong to the area which is common to
"p^ .PIT* and ^h^ - that 'pale', 'colourless' - luminous, not
chromatic - quality to which each of them might refer within a
given context.
192 COLOUR TERMS IN MH

C.5 Under sn*

5.1 Biblical Hebrew: SHiM
T ; '•.
Scope: 'gold-coloured1, 'bright1, 'shining'.
5.2 MH. The equation 1MT = XUto 'of gold colour/sheen' - first
found in Ezr. 8:27 in a description of copper vessels - can be
found here as well. Thus M. Midd. 2:3; 1J1H!>J dU> 1'hB onywn ^D
T; -
,:i»n2tt
. .-
iiusnjiy
' . .. •
T
-»jon onaiK an....^ ins nwyiw
"
>:i3n...:irw-*>
TT-V
JTprtf
Consequently, a Hif. formation was constructed from ^nt. However,
this seems to indicate a specification of iPH^ri rather than an
equivalent. We read in the Mishna (Hull. 1:5): !?1D£J "mittl leo
o^ioa nTii ntn sirpsn n!?^nn .T'lin^ ^vos njT> -0:1:3 iu;3 n^i> *jni
But, on the other hand, in the Sifra (Lev. 1:14) and the Talmud
(Hull. 22b); .i^nnira 7-p-iKo •pTinn •»nni»K)3
.i±»n^iya TV^IOO rm* ">n ^nn^Kni
The biblical ^rutfl (Ps. 104:15), in»W3 Q>J£3 !?»n^n!? = 'make
shining, brighter' reappears in MH, although rarely, with the
same idiomatic meaning. Thus we have tinfl^ ti'Jfl ^Tl^Q (Pesik.
R. 20) = 'lights', 'brightens like the sun'. Again, the basis
for the metaphor lies close to the referent of 3My and its
derivatives. Similarly, new lexemes are derived from \/in^ \ cf.
Hif. /in^ - 'brighten', 'make glossy' (Lev. Rabb. 5).
MAN-MADE COLOURS 193

D. MAN-MADE 'COLOURS': DYES, PIGMENTS, DYEDMATERIALS,

MAKE-UP, WRITINGMATERIALS

D.I Terms for painting, dyeing and colouring.

1.0 In contradistinction to biblical Hebrew, which does not
possess even one term denoting the activity of painting/dyeing,
MH has quite a number of them. These can be divided into the
following groups:
1. Terms signifying the general notion of 'shade'
(chromaticity) or 'dye' (with no specific reference to any one
given hue) and/or the activity of dyeing/colouring.
2. Terms signifying a particular notion of a certain
shade/hue, but also serving, in some contexts, as 1. above.
Within and across these two categories a further distinction
should be made between two sub-groups: one referring to various
activities and concepts connected with textile dyeing, and a
second referring to other activities/concepts (cosmetics and so
on).
1.1 General terms
The semantic void of biblical Hebrew is filled by borrowing
either from other languages - y^X ,y$X etc. from Aramaic; TU , Tll>
from Persian; 0113 ,d11!3K from Greekj or from another field in
Hebrew itself - DO?
1.1.2 )/y2X - originally 'wet, dip, immerse' (biblical Aramaic:
Dan. 4:12, 20, 22, 30; 5.21. Pa. and Hithp.) -^ 'dye' is the
dominant base in the sector. The Qal refers to the activity of
textile dyeing: the pass. part, signifies a 'dyed' condition (for
material, Num. Rabb. 2; hands - M. Neg. 4:7) and by extension 'a
hypocrite' . The activity itself is called hy^if, the professional
who engages in it is yZX, and 'dyed/colourful clothes' are
T—

V}i)m (*T>:i) . yi* denotes 'dyed material', but also 'dye'

(M. Nidd. 9:6, in opposition to DOS = 'stain') and 'coloured
appearance' (Num. Rabb. 2). JOli? is more specialized - it
denotes 'material for dyeing' and especially 'vegetable dye'
(M. Sheb. 7:1 ff.; see below, 4.1).
1.1.3 yu , ym = 'shade, hue' (Tal. Erub. 53b., Nidd. 24b; very
frequent) is of Persian origin. Q1"O ,D11DK is a loan from
1
Greek/Latin chroma.. Within some contexts OIID refers specifi-
cally to 'paleness' or 'greenness', but its general denotation
194 COLOUR TERMS IN MH

as an equivalent of H> , IIIA (although it is less frequent) is

beyond question.
The last term in this series is dO , 11M3D - in biblical
T i- 2
Hebrew 'spice used for preparing incense' (in P and Ch.), an^ in
MH - 'drug', 'medicine', but also 'colouring agent, dye' (cf.
below under 'Writing Materials').
To summarize: three out of the four new terms in the series
yiX , lU/lIU ,D11D/Dn^K, and dO/lttttO are loans (from Aramaic,
Persian, and Greek/Latin respectively), and one a specialization
of a native term. QO / 10)30 and y^X signify 'dye' too. In
addition, the biblical usage of 'MK'UD1, "pyip1 is resorted to
many times.
1.1.4 Biblical y"m ,iYUX is the base from which MH v^S Pi.
and 1*5f = 'someone who engages in the plastic arts' are derived.
As such 1*JJ and 1*iJ refer to the activity of painting and
drawing as well - cf. Gen. Rabb. 1, where 1QBD appears in the
same context.
1.2 Another category is that of terms which are originally
restricted in reference to a certain type of colouring activity,
mainly make-up, and/or to a certain shade/pigment used within
that area; from these the scope is widened to denote the process
in general, and even to 'cover with paint'. Thus plO = a type
of 'reddish1 make-up for the face and the verb formations
related to it have come to denote 'paint', and not only the
face, especially in the syntagm 03151 pID (M. Kel. 15:2 and
more). Similarly, d!)"13 - from d313, basically 'pale', 'saffron
" '* ' 3
coloured' - became 'paint', 'polish'. Within the same sphere,
4
/Opfl Qal, from op*3 = 'rouge', signifies 'rouge1, 'make-up the
face1; but its derived form - dplfl / OU^fl, with a dissimilatory
/1/i also denotes 'paint' (the body, and not with Op'3 only).
Viewing the latter verb as an enlargement of y'Opfl seems more
c
convincing than to derive it from /plfl, especially in the light
of the parallel developments gMX = 'hypocrite' (< y^) - D-lp^a,
0.131£3, ='deception', 'deceptive appearance' (-=C Dplfl)0 Finally,
we have tPO Pi., chiefly in the series V;?3 - 1^3 - T^D,
meaning 'to paint', 'give a colourful appearance'. Jastrow
derives it from biblical H*3 = 'soot'. Unless we assume that
the verb has undergone the expansion process described above, it
should be considered a derivative of a homonymic I"P£J root; cf.
MAN-MADE COLOURS 195

6
Moresheth for the latter view and for quotations. There is
no doubt, however, concerning v^tJ^S Pi./Pu., Nif., Nithp., all
denominatives from QMS. These refer to various aspects of
'blackening', 'becoming black', and not necessarily through
7
touching charcoal (as in Tos., Maas. Shen. 5:13).
196 COLOUR TERMS IN MH

D.2 Colourful Textiles

(Cf. Herszberg, A1K.1, p. 267f£.)
2.0 In biblical Hebrew primary terms are not customarily used
for denoting 'dyed shade', 'dyed/colourful cloth1. When they
are, it is within sections that are late (post-exilic):
IBHaX^) tJTTK ylttt (Isa. 63:2) and CP^!? d'T^ (Qoh. 9:8) are
the only instances. In MH, though, we find "TOi? that is not only
1^!?, but linta and OITX (M. Bab. Kara. 9:4) as well. Garments can
TT . T T
be Online; or tPJ^, thus by-passing the need for specific terms,
while 'red' dyed garments are defined and discriminated by
existing biblical terms (IQA'IK ,'3Kp or their equivalents (JPIlMt).
2.1 inAIN - denoting the cloth itself and its 'dyed' property -
is far from specific: in the Temple Scroll we encounter the
specification D1TK in.X'IK within a text based on Exodus 25, where
1QA1N only is mentioned. An additional designation is
il/Nl.lfll-lfl-11*113113, a loan from Greek, mainly a reference to
""" '' garments' in the Midrashic literature. 2 The Aramaic
'purple
TU1K is of minor importance in MH, for the standard biblical form
is retained.
">1W , *3t!> riy^in.iiy^in = 'scarlet, crimson' are used
sparingly. The chief term is the Aramaic JT>liflT, a culture word
with cognates in Akk. (inzah(u)retu) and in Syriac, but with no
3
satisfactory etymology.
2.2 n!on (Aram. K^Ofl ,Kfltori) denotes 'dyed wool of a certain
colour1, as in biblical Hebrew. In addition, it serves as a
substitute of sorts for the empty 'blue' slot (cf. B.6 above,
p. 185). The colour reference is quite extensive - from 'sky/
lapis lazuli blue', to shades close to 'green' ('HID ^ fitOfl)
4
and darkest violet/purple.
2.3 'Colourless'—'white' kinds of cloth in biblical Hebrew
are W = V-13, Ofl1D,1in (the latter two in Esth. only), T!l ^"m,
and once D'33!? D'HAS (II, B.2.6 above, p. 90) - a forerunner of
MH usage. In MH 03"O hardly features - instead ISA 10£ is
used. 5 Tin, unless reflected in the Aramaic KT>K (with a loss
of the original /fl/, although Jastrow relates the latter to
i/IIN), disappears too.
V-13 supplants W. When the writer of the archaizing War
Scroll from Qumran uses WW, as is dictated by his biblical source
MAN-MADE COLOURS 197

material, he adds in!? as an explanatory gloss (II, E.2.3.1 above,

pp. 148-149).
The more frequent designation is simply 0*3^ D*"U2l or
H'b *!7i>/*TA3 - the primary term is expanded to denote 'man made1
appearance, as in p^ 3"l)> (// y)3X 'flW). An equation of
•p^ 'ma = V'13 = VK3 seems to be in order. Thus M. Yoma 3;7
T T i •; • "
supplies us, as specifications of the high priest's "pb 'TA!l,
with the names for two types of V-12 : VD-l^O, from Pelusium in
Egypt; and "plTJfl from India.
2.4 No specific terms denoting textiles of additional colours
+ the dyeing agent used for producing them are extant, although
7
information concerning dyeing processes is readily available.
In contradistinction to the linguistic-literary lack,
archaeological finds from the Bar-Cochba period vividly
illustrate the realia that is only inadequately reflected in
8
linguistic usage.
198 COLOUR TERMS IN MH

D.3 COSMETICS
3.1 In biblical Hebrew only terms referring to eye make-up are
preserved. The earlier ,1'IS is probably the equivalent of
bhia ,b1M3. Therefore, the development of the tTD sequence -
- T . . . .
MH. Qal, blhS; Aramaic Qal Ithp.; *13, *13, 'pftnS) O Hebrew 103);
T —• - ••*«. •;
and birGQ - while "113 displays no productive potential, is hardly
surprising. No change is to be detected in the extra-linguistic
reference, which is quite problematic.
KIp'O and its denominative f'lpO (Qal, but especially Pi.)
also refer to eye make-up, like the biblical hapax of a different
sense (Isa. 3:16). Therefore I find it difficult to accept the
equation /p"10 = /IpD (through metathesis), or an interpretation of
K"lp>D as always denoting a 'red' paint. Perhaps the Qal should
be differentiated from the Pi., the latter being a secondary
development based on the biblical occurrence?
3.2 No specialized biblical lexeme refers to face make-up,
whereas in MH we have op*S (and its denom. Qal), DDID/opIS and
1 2
plo (£p"lK;). Op»3 (from Greek, Latin ) is 'rouge1 - cf. Tal. Bab.
Shabb. 64b, olpflll K^l birOJl tftv. The secondary formations
Dma/Opia/OlSVa/OlpVQ signify 'painting' (face), 'dyeing' (hair)
—> 'deception'.
P"l0 and its Pi. denominative, often appearing with 03*13, is
"'"'" 3 r* '
according to Herszberg the equivalent of biblical Hebrew 1Bt».
4
Its etymology - biblical plW*, Akk. sarku = 'dark blood'
points to a positively 'red' effect of some kind. It is used as
dyeing agent for general purposes, but also as a cosmetic for the
face (Tal. Sanh. 14a, among others).
3.3 T>0 = (lime) was apparently used as a hair preparation
(cf. M. Shabb. 8:4).
MAN-MADE COLOURS 199

D.4 DYEING AGENTS AND PIGMENT SOURCES

4.1 y:m , pi. D^jmx/'pym is the general term applicable

to the notion of 'vegetable dye': M. Sheb. 7:1 ff. lists
vegetation bases used for dyeing under the heading V^I^U V-W-
Unlike WIN,n!on ,'JW / JPTinT, our terms do not signify
'coloured appearance' of a certain textile, but the name of a
vegetable or fruit base used for extracting a required dye.
4.2 On the basis of the sources Herszberg supplies a full list
1
of dyeing agents. The list is reproduced below, together with
some remarks concerning linguistic points or aspects of realia
related to various items.
D^OD'K - Wade, for producing dark to blue appearance,
'""" 2
from the Greek.
3
HK'ia - Alizarin, dyer's madder; for red.
d>T!AK ''D't'p - Nut shells and pomegranate peels; for
a^imi ,, ,
yellow-red.
4
nflDTi - Cyclamine; ' red.
T
'• ' S
fl/KXIp - Madder; red.
KJ(y)t?ifi - A tree; 6 from yVia = scarlet,
crimson.
7
K/M35 - Arabic and Latin cognates, originally from
India; a gum compound used for achieving a
bright red effect,
ynn - Safflower (cf. Syriac, Arabic, Phoen., Akk
Heb. Vl*in); for yellow to red and pink.
•p2*^n .Vyy1? - Signification unclear.8
g
I^K-K^p. - Indigo; from Greek; for producing blue to
violet cloth.
tb-13 - Saffron (cf. II, E.4.1 above, p. 153) u
in combination dyes; hence probable Q31.3, in
an expanded sense = 'to dye, paint, stain'.
A1K - Red berry.10
1

Most agents are used to achieve dyes which may produce red,
yellow, or blue; linguistically, the terms themselves supply no
clue as to their referents. Finally, if we are to return to
Yadin's finds, and if these are to be taken as indicative, some
agents - namely indigo, alizarin, saffron, and carminic acid -
11
were much more popular than others.
200 COLOUR TERMS IN MH

D.5 WRITING MATERIALS

The relevant biblical terms are 1>"T ,1WU> and perhaps TW

r r
r —T •/••'
(II, E.5, p. 154). MH sources list m ;00 (D.1.1.3,pp. 193-194);
K1R>0 ; 01 nip (from Greek - a kind of gum); and BilUWp/OlJOp^i?
T . -. - ; - ; -: -
(Greek, = sulphate of copper) - all in M. Shabb. 12:4. Other
materials are nil"! (M. Nidd. 2:7, HIDD 11 hW, and explained in
Tal. Nidd. 20a. The Aramaic idiom is KIrt->TT Kniltl) J lilWpin^W
- 'coal, soot' (Tal. Gitt. 19a); 'J>'0 ,1^>n, (from Greek, 'black',
'dark pigment1, Gen. Rabb. 1); and 5lhD (D.3.1, p. 198). It
seems that 1**T = 'ink' is the superordinate term in this sector;
therefore it does not refer to any specific colour quality. The
other terms do represent colour qualities within the dTK/lflKJ
ranges. The number of loans from Greek here is three - 1^'D.j
Olttlp, DinJp^p/DlJlJp3p - which is quite surprising. This perhaps
represents the borrowing of practices together with their names.
MAN-MADE COLOURS 201

D.6 WHITEWASHING

TO typifies a certain "P^. appearance in M. Neg. 1:1,

^3^Mh TD3. It refers not only to the natural mineral ('lime'),
but also to its processed appearance ('plaster') and colour.
Thus Qal, Pi., Pu. = 'to whitewash', 'plaster'; *J*'*§.= ' 1 imeburner
'plasterer' .
On the other hand, TA (also 'lime') is not as productive in
this area. Noteworthy is VA with an opposing reference, as in
1
'"PAM m^U1 (Tal. Bets. 15a) = 'inkstone',, 'sulphate of iron1.
Rashi comments on Ex. 16:14, l n e yiX T»03 Kim (based on Tal.
Hull. 88b., "p:nTm Tun) . As such, this "m merits a separate
lexical entry as a homonym of V> = 'lime'.
202 COLOUR TERMS IN MH

E. SUMMARY

1. Whereas no new primary terms are created to fill obvious

gaps in biblical Hebrew (for 'blue', 'brown', 'pink', 'orange',
and 'grey'), existing primary terms continue to function as such.
P")^. and 3MS, whose status within this stratum has previously
been weak or not attested, emerge as more securely established.
Within the existing boundaries there appear new constructions,
either of verbs or of nominals (the series ft 4 3 3!? ,H-1131K), which
are generated by the old terms.

2. Further specification is achieved by the introduction of new

subordinate lexemes. These are derived from native stock which
features in OT language, but built upon new patterns (nitt'tt'TK);
brought in from neighbouring fields (^jl^il 'T*0?'?)' or created
through a process of morpho-semantic contamination
dn^^V^ritf + v/^n^ ; Qr\V <Y™>> + /dirf ) . At the same time
some terms which were rare, isolated, or idiolectical within
biblical Hebrew are dropped out of usage C^^pn ,15*1011*) or
returned to their original slot outside our field ("]"(_* ,H"tf).
Therefore, it seems that despite the accidental manner in which
biblical language has been preserved, statistical/distributional
factors should not be disregarded out of hand.
On the secondary/tertiary levels, as on the preceding one,
loan words form a small minority.

3. In the absence of monolexemic or unilexemic terms, analogies

are resorted to - the shade/brightness aspect referred to is
conjured up by means of a comparison, as in biblical Hebrew,
albeit with much greater detail and precision.

4. The greatest development is to be found in the section

covering man-made 'colours' (D. above), that is, the linguistic
sector which reflects technology and material culture best.
Whereas in biblical Hebrew no term for dye/paint or dyeing/
painting is extant, MH employs quite a number of them, both
general and specific (D.I). Although some OT terms are rejected,
this is usually done in favour of a modern equivalent (fPTinT for
»aw; plO for "lEW,
r
although p'lb* itself disappears) . The
*T TT ~~ T '
profusion of data concerning dyeing/painting (D.I, 2, 4), make-up
(D.3), and writing materials (D.5), together with the fact that
SUMMARY 203

quite a few terms are borrowed - mainly from Aramaic and from
Greek/Latin - probably demonstrate not only a growing awareness
of and preoccupation with colour phenomena, but also a growing
need and wish to express fine distinctions in a more precise
manner. The language must be a reflection of superior technology,
brought about by changed conditions and the passage of time. The
process of linguistic expansion in this field is quite noteworthy
even when MH is isolated from Aramaic, as has been done here; and
even though throughout the life-span of MH its co-existence with
Aramaic gave rise to bi-lingual interference, as much as to
supplements of missing terms from Aramaic lexical resources into
MH (and, to a lesser extent, vice versa).
This page intentionally left blank
IV SUMMARY
This page intentionally left blank
 207

SUMMARY

1. The chief factors which govern the psycho-physical sensation

of colour are chromaticity, saturation, and brilliance. All
three aspects are important for defining the reference of a given
colour term, and the coining of monolexemic terms in biblical
language is less hue-motivated than, for instance, in modern
European languages.

2. The colour field in OT Hebrew - or in any other language, be

it ancient or contemporary, 'primitive' or 'modern' - must be
discussed against the backdrop of comparative scientific work,
including that of anthropology and anthropological linguistics,
and not treated as an isolated phenomenon.

3. In biblical Hebrew the field is structured and its

components are arranged in the following manner:
a. Basic colour terms - dlK ,1^ ,°inv .p'lV , ?p"P ,?3iniJ +
T TT T T •;•,' T
basic verb formations directly derived from them (but excluding
participial forms).

b. Secondary terms - .IT* ,hS ,in>?*; dTOTK ,tttKtt ,V»K* ,p1K;*;

• "~ — T — : — : TT; T T
-irnhei* ,o.in; Pji?v; srun.
c. Tertiary terms - »aimK , ^^on and n-l^DQ ,V-1Bn,
-m-ian* + D^O; o*renK •}&; /viTT Qai; v/iwn Qai, nns* ,nn3.
d. Terms for pigments, dyes, and paints - n^Dti, 1»A1K = 11A1K,
>jw, y"y-\f\, >i» n#in, ^>»T»D, MKIS, -TOW; y-ia, m, i3n, OQID, m*,
T>W; ^5Fo Qai, T»Q, 1'T; D-J1)?? 10£»?

Apart from these classes, two additional categories were

found to be peripheral to the field.
a. Proper names and names for various objects/concepts
which are related to colour terms either etymologically,
phonetically, or by way of association; and

b. Terms for speckles, stains, and other 'coloured' areas -

pna ,rnn£ ?q-it>y* ,TTHyp/i^to*, T«a* ."rb^ ,Tpy* ,K-I!?O ,inj .mirmn*
4. Conclusions - biblical Hebrew. The basic term category is
not well developed. It fits stage III or IV of the evolutionary
scale described by Berlin-Kay. The DTK segment is the best
developed. Surprisingly, the inw segment is less advanced than
DTK
T and 13^ before the exilic period. A similar state of
TT
208 SUMMARY

affairs appears on the secondary/tertiary levels. Within the

sector comprising terms for pigments, dyes, and so on most data
refer to textiles. Here too DTK-governed terms form a majority,
while "p!? - governed terms come second.
As for peripheral terms, proper names are difficult to tie
up with colour terms - often they are generated by homonymous,
not identical, sequences that are semantically opaque. It is
easier, and perhaps more meaningful, to establish connections
that may be called folk etymology or literary associations.
Other terms which contain indirect colour allusions include some
relatively clear-cut cases (names for precious stones, metals),
although the inter-relationship is not always easy to define.
5. In MH, as is to be expected, the field becomes better
defined. This is the result of its expansion and the demarcation
of existing segments by the addition of new terms. On the other
hand, no new segments are actually carved for basic 'blue', 'grey'
'orange', or others. The development takes place within the
previously defined segments.
The new terms are either of biblical Hebrew provenance, or
else they are loans from Persian (as in late biblical Hebrew),
Greek, and Latin, but mainly from Aramaic. The best-developed
sector is that of pigments, dyes, and painting. Two possible
reasons for this might be either that the language adequately
reflects the superior state of technology, or that the type of
literature preserved simply contains more information pertinent
to our study. If the latter reason is the more weighty of the
two - which cannot be ascertained - then the real advancement is
less spectacular than it appears to be. Finally, bi-lingualism
(Hebrew-Aramaic), which accounts for the borrowing and/or
Hebraizing of many Aramaic forms, might also have blunted the
motivation for further development.
 V SOME REMARKS ON
COLOUR TERMS IN MODERN
SPOKEN HEBREW
This page intentionally left blank
 211

SOME REMARKS ON COLOUR TERMS IN MODERN SPOKEN HEBREW

Although, strictly speaking, colour nomenclature in Modern

Hebrew is a subject beyond the scope of this study, I feel that
it is relevant to the present study inasmuch as it illustrates a
conscious, even artificial utilization of old word-stock and its
adaptation to changing needs and changed times.
When we deal with the renewal of Hebrew as a common language
for everyday speech we usually fix the origins of this phenomenon
in modern times at about 1880. More specifically, 'Modern
Hebrew1 is the language used among Jews in Israel (Palestine) as
the chief medium for communication especially since the first
World War. The waves of Jewish immigration that followed that
war, the Russian Revolution, the political conditions in Europe,
and the Second World War - in short, the chain of historical
events which transformed a small Jewish settlement into an
independent state - intensified the need for a language that
would preserve as much of its ancient features as possible while,
at the same time, function as a suitable vehicle for communication
in a modern Western-type (albeit genetically mixed), and technol-
ogically advanced, society.
Colour terminology gleaned from previous strata of Hebrew,
as is the case with other sectors of the vocabulary, was felt to
be inadequate for contemporary usage already during the 1880's.
The European education of the new Hebrew speakers made the gaps
in Hebrew lexis - even after all available diachronic sources
were delved into and consulted - acutely apparent. It seems
that the first task undertaken in this direction was that of
1
completing the list of the eleven 'basic1 colour categories.
To the existing stock of 13t» ('white'), Idly ('black'), DTK ('red'),
plT> ('green'), a'nx ('yellow'), and D-lh ('brown') - all found in
T T
biblical literature, although not necessarily relating to the
same points of reference - five more terms (for 'blue', 'purple/
violet', 'pink', 'orange', and 'grey') were added in quick
succession. DH3 ('orange') and !?rO ('blue') were coined as
derivatives of biblical and MH forms and bases (0113; VJPy flt>M3 -
•' ^ -T
Ezek. 23:40, and b£l3 in MH and in the Aramaic of the Targums ),
3
already in 1887. By the beginning of the 20th century Ben Yehuda
had already 'invented1 new terms for 'pink' (Till, from MH Til,
212 COLOUR TERMS IN MODERN HEBREW

4
believed to be of Persian origin ), and for 'grey' (113K, from
5 "^ .
biblical I^N, = 'ashes'). A word for 'purple/violet' - t>1>0,
from the name of the plant ^A^O = 'violet', mentioned by Rashi -
'T ' f
was added at about the same time. Thus, by the beginning of our
century all eleven primary terms were available for everyday
speech. Furthermore, at the end of the entry yiX ('colour, paint';
see discussion of biblical and MH ysx and forms related to it) in
7
Ben Yehuda's dictionary a list of chief colour terms appears.
The items on the list are mainly of primary and secondary status.
The 22 terms fall into the following categories.

a. Biblical terms (13): ihw >EJ1K, VBK, IIS, d.lh, ^fcafl, V-IBH, !?1*,
. . T T T T ' «- T T
nna,
V ••
iat>, snx,
TT T
ins,
T
pie;.
T

b. Terms found in MH and in Aramaic, or adapted from biblical

Hebrew (7): DDK ,10N ,111 ,1-111 ,Dn-J 8,o'nu> ,qhu;

c. Newly coined terms (2) : !?h3 , !?AO

Out of the 22 lexemes, 3 are marked by the symbol indicating

an original contribution by Ben Yehuda himself (&DN ,10K ,111).
T T T
Another lexeme which Ben Yehuda tried to reintroduce - >lon or
30n ('variegated') - never entered colloquial speech.
9 Sivan
is probably right in saying that Ben Yehuda wished to reserve the
-c- 10
pa ol pattern for colour names only. This, and the practice of
drawing heavily on ancient Hebrew sources (while at the same time
leaning on Arabic usage as well) were the two prominent principles
which governed the introduction of prescribed terms in years to
come.
The prescriptive nature of proposed terminologies pertaining
to various fields became even more emphatic with the periodical
publishing of word lists for specific subjects. These lists came
to fill serious gaps in the vocabulary, so that foreign words
would be used less frequently or - preferably - not at all.
Every issue of 1331W!? from the first one (1928/9) until the late
1940's carried such lists, the result of work done by sub-
11
committees of *|ia5n lyi (the Committee for the Hebrew Language)
and continued until this day by the Academy for the Hebrew
12,13
Language.
A prescriptive (semi-official) list of terms for colours,
colouring and painting was compiled in 1931-1934 and published
COLOUR TERMS IN MODERN HEBREW 213

in 1331^ VI (1934-5, pp. 83-87). Again, the most important

principle is the utilization of material found in the ancient
sources - and first and foremost, in the OT - and its application
to modern usage inasmuch as this was at all possible before
14
coining new terms. Some colour terms which were already in
existence at the time were granted a stamp of approval, while
others were newly suggested and subsequently either accepted or
else rejected by the speech community. This list will be called,
henceforth, "list A".
On reading the list the guidelines according to which it was
15
compiled become fairly clear. The main ones seem to be:
1. Hebrew forms (of all periods, but mainly based on biblical
Hebrew, MH, and recent additions) and Hebraized forms are
recommended in almost all cases. Some words are derived from
Aramaic (fl^lfiT, p. 85, and related lexemes; lit! and TlH, p. 86,
although the latter is a hapax legomenon in biblical language).
The foreign 1*^3 (p. 86) is an international culture word; thus
its inclusion (instead of its biblical doublet ^tt13) can be
justified. The adoption of the pa ol model as the basic pattern
for colour-denoting lexemes is in the spirit of the innovations
proposed by Ben Yehuda and his contemporaries.
2. Traditional Hebrew words are given a semantic shift of
sorts - their form is kept, while their designations might be
16
changed. So, for instance, Flton and its derivatives (p. 87).
3. Old forms are utilized as productive bases for new forms -
panon (p. 85), pvnn (p. 87).
4. For the specification of secondary or tertiary terms:
a. Whenever possible one single lexeme is prescribed, be
it an ancient Hebrew lexeme (VQK, p. 85), an Aramaic base (POO,
p. 85), or a derived innovated form (ton, p. 87).
b. The exact chromatic value a term represents within its
sphere is specified by appending to it a name for an ('natural')
entity which characteristically exhibits that shade. The resulting
syntagms are constructed as compounds - KVW p*1» - GrasgrrSn,
n ir''~
JT>T pT> = Olivengrun, Ittfl pT = Palmengrun (p."''85).17 More rarely,
• - -; T T -i
a compound of two colour terms is prescribed for borderline cases:
thus DTK-Din (p. 85) and I'lTO-VhS) (p. 86) for peripheral colours
that are situated away from the focus of their sectors.
214 COLOUR TERMS IN MODERN HEBREW

c. Degrees of brilliance/saturation are usually indicated

by appending an adjectival modifier to a (primary) colour term:
Vrm DTK = hellrot, flb-A 0"TK = hochrot, pfi^ DTK = tiefrot (p. 85) .
These look like caiques, or loan-formations based on European
practice.
The self-evident purpose of such a list is the controlled
addition of a whole body of new terms to the vocabulary of the
language. Pre-existing terms, while consciously exploited, must
perforce lose a certain amount of their original sense, and in
most cases they become narrower in scope. The new presence of
numerous lexemes within slots that were previously vacant or
else occupied by a single term or only a few terms makes old and
new mutually exclusive on a synchronic level. OIK, for instance,
is presented as denoting a much more specific value. While it is
still a blanket term (rot, p. 85), because there are so many
more monolexemes and compounds subordinated to it it now
designates - more than any other chromatic property - what in
biblical language could be properly called its focus = 'blood
18
!
colour'.f

To be sure, this dual function of a term as both superordinate

and hyponym in its own sector is not unique. It is already found
in biblical Hebrew (see DTK there), and the linguistic principle
behind the phenomenon is well recognized. However, the
introduction of new terms restricts the scope of the hyponym by
fixing other terms at its boundaries. The availability of YTl
(p. 85) would presumably exclude the usage of DTK, or DITtf1! h^,
from the description of 'healthy complexion', unless a direct
biblical allusion or association is intended. The list was
The list was published over forty years ago. Today any native
speaker of Hebrew would agree after a cursory glance that not all
of the old or innovated forms proposed fared equally well. Some
do not feature in standard speech, or perhaps never even entered
the parole sphere. These 'reject' items were either not accepted or
else have not survived into the 1970's, were relegated to the
level of literary/poetic diction and restricted to it, or acquired
meanings not recommended in the list. Thus the Aramaic-derived
series pno ,itt30 ,|7fflO ,p*non ,j7»fitsn does not refer today to a
f ~T '"" * '. • "~i '

specification of 'red' = 'blood-red', as suggested (p. 85).

Rather, this notion is signified by the hyponymic CHN itself.
COLOUR TERMS IN MODERN HEBREW 215

p'ODn means simply 'become red' and is mainly used in a syntagm

with'* D'UO (face), or even without O'JS, meaning 'to blush'. 19
On the other hand, this phenomenon of rejection is, of
course, a by-product of the process of expansion. The many
additional terms that have entered the language since have
sometimes supplanted the proposed words, while at others
segmented further the net of term-bound areas within the sectors
themselves or across them. The extent to which Hebrew colour
terminology grew over the intermediate years can be gleaned from
consulting the list in the Rabin-Raday Thesaurus (from now on
"list B"), while a few more illustrations and observations
concerning the tendencies of the 1331g>5 list (A) and the
suggestions incorporated in it seem to be in order at this point.
ton = hellblau, Himnelblau (A, p. 87) is not often heard
nowadays in colloquial speech. The substitute employed is t>D3
(p. 86), together with an adjectival modifier (1*M3 !?hs) or a
nominal modifier (D'OW !rtlD). Otherwise, the colour is simply
• -T T
referred to as the colour of n^Dtl. In other words, t>h3 and !Ofl
v: T T
are felt to belong to the same non-linguistic basic category.
Hence, the most obvious mode of differentiation between them -
that of using two separate monolexemic terms, with or without
21
modifiers - is only rarely practised in common speech.
List A follows biblical and MH sources in prescribing the
employment of the Pi. theme for naming the activity of painting
into a certain colour, and extending it to all basic terms and
to some secondary ones (0;fN >P>30 - p. 85; j7T> ,^flD , ISt^ - p. 86;
ttX ,~irn». ,b3fl - p. 87). In fact, this recommendation seems to
have been rejected. In practice the verb /yn^ Qal - together
~ c~
with the term for the colour referred to (in the pa ol form)
and with the prepositions t> or U affixed to the latter - is
commonly prevalent: pVU) )m,tnK(3) )m (or dTNt> ,P^» etc.).
This, then, is an illustration of the rejection of a traditional,
compact pattern in favour of a compound which is possibly
influenced by foreign modes of expression. Similarly, syntagms
like pT> .1flil»"ni Moyj are much more common than single forms
T . —r T T-"
such as the proposed Tlinn
**T*. '
(p. 85), pvnn
•• T ' '
(p. 86) - although
D^Ktin (p. 85) does feature in the language.
In list A nnp (nnp "p!? = stumpfweiss, p. 86) and rDA
. . "•'*-,'" TT ~ T
(ttia QTN = hochrot, p. 85) are proposed as modifiers for
216 COLOUR TERMS IN MODERN HEBREW

brilliance/saturation designations. These have not been

accepted. On the other hand, W^Q ,pth (not included in the
list) are often heard in this context. Nominal syntagms of the
adj. modifier + colour term, or colour term + name of object
(in the construct state) are easily generated, thus adding
continually to the stock of more or less permanent colour
terminology.
The few observations presented above deal chiefly with
processes that are fairly normal in the life of a language. The
disappearance of terms and the substitution of others for them is
inevitable whether the terms stem from organic growth or are the
fruit of conscious invention. Far more important is a basic
difference between the at least partially prescriptive list (A)•
and actual practice. While the first speakers and teachers of
Modern Hebrew always sought to exclude foreign words from the
renewed language, their wish has not been fulfilled. As in
other fields, foreign colour lexemes - usually those appearing in
more than one European language - have never been absent from
current speech, even when original Hebrew terms were potentially
available in monolexemic formations or in syntagms of the types
described above. So much so that some of these terms came to be
22
'legitimized' by respected Hebrew dictionaries. In Rabin-Raday
(list B) we find, among others, 'JK'iPD (after the colour used by
the painter Titian) under the heading DTK (p. 868); 'T3 = beige
under D-in (p. 869); noi|7 mo (p. 869); >pn (khaki), taio = topaz
and ^aili^i) = chlorophyll under p"P (p. 869); T'pYlO = turquoise
under !?ns (p. 869), and so forth. There is a definite tendency
to use foreign elements as designators for colour properties,
whether on their own or as part of a syntagm, in their original
form or in a Hebraized form. In other words, the common practice
which the compilers of list A (1935) intended to hinder proved to
be necessary in at least some cases. Hence, these terms are even
'legitimized', or at any rate grudgingly tolerated. Again, this
tendency is, of course, not unique to the colour field only, but
characteristic of Modern Hebrew vocabulary in almost every field
and/or level of speech.
To return to the two lists themselves. Under the heading
ySS, which introduces list B, we first find the terminology
COLOUR TERMS IN MODERN HEBREW 217

referring to the activities of painting and dyeing in general

(pp. 867-868) that is analogous to the same section in list A
(pp. 83-85). Then follow terms describing actual colours.
These are arranged in alphabetical order under primary headings
(D1TK ,TIDK etc.). When we compare the two, the quantitative
difference between them is immediately apparent. List B is so
much more voluminous than its earlier counterpart. This cannot
be attributed to the growth of the language during the intervening
years only. It must be borne in mind that the Thesaurus (list B)
is a diachronic multi-level compendium, not a dictionary of
spoken Hebrew. Many forms there may be unknown, or seem obscure,
to the common speaker; alternately, they exist within the realm
of formal diction without functioning on the standard level. In
a sense, then, D*5)3ft "1S1K presents us with a list that belongs
to the 'competence' level, while no adequate indications for the
'performance' status of single items is actually supplied.
Therefore it is not easy to draw specific conclusions in regard
to colloquial speech - this will require controlled field work -
apart from a few additional observations that are pertinent to
our subject.
1. The productive tendency - the creation of new colour terms
on the basis of ones that already exist in previous strata of
the language - is still prominent (list B, p. 868: »:jn:nK< 10A"1K)
2. Although the traditional pa°ol pattern is retained whenever
possible (especially in the case of primary terms), another
popular structure is that of a nominal form - in some cases, the
name of an object which characteristically exhibits the colour
referred to - to which the gentilic ending /">/ is added. Thus
>in>1K (above), ^fiBhJ « nwhi = 'copper', 'brass'), *VSO (<C.V3D),
and so on.
3. In biblical Hebrew the majority of colour lexemes are either
nominals which may function adjectivally as modifiers, or verbs.
In MH we already find DTK with the approximate meaning 'redness',
and there are other nominal lexemes conveying the abstracted
23
quality of 'redness', such as tT'O^tt'TN. In our list B there are
many more terms that refer to the quality of the colour itself
rather than to the colour quality of the thing/object described.
Thus from OTN, in addition to tHIN, we have also tl^OiaTK,
218 COLOUR TERMS IN MODERN HEBREW

imam™ ,n.io-TOTN ,M»*TK ,rpm*nK (p. 868).

While the new verb formations follow the usual models of
Qal, Pi., and so on, new nominal patterns are utilized for
filling areas which are unoccupied in biblical Hebrew.

4. In both A and B, and in contradistinction to biblical usage,

there appear many syntagmatic expressions which function as
designations of single extra-linguistic colour concepts. These
belong mainly to two types: (a) colour terms + an adjectival of
a general modifying character, and that for the specification of
saturation/brilliance or chromaticity, or (b) a construct - a
colour term + a lexeme denoting an entity which is typical of the
colour evoked. The comparative structure (0^3 DTK - list B,
ltl 869
p. 868; 3Pi?" ,l*y?> P- ) isdocumented too.

5. As mentioned above, the inclusion of foreign terms in list B

should be understood as proof of their popularity.

To conclude: the process of expanding the colour terminology

both by utilizing the sources and by borrowing and adapting terms
from other (modern) languages continues apace. The enormous
difference in bulk between the catalogue of colour lexemes in
biblical/MH on the one hand, and Modern Hebrew on the other, is
an adequate indication - although only one of many others - for
the difference in character and in technological achievements of
the societies which used, or use, Hebrew as a means for daily
communication.
VI NOTES
This page intentionally left blank
NOTES TO PP. 1-12 221

NOTES
Part I
A

1. Optical Society of America (1953), 220.

2. Ibid., 45; R.M. Evans, An Introduction to Color, 77.
3. Evans, op. cit., 59.
4. Ibid., 77.
5. Optical Society of America, 12f.
6. Ibid., 45.
7. M.D. Vernon, The Psychology of Perception, 219-220.
8. Ibid., 219.
9. H. DUrbeck, ZDMG 118 (1968), 23 (quoting E. Hellmig and
D. Katz).
10. Vernon, op. cit.
11. Ibid.,92-93.
12. Cf. II, B.O., p. 49ff.
13. E. Ullendorff, 'Is Biblical Hebrew a Language', Bulletin of
the School of Oriental and African Studies XXXIV (1971), 241.
14. S. Ullmann, Semantics: an Introduction to the Science of
Meaning, 125.
15. J. Lyons, Introduction to Theoretical Linguistics, 402-404.
16. S. Ullmann, Principles of Semantics2, 183-248.
17. For a more detailed description of ifl^ »^i?^l? see H> ^.5.
For the description of v^MS and its derived 'forms in MH,
see III, B.5.
18. BOB, 843; KB, 795; Ben Yehuda XI, 5402.
19. So KB, 795; Qimhi, D">ETmn -|fl&, 614; for a textual occurrence
see Tos. Neg. 1:2.
20. E.g. Jastrow, 1264.
21. Z. Ben Hayim, 'X D^n ^ph', Tarbiz 12 (1951), 75-77.
22. J. Barr, Comparative Philology, 125-151; and examples cited
there and on 320ff.
23. Ben Hayim, op. cit., 77; also M. Moresheth, t?jnon ll^O!)^
nilKJ&n nw^l Knnnnu; (Dissertation, Jerusalem, 1972), 187-188.
24. Ben Hayim, op. cit., 75.
25. For a summary of the dimensions and the variables discussed
here cf. Optical Society of America, 49-57, 221-293; G.A.
Collier, Language 49 (1973), 245-248.
26. Optical Society of America, 57-58.
27. S. Skard, 'The Use of Color in Literature', Proceedings of
the American Philosophical Society 90 (1946), 174.
28. W.E. Gladstone, Studies on Homer and the Homeric Age, III,
457-499.
29. L. Geiger, Contributions to the History of the Development
of the Human Race, 48f.
30. DUrbeck, op. cit., 25.
31. A. Maerz, M. Rea Paul, A Dictionary of Color, 8-11.
32. Vernon, op. cit., 71.
33. Ibid., 93.
34. Ibid., 69.
35. B. Berlin, P. Kay, Basic Color Terms: Their Universality
and Evolution, Appendix II, 134ff.
36. Gladstone, op. cit., 457. Cf. also Skard, op. cit., 165-166;
Durbeck, op. cit., 24-25, B. Landsberger, JCS 1967, 139-140
(especially n.7).
222 NOTES TO PP. 12-14

37. There is no term for 'blue' in Hebrew, Akkadian, Aramaic,

and Greek, to name only a few of the literary languages of
the past.
38. H. Magnus, Untersuchungen uber den Farbensinn der Naturvolker
(Jena: 1880). His findings are summarized in Berlin-Kay,
op. cit., 139-146.
39. G. Allen, Colour Sense (London: 1879), reported in Berlin-
Kay, op. cit., 137.
40. ibid., 147-149.
41. DUrbeck, op. cit.
42. A.C. Heinrich, 'A Non-European System of Color Classifi-
cation', Anthropological Linguistics XIV (1972), 220-227.
43. DUrbeck, op. cit., 24-25.
44. P. Fronzaroli, Studi Linguistic! I, 383-386.
45. Berlin-Kay, op. cit., 139-149.
46. Vernon, op. cit., 69-70.
47. R. Gradwohl in his Die Farben im Alten Testament recognizes
the multi-dimensional nature of colour phenomena, and
attempts to define lines of lexical development (esp. 89-98).
His study is mainly etymological-comparative. He uses
traditional "European" criteria of hue-discrimination.
Apart from a few general statements, he does not deal with
the complexities which colour sensation attributes impose
on colour terminology. His methodology and his conclusions
will be referred to in the appropriate sections of this
study.
48. When the word 'primitive' was used on previous pages, the
context was always that of representing the opinions of
scholars who use it to signify cultures that are thought
to be less technologically and materially advanced than our
Western, post industrial revolution civilization. As for
myself, I hesitate to use it because of the pejorative sense
of "simple, rude; uncivilized or of rudimentary civilization"
(The Concise Oxford Dictionary,5 969), which is prevalent in
colloquial - if not scholarly - discourse, and imparts a
feeling of superiority. Therefore I opted for the more
precise, if admittedly clumsy, modifying noun-phrase.
49. This example, as an illustration to the dimensions of
colour nomenclature, is discussed by Fronzaroli, op. cit.,
384-385, where Guillaumont's opinion in his article in
Problemes de la Couleur is mentioned as well.
50. Berlin-Kay, op. cit., 2.
51. V.F. Ray, South-western Journal of Anthropology VIII (1952),
252.
52. H.A. Gleason, An Introduction to Descriptive Linguistics, 4.
53. E. Nida, 'Principles of Translation as Exemplified by
Bible Translating1, 1.
54. E. Sapir writes: "... the "real world" is to a large extent
unconsciously built up on the language habits of the group.
... We see and hear and otherwise experience very largely
as we do because the language habits of our community
predispose certain choices of interpretation" - quoted in
B.L. Whorf, 'The Relation of Habitual Thought and Behaviour
to Language', in: Language, Thought and Reality (1956, ed.
J.B. Carroll), 134. In another article,'Language, Mind and
Reality', Whorf summarizes an example by saying, "... we
all, unknowingly, project the linguistic relationships of a
particular language upon the universe, and see them there
...". (262).
NOTES TO PP. 14-23 223

55. Berlin-Kay, 5.
56. See criteria for the classification of biblical colour
terminology-II, A.
57. Berlin-Kay, 2.
58. Ibid., 17-36.
59. Ibid., 16-17.
60. ibid., 36-41, with examples.
61. J.H. Hill, K.C. Hill, 'A Note on Uto-Aztecan Colour
Terminologies', Anthropological Linguistics XII (1970),
231-238.
62. D.L. Snow, 'Samoan Color Terminology: a Note on the
Universality and Evolutionary Ordering of Color Terms',
Anthropological Linguistics XIII (1971), 385-390.
63. Heinrich, op. cit.
64. J.A. Frisch, 'Mohawk Color Terms', Anthropological
Linguistics XIV (1972), 306-310.
65. H.B. Broch, 'A Note on the Hare Indian Color Terms',
Anthropological Linguistics XVI (1974), 192-196.
66. G.A. Collier in a review of Basic Color Terms, Language 49
(1973), 245-248.
67. See, for instance, Frisch's article (note 64). Questions
concerning the validity of the system - as defined in the
book, and without modifications - are raised by Snow (note
62) and Hill and Hill (n.61).
68. Although, strictly speaking, a detailed structural study
of the colour field in cognate languages lies outside the
scope of this study.

IB

1. E. Ullendorff, Bulletin of the School of Oriental and

African Studies XXXIV (1971), 245.
2. For the problem of plene and defective spelling cf. A.
Murtonen, 'On the Interpretation of the matres lectionis in
Biblical Hebrew', Abr. Nahrain XIV (1973-1974), 66-121,
and the literature cited in the notes thereof*
3. Ullendorff, op. cit.
4. C. Rabin, mm 3 D^Q, Enc. Miq. IV, 1069.
5. Ullendorff, op. cit., 253-254.
6. For the status of Mishnaic Hebrew (MH) vis-a-vis Biblical
Hebrew, see part III.
7. Fronzaroli, 'Struttura dei Colori, 385.
8. CAD, A, 300ff.
9. Fronzaroli, op. cit.,
10. Ullendorff, op. cit., 246ff.
11. BOB, Mgn, 641.
12. Gradwolil, Farben, 90-91.
13. See, for instance, lists of Mishnaic lexical innovations
(other contemporary Hebrew sources excluded) in C. Albeck,
naflflt? K1D3, 134-215; C. Rabin, Hilt D?!?n, Enc. Miq. IV,
1070-1080.
14. DUrbeck, ZDMG 118 (1968), 26.
15. BOB, 10.
16. Ibid, 71.
17. F.R. Palmer, Semantics. A New Outline, 91-97. Fronzaroli,
op. cit., 378-380.
224 NOTES TO PP. 24-4

18. Palmer, op. cit., 49. A discussion of the scheme follows

on pp. 49-51.
19. N.H. Tur-Sinai, 130m IIW^H (Jerusalem: 1951), 387.

C_
1. F. de Saussure, A Course in General Linguistics, 65-70.
2. F.R. Palmer, Semantics, 26.
3. Berlin-Kay, Basic Colour Terms, where 'term' is used
throughout the study.
4. de Saussure, op. cit., 65.
5. Palmer, op. cit., 30.
6. Ibid., 30-34.
7. E.A. Nida, 'A System for the Description of Semantic
Elements', Word VII (1951), 2-3.
8. Lyons, Introduction, 196f.; Palmer, op. cit., 37-39.
9. Lyons, op. cit., 197; Palmer, op. cit., 39.
10. Ullmann, Principles of Semantics, 157; see also survey by
S. (jhman, 'Theories of the "Linguistic Field"', Word IX
(1953), 123-134, and Ullmann1s discussion of the concept
of 'fields', op. cit., 152-170.
11. T. Donald, 'The Semantic Field of Folly in Proverbs, Job
Psalms and Ecclesiastes1, VT XIII (1963), 285.
12. Thus also Oilman, op. cit., 134.
13. Principles, 254-256.
14. Fronzaroli, 'Struttura dei Colori', 382.
15. BOB, 850.
16. Ullmann, op. cit., 254-265.
£
1. In Oudtestamentische Studien XIX (1974).
2. i:mttfr XXXII, 405.

E_

1. As a matter of fact, even group 4 is more 'indirect' than

'direct1, in the sense that it does not include terms
which refer to. natural colour phenomena, but to man-made
appearances of colour. Characteristically, many names
for dyes have a double reference: to the dye, and to the
material that is habitually treated with that dye.
Part II

A
1. Berlin-Kay, 6-7.
2. In contradistinction to 'form' or 'function' words; cf.
Palmer, Semantics, 37.38.
3. Active verb forms are listed as primary colour terms when
they share the same consonantal root with the primary
nominal form from which they are derived (e.g. QTK*, D'^Kil
0;JKtyi1*> inunediately after DTK) . It is clear that because"
these verbs are denominatives, they are genetically
secondary. On the other hand, they carry the wide reference
of the nominal form, and thus belong with it. The Pu. form
d^TKO functions as a nominal, is limited in its application,
NOTES TO PP. 41-49 225

and its reference is possibly narrowed to 'dyed red'

(BDB, 10). Therefore it was excluded from the group of
DTK - verbs under the 'primary' label.
4. 'Dead' in the sense of 'not being used for daily communica-
tion'. For the problem of the "life" or "death" or Hebrew
from 200 A.D. to its "revival" at the end of the 19th
century, see a recent article by J. Fellman, Anthropological
Linguistics XV (1973), 250-257, and literature cited there.
5. See Section I, B.a.
6. Ibid.
7. Fronzaroli, 'Struttura dei Colori', 382.
8. For an analysis of the construction yilh l7lp*T (when the
first member of the combination is the qualifying, and not
the qualified element) and similar instances see under
frill ,i>n'- ~ u> B-4-
9. Cf. II,VC.O.
10. Gradwohl, Farben, 4. Cf., however, II, B.2.0.1.

IJ

B.O
1. Cf. II, A.I.
2. The following lexemes are excluded from the above count:
a. 13^ in the combination DlftTK 1^ (Lev. 13:19, 24, 42,
43), which seems to belong to the tertiary layer of
colour notions, and which will be discussed there.
The same applies to "p^ in the syntagm HJ^ ^<&
TT T:
(Lev. 13:39).
b. mNn (in Nah. 2:4, and 6 times in Ex. in the syntagm
D^tt'TKn D^K filly) - because this passive formation is
of secondary nature and functions as a qualifier which
appears - in all instances but one - in one and the
same syntagm. Therefpre it is not considered a primary
term.
c. "lirHJ (Lam. 4:8), be its original reference what it may,
is first and foremost a name designating a certain
substance, not a colour notion. Like >!pK7 ,^!?n , 1?1
in other instances, it strengthens the colour^allusion
of .IBh (ibid.) but does not create it. The synchronic
existence of 1(l)hW in the language also argues against
accepting lirHU (like p*V, in some instances; see under
PIT) as a direct colour term.
3. For the implied - although not stated - analogy of l^tj. to
n>nr2n "py by juxtaposing Ex. 16:31 and Num. 11:7, see II,
BT2":4. "
4. The point of departure for the description of colour terms
are lexemes that are used in order to convey a direct
colour sensation. Until, and if, another connection will
be established for such groups as IThW - 11HU); .DIN-D1!,
?OJ3& the non-direct 'colours' will^be considered as
'derived terms' although, etymologically and semantically,
this may perhaps emerge as incorrect in a detailed analysis
of each of these lexemes. The terms 'derived', 'derivatives'
are here employed to denote secondary formations that share
the consonantal skeleton of the primary term (that is, the
root), together with variations of the vowel patterns, or
the addition of certain consonants either as preformatives
or as aformatives.
226 NOTES TO PP. 52-58

5. The division into a 'physical' group and an 'abstract'

group should be understood as:
i. 'physical', terms that denote objects which may be
perceived by the (human) senses.
ii. 'abstract', terms that denote entities which cannot
be directly perceived by the senses alone, but have
to undergo a cognitive filtering process.
In other words, the division is that of 'concrete' and
'figurative' use of language, without using these over-
worked terms, but also with no adequate new catch-words
to head the old categories; hence the inverted commas.
6. Dan. 11:35 (Hif.?), 12:10 (Hithp.) and Ps. 51:9 (Hif.)
are actually connected with persons; however, the verbs
are linked to terms that, in these instances, refer not
to persons per se, but as sinners, that is, to people of
a certain moral or ethical state. As such they belong
under this category rather than the first one (i.e.,
'physical', human). For the non-colour - but colour-
derived - reference of this usage see under 151?, II,
B .9/. .Q O . ^"^"
7. In other words, as far as 'hair' is concerned, there is ,
an appositional (achronistic) set of "\& - ':JO}tf7& -1(1)ni>7.
As for skin sores (or lesions), the apposition there is:
•pt}. - DTMTK - 0*m*TN 13!>. In the latter case, then, three
levels (from the primary to the tertiary) of colour lexemes
are represented.
8. See discussion of ."IT* ,h^ under 'Secondary Terms', II,
C.2.2.
9. Fronzaroli, 'Struttura dei Colori', 380.
10. Gradwohl, Farben, 23, where injf follows v'O'W, v"lOfl, pl'UT*,
V^ltfl , and their derivatives as a term for some kind of 'red'
colour.
11. Fronzaroli, op. cit., 384.
12. Gradwohl, op. cit., but see Blau, "miU/t? XXXII, 405, in a
review of Gradwohl's work; and under iVtf (II, B.5.).
T
13. Cf. II, D.I.I.
14. ^iliO? in its context is of dubious status, as pointed out by
BH , both grammatically and reverentially; for further
discussion, see under 3nsO (n, C.5.).
T:
15. II, B.4 and C.4. '''
16. Op. cit., 385. The opposite view is expressed by Blau,
op. cit., 403.
17. For illustrations of this process see figures 4-10 in
Berlin-Kay, 17-22.
18. Fronzaroli, op. cit., 387.
19. Cf. also Gradwohl, op. cit., 52.

B.I
1. All instances will be discussed in the order in which they
appear in the Hebrew OT, nominals dealt with before the
verbal occurrences connected with them, on the basis of the
assumption that colour-denoting verbs are normally denomina-
tives in origin. This assumption is borne out by the fact
that - even after allowing for the arbitrariness of textual
preservation - the number of nominal forms referring to
colour exceeds the number of verbal formations. As far as
NOTES TO PP. 58-64 227

this semantic field is concerned, it would seem that

stative phenomena are usually defined by a compact unit
(i.e., a single lexeme) before the process or the activity
is similarly delineated.
2. BOB, 9f.
3. KB, 12f.
4. Vol. I, 29.
5. The various occurrences were examined according to the
editions of Sperber, Diez Macho (for Neophyti J), and
ni!n"U niNlpn (Makor, Jerusalem 1973).
6. For the translation of each OT occurrence into Greek and
Latin see Gradwohl, 7ff. passim.
7. Gradwohl in his summary to the "Rot" section, 26-27.
8. Berlin-Kay, 13.
9. These attempts will be discussed under OIK ,t)1 .fiOIK.. The
referential connection (as distinguished from the
etymological one) between DTK and 01 will be touched upon
under 1.1 (Gen. 25:30) and under 1.3 (2 Ki. 3:22) of this
section.
10. Genesis, ICC, 361-362.
11. A. Dillmann, Genesis, 320-321.
12. T.H. Caster, Myth, Legend, and Custom in the OT, 368, n.13.
13. C.J. Ball, Genesis, 80.
14. Ibid.
15. So also Gradwohl, 7.
16. Sperber, I, 38; niKlptt, I, 63.
17. Blau, i:m«ft XXXII, 403.
18. D. Daube, Studies in Biblical Law, 191-196.
19. G. Henton Davies, Genesis, 209-210, cites Daube's theory
and seems to accept it.
20. G. von Rad, Genesis, OTL, 261 (though cautiously).
21. W.H. Bennett, Genesis, Century Bible, 265.
22. BDB, 727. So all commentators, from the ancient Versions
to Jewish medieval commentators - Rashi, Qimhi and others -
and modern scholars.
23. Jastrow, 536 (•phfl'bp) , 1046 (t^Tr^); and he quotes B.T. Ab.
Zar. 38b, where the practice of bo'iling lentils in vinegaT
or in water is discussed.
24. Gradwohl, 7.
25. H. Gunkel, The Legends of Genesis, 57.
26. Gradwohl, 9.
27. N.H. Snaith, Numbers, CB, 270-271.
28. Snaith, op. cit., cites material reported by Frazer,
The Golden Bough, II, 142, 254-311.
29. G.B. Gray, Numbers, ICC, 248.
30. Ibid.; for similar rites and for the passage in general
see 241-247.
31. Berlin-Kay, 17.
32. For instance: *D 1^7 l^n n»11K .113 inifl Dl^ T^ttt in
.anr n!>pwn !?Kiun
•pn Ntn nniiK mot> ^KIW isiusuw nwya
...inn a"iDy !?XK nmK INJM D"nKVQ>K^ifl
33. Gradwohl, 9. See, for instance, Sifre for Numbers (ed. H.S.
Horovitz, 1966, 152); ,1O»nn IK niKPmN^.iMPWl nmnn...
5"ti nn xn CPIIKJK tpnin >in cno ru VK IWN iaiK Kinws .D»ma5
nimniK^ hmnn .in^nn. So also Rambam, nun nJWtt (Jerusalem,
1962; Comm. by M. Reich, 119): rwan iltt'tin Mlim IfiKiW nt
.nio^ttiK!?
228 NOTES TO PP. 64-71

34. Gray, op. cit.

35. Cf. Rashi, Qimhi, J.A. Montgomery (Kings, ICC, 361), J. Gray
Cr and II Kings?, OTL, 489), They rationalize the
improbable reasoning of the text - how can the water be
bloody-looking by merely reflecting the sun rays? - by
assuming that the flood waters possibly carried reddish mud,
or sandstone; however, all commentators agree that the
Moabites were deceived by the combination of the two
elements: a sudden flood, and its strange colour
(Montgomery, op. cit., 362, even considers a mirage!).
36. Gray, op. cit., 480f.
37. Gradwohl, 4-6.
38. Ibid., 5.
39. CAD, III, 79.
40. Montgomery, ICC, 362, cites, and marks as "wrong", the
omission of D?T3 in LXX .
41. Gradwohl, 8. "*""
42. Ibid/, 27.
43. C. Westermann, Isaiah 4O-66, OTL, 380-381.
44. So also Ibn Ezra to this verse. He observes laconically
abcdefghhijklmno
45. So. Skinner, Cambridge Bible, 196. T.K. Cheyne, Isaiah,
mentions the following emendations to v.l: instead of OTTKtt
read O^KJ?; for fll^O read 1^'an (so also BH3) - but rejects1''"
them as unnecessary (162) . "As he points out, the allitera-
tive element is probably a by-product of the text, and
should not be carried too far. Westermann, op. cit., 381,
also rejects the emendation, claiming that there are no
compelling grounds for the alteration, and if we accept
it we lose the exact reference to the actual scene of
events. Gradwohl, 10-11, does not accept the correction
either.
46. Cf. II, D.I. 4.
47. I.e., the treading of the wine press as a recurring simile
for YHWH's action in judgement - Joel 4:13; Lam. 1:15; and,
in the NT, Rev. 14:19, 19:15 - so Westermann, op. cit. ,382.
48. Gradwohl, 8.
49. Pines, Enc. Miq., VI, 665.
so. AS in tr^ina rntnpa.
51. Fronzaroli, 'Struttura dei Colori', 379.
52. Herszberg, ZDPV 69, 177- as noted in Klpon 11K)» 1S1K, I. 32.
53. G.H. Smith, Twelve Prophets, 278, 298.
54. Gradwohl, 8.
55. E.G. Mitchell, Zechariah, ICC, 119.
56. Ibid., 119-121.
57. (for 1:8). ."Dm (Dtf \lfrtf Ym 'PNI flNI HD3 DTK 010 ty"
T
58. Mitchell, 179 (forv.6).
59. G.A. Smith, Twelve Prophets, 276; Mitchell, op. cit., 116.
60. These terms will be discussed under the 'secondary' and
'tertiary* headings, where attempts to clear up the
difficulty of the lack of tPJsTTK.in 6:6-7 - and the textual
problem of O^MJK.t^Tj^ (v.3) ->rwill be referred to.
61. M. Dagut, A Linguistic Analysis of Some Semantic Problems
of Hebrew-English Translation, 20ff .
62. TO DI13 (in BH there are two suggestions for emendation: to
read,' ''following the LXX, Tfll DtO; for some reason, this
regular hendiadys construction seems more acceptable; or
to read T£J)3 0113 (as if the B was omitted by haplography)
NOTES TO PP. 71-76 229

will be discussed under (Indirect) Allusions,to Colour.

The last part of v.ll, 311JJD Ittlhtf (which, BH proposes,
should be omitted on metric grounds, but whose status
vis-a-vis the referential connotation of T3 dfO should
also be clarified) will be referred to under ihtU in this
section (Primary Colour Terms, B.3). Inasmuch as the two
syntagms will be mentioned in this part (1.6), it will
be done in general terms only, as analogies and background
for D1TK1. n*.
63. See under secondary terms subordinated to "IS!?-- II, C.2.2.
64. Rashi, tTlVm mKlpfl for this verse; MartiniTG. Currie,
Proverbs, Ecc., and Song,, Century Bible, 335.
65. Numbers, 271 (to 18:7).
66. E. Ullendorff, Ethiopia, 126. Even Gradwohl, 6, hesitantly
says that this might be the case.
67. Pines, Enc. Mig. VI, 665.
68. See summary to the 1^5 section in this chapter (II, B.2.9).
69. See discussion of DTJIrTK "\& (II, D.I. 2.).
70. For further discussion of v.ll see under 1DW (II, B.3.1.2.).
71. MT O^IlBtt.
1Q Isaa, as well as four Hebrew MSS, have ^B>3, as is the
case in the LXX, the Vetus Latina, the Syriac, and the V.
G.B. Gray, Isaiah I-XXIX, ICC, 29, sees CPJty as pi. of ^iB,
'scarlet-clothes', and compares to Q'):T2l> "linen-clothes',
and Prov. 31:21 D^W v& rims to o >^n nwfi KT>n H"y.
For the latter verse, however, BH^ suggests an emendation
(because of the context - one would expect warm clothes to
combat the cold, not colourful ones). In addition, here
the final D might have been a dittography of D**mn, the
first word of v.22. Also, many Hebrew MSS read here ">W.
According to C.D. Ginsburg, Later Prophets, 2, the VI*SO
for D'JBO, Isa. 1:18, is *3^. On the grounds of the
dubiousness of Prov. 31:21, the evidence of 1Q Isaa and
other Hebrew MSS, and the use of "Oty as 'scarlet clothes'
(in the plural, 2 Sam. 1:24, Jer. 4730) it seems better to
emend our instance to '3^..
72. BH has a suggestion to read OKI, following many Hebrew MSS,
and the Versions. Gray, Isaiah} 30, says this emendation is
possibly called for, especially by analogy to the structure
of vv. 19-20: OK (v.19)...OKI (v.20).
73. Gradwohl, 9, 27^
74. BOB, 9.
75. GB, 13.
76. Kipnn Tiuft "unN, i, 32.
77. Gradwohl, 9.
78. Fronzaroli, 'Struttura dei Colori', 384.
79. For a discussion of y£in and 'aw see under II, E.2.1.
80. See sections II, B.2.0 and 2.I.T.
81. So, for instance, Ibn Ezra, who clearly states that the
picture conveyed in v.18 refers back to 12m (16) etc., i.e.,
the possibility of repentance and a change in the present
state of affairs.
82. Gray, ICC, 27-29.
83. Ibid.26-27.
84. So K; Q 01M .
230 NOTES TO PP. 76-78

3
85. According to BH , some words are perhaps missing here;
indeed, as the verse stands, its second part - 1*3 *y— *3 -
does not make much sense, although 13>y Ol^O IIU has become
an (opaque) idiom in later Hebrew, meaning: loves to drink,
or even habitually drinks. G. Currie Martin (Proverbs,
Century Bible, 1908, 148) suggests something along the
line of 'when it (the wine) is red, when it gives its
colour in the cup'. Similarly McKane (Proverbs, OTL, 197
394): 'red wine sparkling in a cup'. The interpretation of
the two above-mentioned scholars might, however, be based
on the text itself, with U*y explained as 'its (of the
wine) appearance', the way it looks, in the cup. Furthermore,
the problem of the subject of the IJPy . „. >D part is not
solved: from the point of view of contiguity the grammatical
subject should be 1J', but the verb "|fl* is not usually
applied as a predicative element to a non-animate subject.
Disagreement in number excludes the tPlhKfi and Q^K^ of
v. 30. Perhaps we have an oblique subject here - whoever
is fascinated by wine, his fate would be ... (v. 32ff.).
In general the passage, with its changes of pronouns
and grammatical number, is far from smooth, and reads
like a collection of random sayings arranged together
because of the subject matter (exaggerated wine drinking).
86. D^IWMi "l^ntl* does not make sense here. BH3: perhaps an
addition from Song. 7:10 - itself an obviously corrupt
verse - where it is proposed to correct the MT *Tl*T^ "l^lfi
Q->W) it\!w a:m D>IB»D^ (subject - wine) - to asni i!nn
d^itai a->ii£jp >3ftf> (BH^ iMd.).
87. Gradwohl'/10. '':
88. BOB, 10.
89. KB, 12-13. Similarly also Gradwohl, as above, and 1S1K
Kipnn iit!>5 i, 32.
90. C.H. Toy, Proverbs, ICC, 439, justly draws our attention
to Isa. 63:1-3 (1.4).
91. C. Gordon, Textbook, Gloss., no. 483; G.R. Driver,
Canaanite Myths, Glossary, 154.
92. CAD, III, 79.
93. So Currie Martin, 148; McKane, 394.
94. Ullendorff, Ethiopia, 126.
95. In BH there is a proposed emendation of M^VTJ to n>iyj.
A.W. Streane (Jeremiah and Lamentations, Cambridge Bible,
385) sees in M**PT3 a reference to the Rechabites (Jer. 35).
The Targum retains the MT form. Although Rashi says (in
his first comment): iroi in 1ED i1»1B ,V1>n, he goes on
to state that, in fact, the text mentions "real"
Nazirites.
96. The combination 0»J»3£))3 d^y in*TK is problematic because of
QXy (for discussion of reference see below), its semantic
reference, and its syntactic position within the utterance.
Oi^y is missing in LXX, and the Syriac has the equivalent
of Qil^nxy; V - ebore antique; and the Targum, within a
fairly free paraphrase, lT*n lp'00 (= their faces,
appearance). _In BH3 we find three suggested corrections:
a. D^JflQ Qliy DTK; or b. Q»n'J3 *Sytt DTiy DTK; or c.
d'man QnflK;TnmK".T I would suggest that to~omit Dify -
following the LXX - should be sufficient, and would not
impair the sense of the utterance. Gradwohl, (9-10) who
tries to retain ti^y by comparing it to Prov. 15:30 (//iJtO ,
NOTES TO PP. 78-82 231

16:24 (//WSJ) and Ps. 35;10, still has to define D2y

in this particular context and to rearrange and correct
the syntactic irregularity; as it stands, ttty is a
hindrance rather than an asset for the understanding of
cpjpjfln .innK.
:T
97. Fronzaroli, 'Colori', 382.
98. Excluding Job 30:30 - *5yfl ini2> my- because here the
basic lexical unit is ... 0 IflW, and not in^.
99. For -1ST , tins see under subordinate l^ terms. (II, C.2.2).
TT
100. Gradwohl, 10. .
101. See niK31 »JN niinw, Song. 1:5, under inta, II, B.3.3.
T
102. Gesenius, 13.
103. A.W. Streane, Jeremiah and Lamentations, 382.
104. Y. Sukenik (Yadin), nriyio nnuw 'anwi d^oiai n»»cn ^iKQ,
nWttlN Tarbiz XVIII (1947), 125-128.
105. C. Virolleaud, 'Ras Shamra 1179', Syria XXI (1940), 274ff.
Cf. also von Soden, AHW, II, ad loc.
106. In BASOR 102 (1946), 7ff.

B.2
1. For the etymological relations and origins of the various
lexemes in which "p^ is manifested in various indirect
forms see under the appropriate heading (II, F.I.1.2
and F.I.2.2.).
2. Gradwohl, 4, 34.
3. In work done in Oxford, 1976. Prof. Barr told me about
Dr. Fenton's research into this matter in a letter dated
19th October, 1976. Later Dr. Fenton discussed it in
greater detail. My thanks to both of them.
4. "}& hiai in Pseudo-Jon, for Gen. 30:37, Hos. 4:13 - the
equivalent of the Hebrew il3l!? (see Ginsburger's edition) -
is a loan, not an independent Aramaic form.
5. For /Tin derived lexemes, see Jastrow, Dictionary, 438,
440, 452; for /^n - ibid., 464.. The same applies to
Syriac - see K. Brockelmann, Lexicon Syriacum, 223 -
\/hwr, 232 - /MJb.
6. Cf. Gordon, Textbook, Glossary: no. 862 for hlb - 'milk';
no. 1357 for lbn.II (Ibn.I in his list is the root
underlying the lexeme for LBNT = 'bricks'). Also:
Grondahl, Personennamen...aus Ugarit, 154 (for \/lbn ) , 135
(y/hlfc) .
7. Cf! E.W. Lane, Arabic-English Lexicon, II, 623.
8. CAD, IX, 8f. (labanu A. = 'make bricks'), lOf. (labanu
B. = 'prostrate').
9. F. Delitzsch, Assyrisches ttandworterbuch, 649.
10. ABWt II, 857.
11. Ibid,, I, 309; CAD, VI, 36. Gordon, Ugaritic Textbook,
glossary, no. 862, quotes a Neo-Assyrian hilpu = 'milk',
but states that this is a borrowing from West Semitic.
12. Cf. I, B.a and B.b.
13. inn A>J13 (Dan. 7:9), although of a different structure,
is the Aramaic equivalent of the Hebrew metaphor.
14. II, C.2.
15. For the identification of biblical Jl)HS see E.V. Hulse's
article in PEQ 107 (1975), and J.F.A. Sawyer's note in
VT XXVI (1976), 241-245.
232 NOTES TO PP. 82-86

16. For Ex. 16:31 see B.2.4.

17. Cf. II, B.2.3.
18. Could it be significant that 'milk' is not chosen as one
of the specifications of p!?. in this passage?
19. BH : D»?j7Jn (with LXX); inTvv. 32, 33 &»*Tj£) and dW^O
appear as"~'a pair (in v. 32 the pair appears twice, but^
perhaps the first occurrence should be omitted, cf. BH ).
20. Diez Macho, I,. 197.
21. Jastrow, WMW ,JiniW II, 1536; KftlD II, 966.
22. Cf. Jer. 13:23. T>m:mn 1031 1Yl)> *W*D Iflrpn, where the
meaning of Itt'm'nn* - whether 'spots' or 'stripes' -
depends on the identification of 1)33.
23. In his commentary, 12.
24. BOB, v'Tpy, 785.
25. Skinner, Genesis, ICC, 392.
26. BDB, IQ^, 856; Jud. 6:37 - in*n Jin - is the clear
occurrence in which IKJif retains its basic meaning.
27. Sperber's ed., Ill, 18. Similarly also in Dan. 7
KpJ "inJI-3. 10K1 13M1, again employing the literary
convention of (*py) IBS // \^0
28. The text reads: inun''''^! ^piTi "]& 1QSD 101-pfl INJin 10H
•OIIK Kn»fli...iNiin mom dniKDi (16 t"»p 'nn) ia:o ^v)
I:P^ >trcD ™o dm •mno d'OKO d^xon T»rr dK Vt nx
10^5 i^n 1 * TK ytnn:> IKO d^nnK T>IT» dKi .d^nisiy ty IS^K
.^^wn iQD nkn p!? U->KE;
29. Elijah ben Salomon Zalman, whose commentary appears in
some editions of fllfr'm mKlpQ.
30. Gradwohl (p. 57) presents the data but does not draw any
conclusions.
31. BH: delete d">^n as a gloss to the parallel d'-^ttK, and so
also for v. 6c""-(in»n VIM !>N 1N** &>Ynm). For "the
position of vv. 2-3 vis-a-vis vv. 6-7 see commentaries
and Fronzaroli's article (mentioned in section B.I.5 of
this ^chapter) and notes thereof.
32. For VQK* and Yl^* cf. II, C.I.3.
33. See, for instance, commentaries mentioned in B.I.5; and
Abrabanel's commentary for ch. 6 (Vol. IV, p. 214).
34. G.E. Post in J. Hastings' Dictionary, II, 418.
35. R.H. Charles, The Revelation of St. John ICC, Edinburgh,
1920, I, 162, with references, quotations, and a short
discussion of the Zechariah passages (pp. 162-163).
36. For a summary of various interpretations see B. Oppenheimer,
nnsr imiTn, 71-85.
37. Skinner, ICC, 392; for np3t> see under 'Indirect Terms', I
F.2.2.1.
38. So Skinner, op. cit.\ von Rad, OTL, 293; Bennett, Genesi
299: Henton Davies, Genesis, 227.
39. BDB, 822.
40. Jastrow, 1205.
41. Skinner, op. cit., regards it as superfluous - a variant
or doublet.
42. BDB, 362. In Jo. 1:7 it appears as verb formations, nut
riflKJh., and in the same verse we read iTA'liy U>}^n.
43. WniTe all instances of /"p^ are translated in the LXX by
leuke and its derivatives, in V two terms are employed,
albus and candidus. According to the Oxford Latin Diet.
(Lewis and Short, 1958) albus is "lustreless white" (81)
while candidus - "dazzling, brilliant white" (277; and
NOTES TO PP. 86-88 233

candor, 278); similarly also Simpson in the Cassell's Diet.

(1964; for albus, p. 33). However, the two lexemes are
used in such a way that an attempt to classify the two
modes of translation under "brilliance/lack of brilliance"
motivation is fruitless. Perhaps this can be attributed
to the desire to introduce variety into the translation?
44. For the construct form 111!? see G.-K., 93dd; Gradwohl, 35,
n. 6, implies that the form is built as a noun. Against
this see Blau, i:mu£ XXXII, 495. As Blau himself states
elsewhere (in his fllllifJTl n>ftfl mill), the boundaries
between noun and adjective - in Semitic languages in
general and in Hebrew in particular - are rather fluid
(197). It is therefore surprising that Gradwohl restricts
the nouns vs. adj. consideration to 13^, leaning on such a
slender morphological basis. For the'question of noun
and adj., see Appendix, II, H.
45. niK-lpn, I, 116 (Venezia edition): inUIHD dTlK Tmt>.
46. Bacher, Ibn Janah, P^enta, 152.
47. Qimhi, EPtyUi;, 248.
48o GB '(1899), 249.
49. BOB, 314.
50. KB, 297.
51. Genesis, ICC, 525, although hesitantly.
52. Genesis, 299.
53. Genesis, 110.
54. Of modern scholars, von Rad (Genesis, 415) seems to be the
only one to adhere to the "red/white" interpretation.
55. Which is the basis for the expansions of the Aramaic
Targums: see Rashi's commentary for vv. 11-12, where he
deals more with the T than with the Hebrew text; and
Ibn Ezra for D»1J)? PT: tnKJl'Kf 111^2 Pl!> inmi
drm-ipa) !?m IYT nn
56. S.M. Paul, IEJ XXV (1975), 42-44.
57. The pair MKnn//a^n appears in Ugaritic literature too:
see M.D. Cas'sutoT4n3y nfrtn, (1965), 26. In biblical
literature it signifies not only abundance, but also
lack of commodities apart from 'natural', 'wild' ones -
see Isa. 7:22 and commentaries ad loc.
58. Am. 9:13, Jo. 1:5, Song. 8:2.
59. Which is secondary to Amos 9:13 - !?D1 CPGy P»inn 13'Ufll
rmAiann nijmn, (Bewer, ICC, 141) and therefore
demonstrates even more forcibly the principle behind
the analogy in our verse (Gen. 49:12).
60. Gradwohl, 35; and opinions mentioned in nn. 8, 9, 10 and
11 thereof.
61. Dillmann, Genesis, 465.
62. Gradwohl, 4, 35.
63. Skinner, ICC, 525; and Gradwohl, op. cit.
64. So Gradwohl (p. 35), who cites the Versions and some
bibliographical material (n. 10 ibid.}; BOB, 151; GB, 140;
KB, 168.
65. Kohut, Aruch Completion (Vienna, 1926), V, 10.
66. Gray, Numbers, ICC, Edinburgh, 1903, 105.
67. Rashi, fllKlpn I, 166. Moreover, his task is made more
difficult by his definition of n^"T3, which he understands
as 'crystal', a precious stone (op.cit., 332, for Num.
(11:7). In his commentary he follows R. Assi, in
Tal.Bab.Yoma, 75a.
234 NOTES TO PP. 88-90

68. »njm H'y -O^Kl - Ibn Ezra, niK1j70 I, 166, 332.

69. Gray, op. cit., 106.
70. Ibid., 105.
71. Although cognates, with the basic sense of 'brightness',
'shine', exist for both; see, for instance BOB, |7£\a and
irina, 97. The adjectival Vfl3 (Job 37:21) is a hapax
leqomenon which appears in a difficult syntactic position -
BH ibid, proposes an emendation to a verb form. Within
its context it may be explained by nilfD = 'stains, flecks'.
In later Hebrew, however, T!l3 was taken to mean
'brilliant', even 'transparent', 'pure' - Ben Yehuda,
Diet. I, 469-470. In modern Hebrew it means either
'light-coloured1 or - especially as modifer of 'skies',
'air', etc. - 'pure'/'transparent1/'bright'.
72. BDB, 673.
73. Ibid., 705.
74. Snaith, Leviticus and Numbers, 92, presents G.R. Driver's
opinion (in DB, 575b) that, following the Arabic cognate
si'atu, fiKB should be defined as an inflamed (active)
scar, that'is, by its colour, not its texture. This,
however, is hardly necessary: IlNb might be light-coloured
(njl3t>, Lev. 13:10) or inflamed (the colour of *n icn,
ifcld.').
75. Gradwohl, 35.
76. Modifying mm and of the same status as nj^ nNiy (v. 19);
rm^ IN nmrnx ru:£ mm (v. 24); DTOTK "pt? y;u =nm«j njm
(v.'42) - the latter is established, as some kind of 13^
throughout the chapter; and similarly J1J!1^ yAJfi JlKE> "r~r
.-IKO 11 y nyis nK*inD...nmmK (v. 43).
77. Gradwohl, 57.
78. Ibid., 36.
79. Section B.2.0.2.
80. E.V. Hulse, PEQ 107 (1975), 103.
81. Ibid., 95.
82. Ibid., 103.
83. Ibid., 98.
84. Noth, Leviticus, OTL, with detailed arguments, 103.
85. For instance Y. Kaufmann, nJIOKfl on^lfl I, 114-117,
127-128; M. Haran;
M nn&IQI tllfllpn Tel Aviv, 1973, 183-200;
J.M. Grintz, D*:iro mim D*ai*rp d»n:no," UJIP^ xxxix
(1975), 20-25, 163-181; XL (1976) 5-32; M.~Weinfeld,
"^nn^iOiTi ->nD^nan iypi-«ipm iwyon", yau; IKI i, (especially
pp. 127-128, 131-132, and literature cited there).
86. Although the usage of "p^ as a blanket term denoting any
pale colour in general persisted in MH times; and see
Part III of this study, under 13!?.
87. Which is the basis for Rashi's allegorical interpretation,
niK"lptt IV, 321. See also Ibn Ezra, op. cit., for the
same verse.
88. C.D. Ginsburg, The Song of Songs and Coheleth (1861 = 1970
413-414, with some relevant biblical and extra-biblical
references.
89. Thus Gen. 41:42 (m ^TA3) and Esth. 8:15, as a symbol
Joseph's or Mordechai's changed status, the priest's
clothes, and so on.
90. See: Kassovsky, TIH^nn Tiefr "mK, VII, 16, for the
relevant references.
NOTES TO PP. 91-94 235

91. A. Hurvitz, HThR 60 (1967), 117-121, and also Grintz,

"o^oni? o^rma", mi eft xxxix, 178-iso.
92. For T&a cf. BOB, 93-94; KB, 107.
93. Gradwohl, 37.
94. Cf. II, B.2.1.1.
95. Cf. B.I.7 for a discussion of WTK*.
96. See 2.8.
97. Gradwohl, 37; ibid., 93, where both Duhm and Krauss are
mentioned (note 8). In agreement with this opinion are
the views advanced by: W.O.E. Oestlerley, The Psalms
(1939), I, 271; C.A. Briggs, The Book of Psalms, ICC, II,
4.; M. Dahood, Psalms 51-100, AB, 1970.
98. M. Buttenweiser, The Psalms. 1969 = 1938, 186-194;
L. Sabourin, The Psalms (1969), 47; and A. Weiser, The
Psalmsf OTL, 1962, 402.
99. BH3 proposes .I^WiTi c\Wt\ aflBh; a discussion of this point
can be found 'in : J. : A. Bew^r/' Joel, ICC, 1911, 79.
100. Section B.2.2.
101. So Bewer, op. cit., 77: "white gleam their branches"; 79:
"show whiteness, grow white".
102. G.A. Smith, The Book of the Twelve Prophets, 1928, II,
412: "bleached are its branches".
103. BH3 proposes to emend pWl Yl^l DHl to 1£%1 TUnn^l DJllK.
104. G.-K., 53q., although iaW is recognized as a Hif. form.
J.A. Montgomery, Daniel',:ICC, 1927, 460, leaves the
options open.
105. Sections 2.7.1 and 2.7.2.
106o Gradwohl, 38; Ben Yehuda. , V, 2612 - but this is obviously
on the basis of "pS^il usage in MH.
107. G. Bergstrasser, n>*Dyn llttfrn PI'TpT, 478 - Hithp. verbs
only rarely express a reciprocal-transitive sense when
they are related to Hif. or to Qal forms of the same verb,
rather than to Pi.
108. S. Morag, !?K^JT 100, "OKPn i^tt T"> 3H3, Jerusalem, 1973.
On the relevant plate (p. 149) the word is seen to be the
Pi. infinitive. Morag discusses it on p. 38, where he
defines the MT form as a Hif.
109. So in the Qumran War Scroll: "pl^tt linD I'm ^fin O^TOni
f8.ll), and even 1>Ktt "}& bftt, (8.10); Yadin, non^Q n!?>AQ
•nU?1M *333 11K *33, 290, relates the description to our
passage.
110. Ben Yehuda , V, 2612; Yadin, op. cat., in a note to 8.11.
111. For v/Yi:i = 'purifyj, 'select' 'polish' (Isa. 49:2) cf.
KB, 156. For /qi2 = 'smelt', 'refine' and its
derivatives: BDB, 864; KB, 817.
112. Cf. B.2.6. See also the discussion of VW in section II,
E.2.3.
113. So also Pines, Enc. Mig.r VI, 669.
114. Cf. the NT., Matt. 17:2.
115. Gradwohl, 4,34.
116. Ibid., 37.
117. BDB, rm^, 527.
118. Ibid. Even if MJ^!? is a loan word in Hebrew, it is still
generated from a ^"P^ whose denotation is 'light coloured'
the fact that this denotation no longer exists in the
source language is no proof that it has never existed.
The colour of sun-baked bricks - reddish-yellowish, very
light, although not 'white' - would fit within our
236 NOTES TO PP. 94-98

definition of 13^..
119. Berlin-Kay, Basic Colour Terms, 4, 27.
120. in EX. 24:io,...d»imi o>o>3"i i*3&i! ^^ ii^s "p^Ai nnm -
where fiJ}^ might be interpreted as 'light', among other
possible'explanations. However, because the word seems to
be a construct form of fl.^ = brick (So BOB, 527; KB, 47
has a 'flagstone', from Akk'adian libittu"), it will be
discussed under the heading 'Indirect Colour Connotations'.

B.3.
1. So BH3.
2. BOB, 1007, lists inKf in Jo. 2:2 among other inKJ. = dawn
occurrences, but says (in brackets): "al. blackness, //
^fliyi my".
3. Jastrow, 1551, 1552, 1559,
4. Gradwohl, 51.
5. Fronzaroli, 'Struttura dei Colori', 384, 38
6. BOB, /lem, 364-365.
7. Ibid., 871.
8. 0*?^ (Prov. 22:29) appears in BOB (p. 365) under the
reconstructed sub-heading (HK/n*). Although the formation
is that employed in biblical Hebrew for other colour
terms, it is not exclusive to them. This, and the
parallel D>^)3 *JD!>, exclude d*3Wh from being considered
a colour term. *•
9. Gradwohl, 51-52; in BOB, 1007, an Akk. cognate for "line/
'coal' - suru - is mentioned.
10. Gradwohl, 52, although on 59 he concedes that the usual
Aramaic translation of I'tlE/ is D31N.
T
11. Ibid., 51.
12. For *in^, and its derivatives in Aramaic, see Jastrow, as
in n. 3 supra; for WlK Jastrow, 25.
13. Tur-Sinai in Ben Yehudah, XIV, 7034-5, second colum
n. 2; also cited in Pines, op. cit., IV, 671.
14. Berlin-Kay, 4, 17.
15. The LXX to Job 30:30 has eskototai megalos to "invy;
Gradwohl, 53, says that this translation is influenced
by the rendering of lintfla ;1Ujn (Lam. 4:8). There is n
conclusive evidence for'this. However, there seems to
be a genuine understanding that 1hK> in this instance is
closer in reference to the concept of 'dark', 'become
dark', rather than to 'black', 'become black'.
16. KB, 733.
17. For the textual problems of v. 31 - the need to elide "pK -
cf. BH3 and Gradwohl, 52.
18. Jastrow, 25, 64.
19. See note 16, 3.0.5 above; Guillaumont, 341
20.. Cf. also Lam. 4:7, 8.
21. So3BH3.
22. BH proposes to emend TO DfO to TQ1 DtlD (following the
LXX), or alternatively, to T3£ dn3.
23. See Lam. 4:1 SlBi} Dton
V '.' •*•
NJW*
**' '<,"
5ilt
T» T
dW* rO»K, where ^nt and
OflD are employed as symbols of the (visible) property of
brightness, in this case unnaturally tarnished.
24. Fronzaroli, 'I Cavalli', 593-602 (English summary of the
article on 602).
NOTES TO PP. 98-101 237

25. Gradwohl, 53, states that ITl'mtuy, like a*mTK,p*lpV,

denotes a lighter colour than its basic form (einen
helleren Farbton bezeichnet, also schwarzlicti) ..Therefore,
it is somewhat surprising that he translates filing as
schwarz. The Aramaic T KIl*YTp_ (from Tip.), which he
cites, displays the same wide' range - 'dark1, 'dull
coloured', 'black' - as the Hebrew term 11DU) (Jastrow,
T
1317-1319).
26. Cf. section 3.0.4.
27. Also Qimhi, p">ETiE;, 762: dhan Kin Titiwm.
28. Gradwohl,' 51ff.
29. N. 13 above.
30. Jastrow, 1548-1549.
31. For n^\a see II, F.2.2.2.
32. Berlin-Kay, 38.

B.4.
1. BH : -l-in^. (following Theodotion, the Aramaic T, V)
instead of 1.1 n?.
2. Noth, Leviticus, OTL, 103-105.
3. M. Dahood in his commentary to this chapter, Psalms 51-100,
AB, 133-152; T. Gray, JSS XXII (1977), 1-26.
4. So E.A. Leslie, The Psalms, 73 (following Schmidt);
Weiser, The Psalms, OTL, 483; Dahood, op. cit., 133. For
a late dating: Briggs, Psalms, ICC, II, 96. For .
additional opinions see Gradwohl, 30, nn. 23, 24, 25, 26.
5. C. Gordon, Ugarific Textbook, Glossary, no. 1014.
6. CAD, A, 300; H, 246; Landsberger, 'Uber Farben', JCS
(1967), 144.
7. GB, 344; BOB, 438-439; KB, 406; Gradwohl, 27; Pines,
op. cit., 668.
8. Fronzaroli, 'Struttura dei Color!', 384, 388.
9. Albeck, roum!? tmn, 141.
10. -irun "y\a TWO jaiD rwnan fco "nipa lipv own - z. Kailai,
Enc. Mig., Ill, 889.
11. Although in KB it is derived from •VPI (406) while, if
derived from v^pl*, the toponym would seem semantically
opaque; cf. under 'Indirect Terms', II, F.I.2.4.
12. Gradwohl, 27-30.
13. Although Gradwohl, 28, differentiates between p^* and pV
(on the basis of the different construct formations),
there is no semantic reason to do so. See Blau in 13 31 lgt>,
XXXII, 405.
14. Cf. G.-K., 128r, for a definition and instances of "...
substantives ... used to convey an attributive idea in
the construct state before a partitive genetive".
15. So, for instance (for Gen. 1:30), Bennett (p. 86), Henton
Davies (129), and Spurrell (17) in their commentaries.
Ibn Ezra for this passage states clearly: '10 tOI lay ^3
.mm ^Dtn rmnt> n>>ym DTK!? ima vy in other words, he
understands 3V)y p"V>4< as an equivalent of pil^. 3K/y_.
16. Gradwohl, who admits the possibility of p1!?. = pVl^. = proper
colour term, sees the shift as that of expansion, not
narrowing, and dates it to the post-exilic period. I
think that the passages quoted do not lend themselves to
this line of interpretation.
238 NOTES TO PP. 101-108

17. Cf. BOB, 438; KB, 406.

18. BOB, op. cit.
19. KB, 403, and Gradwohl (p. 30), following BOB, KB, and
Ben Yehuda (Diet. IV, 2153).
20. Gradwohl, op. cit.
21. N. 7 above.
22. Berlin-Kay, 18; see figure 6a.
23. Pines, op. cit., 668.

B.5.

1. Berlin-Kay, 6; section II, A. below.

2. Ben Hayim, Tarbiz XII, 75-77; Moresheth, tyliffl, 188, 190.
3. Gradwohl, 23.
4. Blau, i:meft XXXII, 405.
5. Jastrow, 966.
6. Gradwohl, 23.
7. tntnpQ, 265, Berliner, 122.
8. Ginsburger, 193-194.
9. Neophyti I, vol. Ill, 81, 83.
10. Mentioned in Pines, Enc. Miq., 670.
11. Jastrow, 1282.
12. Moresheth, op. cit., and especially n. 2 (p. 190).
13. Ibid.
14. See Berlin-Kay, figure 6b on p. 19, which shows clearly
that a term whose focus is our 'yellow', when it evolves,
does so at the expense of the area previously classified
under 'red'.
15. Pines, op. cit.
16. BOB, 843f.

C_

C.I.

1. Gradwohl, 12-13, cites LXX, V, and T. Onkelos; Pines, VI,

668 lists the other Aramaic translations for {7*1171* „
2. M. Neg. 1:2.
3. IJbid.
4. Which Gradwohl (13) and Pines,665-666,do not take into
account.
5. Yer. Sukk. Ill, 53d; M. Sheb. 6:1. Also: Sif. )Tl>tt>, 14:2.
T:
6. tntnpn I, 266 (for Lev. 13:49).
7. Although both he and Ibn Ezra agree that in Ps. 68:14
P*lpT is a diminutive and not an intensive form.
8. Qimhi, tPETHg, 8: nWHTTKil SX
9. "mn ion d-nniN w»i mhinty IDI ynon^ ^ISD nr
.Cni^iTA niKipa)
10. BDB, 10.
11. KB, 13.
12. GB, 11.
13. S. Moscati, Comparative Grammar, 79.
14. See, for instance, the list in laJIBft VI, 83-87 - where
DIMTK does not feature, but the model is utilized for the
creation of other derived terms, such as 3'flS >innn^
l£ >13^> ^ ="^m).
15. Guillaumont, "Problemes", 345.
16. Whose conclusion is mentioned in Pines, op. cit., 671.
NOTES TO PP. 108-114 239

17. Gradwohl, 14.

18. Pines, op. cit., 666.
19. Z. Har-Zahav, "fltmnftt rUOpntl niTlS", IJJIEft I (1928),
127-144.
20. J. Zlotnik, "DTniK i7lp1>", 1331^ I (1928), 443-447.
21. Gradwohl (13) supplies a list of emphatic or duplicated
forms conveying the idea of 'intensity' in cognate
languages; these, however, have one double radical at the
most. As such, their evidence for our problems is limited.
22. N. Shalem, "miayi t^jnitfl ITlQEfr", 1331^ IV (1931-2), 61-67.
23. JMd., 63.
24. Berlin-Kay, 150, point out that "As a color system
introduces hue contrast, the importance of brightness does
not diminish - the system simply becomes more complex";
for as emphasized earlier on the same page, "... brightness
is a major dimension of contrast in all color systems".
From the aspect of hue 'pink' is 'virtually identical'
with the focus of 'red'. (iiid.).
25. Sections C.I.3, CU.4.
26. Cf. C.3.1.
27. For BH3, Isa. 63:1 = tntKtt > tPTKO, ?m3tt >1^30 - cf.
B . I . 4 , and notes thereto''.'" ~ rr ' "r'T
28. BOB, 10; KB, 11; Kaftan IIEft ISIK I, 32 - "dyed with siqra";
Gradwohl, 10-11; Pines, 665.
29. Pines, op. cit.
30. Ibn Ezra to Ex. 25:5 (Fleischer's ed., p. 227).
31. M, Noth, Exodus, OTL, 199.
32. So Pines, op. cit.
33. This has been suggested for other verses as well, for
instance Hos. 11:4 - ranN ninhya //... DTK ^nm (which
renders the emendation of DTK to flHK, cf. BH3, unnecessary).
My thanks to Prof. Barr for pointing this out.
34. Gradwohl, p. 11, following Hess (iMd., n. 68).
35. Noth, op. cit., 201-202; Cassuto, Exodus, 326.
36. M. Haran, "nnnn tl30i\ AlVnn :pt>n3n", 33,n. 10.
37. Jastrow, 1009; Levy, W&rterbuch II, 176-177.
38. BOB, 54f; KB, 62-63.
39. Hare'ubeni, "i:me>t> III, 134.
40. Shalem, •miEfr IV, 66.
41. B. Maisler. Bnc. Wig., I. 430-431.
42. BOB, op. cit.; KB^* (1967), I, 61.
43. So Gradwohl, who neither mentions nor discusses it; and
BOB, where it is included in I /Y&K.
44. Fronzaroli, 'I Cavalli1, 602 (English summary).
45. Mitchell, Zechariah, ICC, 178-180, 182.
46. Ibid., 129.
47. Charles, Revelation, ICC, 162-163 (for Rev. 6:2ff.).
48. Ibid., 119.
49. Ibid.
50. Jastrow, 1350.
51. Ibn Janah, tPKnuf, 38.
52. Qimhi, D>l!rm>, 42.
53. niinjm, 405.
54. KB, 63: "piebald".
55. Klpan TlUfr 121K I, 201.
56. Shalem, IJJiefr IV, 66.
57. Which is different from that cited by KB and 11^!? *mK
K-lpnn (cf. nn. 17, 18 above).
24 Notes to pp. 114-116

58. Guillaume, Hebrew and Arabic Lexicography, 7.

59. The lexeme does not feature in MH either.
60. B.Z. Eshel, 1331Bfr V (1954), 5; M. Stavvi, 1311Wt> X (1959),
22; Gradwohl, 22; KB, 932. 3
61. If we accept the second emendation proposed in BH for
this verse and read instead of the MT
•>^in Q'a'iKi jnp'nty D^nwo n^'y .IBQI
the suggested .lllh Om'NI illjnt? O'WO »"riy .lK>hl
(with IQIsaa), then a colour contrast"of OTK // "p^ -
which features in other passages, whether in a context
of vine growing (Gen. 49:12) or not (Isa. 1:18) - comes
into focus.
62. rPg'.l'TO (Isa. 16:8) has the same basic form as D*i?'lW
7zech. 1:8), but within its context ( // ISA) functions
exactly like p^p - it denotes a vine branch'j tendril, or
cluster of fruit (BOB, 977). Therefore, together with the
other lexemes mentioned above, it cannot be considered a
full-fledged colour lexeme. Although retaining some
colour allusion, or connotation, it belongs primarily to
a different semantic field.
63. BOB, 977; KB, 932-933.
64. Gradwohl, 21.
65. Ibn Janah, D?enE>, 536-537.
66. niKlfrtt III, 400.
67. Qimhi, niKlpQ, 811-812.
68. Jastrow, 1345 (for Nhntp .
69. Rashi states:Kin y^T?ln »n)TP Ktn
70. Pines VI, 671; KB, 932; BOB, 977; Leslau, Ethiopia and
South Arabic Contributions, 52 (in Ge'ez, Tigre, and
Amharic).
71. Gradwohl, 21; Pines, op. cit.
72. Gradwohl, op. cit. (and n. 142); KB, 932; BOB, 977.
73. Gradwohl and the others, op. cit.; and Jastrow, NIpJ'O II,
T
986; Herszberg, AlKft S^ym ^Ki\, 235-236. '
74. Albeck, ilJCTrt Kl^tt, 166; Moresheth, ^yiOM 11J7^DD^, 144;
Shalem, •miEfr IV,~ 67.
75. Pines, op. cit.
76. BDB, 977.
77. KB, 933.
78. Leslau, op. cit.
79. Jastrow, y/PIO III, plO, 1030.
80. 17-10: Ben Yehudah VllfJ 4228; J\Fvi): Ben Yehudah XVI, 7627-
7628.
81. Pines, op. cit.

C.2.

1. The Samaritan Pentateuch has Ifitf in both lists of

descendants of Shimeon. The list of Shimeonite families
in Num. 26 has HIT instead of Ihif (v. 13), and so does
the list in 1 Ch. 4:24; ItlS (Q), or IMS' (K) is enumerated
among the sons of Judah in this last list (1 Ch. 4:7).
2. So C.H. Parker, The Tyrian Oracles, 161; W. Eichrodt,
Ezekiel, 379; G.A. Cooke, Ezekiel, ICC, 310 (with an
emendation); BDB, 850, and KB, 801, both with a suggested
emendation to the place name; Gradwohl, 25.
3. See below, and in the next section (2.1,2).
NOTES TO PP. 116-120 241

4. For the location of IIH^D and IMS see Parker, Cooke

(cited above, n. 2) and the literature referred to thereof.
5. Thus LXX, V.
6. Rashi for our passage; Jastrow, 608 OO^jp), 773 (K^tt),
with the relevant quotations from the Mishna, Tosefta1, and
Talmud. A most concise definition is to be found in
Tal. Bab. Shabb. 54a - 101» IS ^ laiD...*!^ 1»* >
n^-m!? i'nlK i»f>pnw
T
7. Gradwohl, 24. '"' ' :~:
8. Jastrow, 1275, Onkelos for Ex. 27:16 (according to
Berliner's ed.).
9. BOB, 850.
10. G.F. Moore, Judges, ICC,148.
11. C.F. Burney, The Book of Judges, 124.
12. KB, 801.
13. Gradwohl, 27.
14. GB, 681; Pines, op. cit., VI, 670.
15. linn: Jastrow, 237: "red-spotted in the face". Cf. also
Ben Yehudah, II, 748, and nn. 2 and 3 thereof.
16. Aruch VII, 14.
17. niK1j7Q II, 70; III, 284.
18. Ibn Janah, D><no, 427.
19. Ben Yehuda. , XI, 5458.
20. Y. Kaufmann, D^oaitg IflD, 136: riinrQftt IK m^.
21. GB, 681. ":":
22. Pines, op. cit.
23. Gordon, Textbook, Glossary, no. 2160.
24. BH3: prp. f^^.
25. Gradwohl, 2.V :
26. Fronzaroli, 'Struttura dei Colori1, 377.
27. ibid., 381-382.
28. Ibid.
29. Cf. Bab. Tal. Men. 86a, 'pJ K!?N IT VK, and its similar
in meaning to y/rOT both inTbiblical Hebrew and in MH.
30. Again, probably through the similarity in form and
meaning with /MDT.
31. BOB, 269.
32. Fronzaroli, op. cit., 387.
33. KB, 255.
34. BOB, 850: KB, 800; GB, 680; Guillaume, Hebrew and Arabic
Lexicography, 14.
35. Note the parallel root ^M2 (BOB, KB ad. loc.) which in
Hebrew apparently means 'parched' (Isa. 6:13 - NOiJ flfiS),
although Arabic and Ethiopic parallels point to a basic
notion of 'be cloudless'. The similarity of reference
between this second root and betwee^ the lexemes tlinsns,
nn*ns ,n>n* - and the compounds MS DH (Isa. 18:4) and
(IS nil (Jer. 4:11) is quite clear. It is possible that
through the basic similarity of form and sense between
the two roots they have become semantically contaminated,
to the extent that the only Hebrew lexeme which is seen
to evolve directly from /*n^ is npij. Leslau,
(Contributions, p. 44) remarks that indeed the Ge'ez and
Tigre root shy seems closer to the Hebrew Vrin^ than to
V'nfi^. For liif D*n see: S. Dubdebany, "ftt tfp3", rafrtTD TOD,
334-338.
36. Pines, VI, 670.
242 NOTES TO PP. 121-125

C.3.

1. Gradwohl, 53.
2. Pines, VI, 671.
3. Ibid.
4. Jastrow, 25.
5. Ibid., 1552. Aruch, VIII, 57.
6. BH , BOB, 299 - both regard this last instance as an
interpolation and recommend its elision.
7. Fronzaroli, op. cit., 383.
8. Gradwohl, 50, where he lists cognates; see nn. 2 to 13
there for authorities cited.
9. Pines, op. cit., 667.
10. S.R. Driver, Genesis, 278; J. Skinner, Genesis, ICC,
391-392: "black or dark brown". Von Rad, Genesis, OTL,
293, 296; Gunkel, Genesis3, 339.
11. The 'black' interpretation, together with an allusion to
Arabic, is already to be found in Ibn Janah, D*gnu>, 146;
Qimhi (DjMJnw, 196) agrees - he too derives the term from
/cian".
12. Cf. II, B.3.
13. Gradwohl, 50-52; Pines, op, cit.; B. Kedar, The Vulgate,
166-167.
14. Jastrow, 1545, (but cf. his dhU? ,dhU> ,CPIW 1548-1549);
Pines, op. cit. ~"r •
15. Jastrow, 702, 'flaming, red'; ^Wti5 I = 'flame, glow', 704.
Such,interestingly enough, is the translation given by
B. Jacob in his commentary to Genesis (English Edition),
204.
16. niKlfta ad. loc.
17. As cited by M. Kasher, ilQ!>U> mi ft ttftnh VI, 1209-1210.
18. Cf. II, B.0.3.
19. Pines, op. cit., 667.

C.4.

1. Pines, 668; Gradwohl, 30.

2. Cf. discussion under aiBTK, II, C.1.1.2ff.
3. Gradwohl, 30-31; C. Levias, Jewish Enc., II, 176-177;
J. Gray, "A Cantata of the Autumn Festival", JSS XXII, 14:
"pale gold" (p. 23).
4. See the discussion of pll*., B.4 above, and notes.
5. For the appearance of the dove cf. Z. Beilin, Klpttft?a 53
(1973), 227.
6. Jastrow, 598.
7. ibid., 595-598; 750 (Kp)*i1& = 'crocus, saffron'); 839
(Kjrhn - the same). T '
8. Cf. cTl.1.6.

C.5.

1. Cf. II, B.5.

2. Jastrow, 822.
3. Rashi, niKIjTO I, 266.
4. Qimhi, d^lBf, 614.
5. Gradwohl, 23.
6. Cf. section C.4.1.
 243
NOTES TO PP. 127-131

D_.

D.I.

1. For the afformative -oni see G.-K., 86f., and Gradwohl, 14.
2. BH^ proposes Q^y or O'y} f°r the tiy, which seems a little
obscure. But se'e Qimhi,' fllKlpQ, who compares the Dy. to
O'»'yi3:a D)? D^KH (Nah.'3:12), where the meaning of Oy (as
here)'is equivalent to \. Moreover, Procksch (in Bl-r)
corrects the text of Nan'1. 3:12 to D'TlDS ptty,
3. There is no logical connection between this part of the
verse and between what precedes it. Goliath underrates
David because of his youth, not because he is *ib*JK and
good-looking. The phrase looks like an addition caused
by 16:11-12, where David is referred to as a young man
among the other D^iyj. Hence, Kittel (BH) rather hesitantly
suggests that the plvrase is an addition. The ancient
Versions, however, retain the phrase in both.
4. See the collection of references quoted in M. Kasher,
op. cit., IV, 1019-1020.
5. Skinner, Genesis, 359f.
6. Gradwohl, 14.
7. Gunkel, Genesis, 296.
8. Which is the usual V translation for 'red hair1; cf.
Kedar, The Vulgate, 168f.
9. Pines, op. cit., VI, 666.
10. or: 1)n>n dl!7;O; cf. Kasher, op. cit., 1020.
11. Ginsburger, 44.
12. Heller, Peshitta, 31.
13. Diez Macho, l) 155.
14. Ibid., in a textual note for this verse.
15. Ginsburger, 194: ""iplOD 11 »n WO».
16. Ibid., 193, 197.
17. See Shalem, "Q^Jttsn JliaKft", llJIgfr IV, 63. According to
him, pttpao - despite its context - is a colour term
meaning 'very red' = "TNJ3 tilTK; if so, it is hardly
suitable for rendering 'JlttTK.
18. Diez Macho, III (Leviticus), 79ff.
19. Gunkel, Genesis, 295f.
20. Ibid.
21. Skinner, op. cit., 359.
22. In the Jewish Encyclopedia, II, 176: 'reddish-brown'
complexion.
23. BOB, 10: 'red, ruddy'.
24. So actually Gradwohl, 14; and cf. dlOTK, C.I.I.
25. Jastrow, 99; Levy, 173;"rotlich oder hochrot."
26. Cf. II, C.I.1.2.
27. Pines, op. cit.
28. See II, B.2.3. and nn. 2-4 thereof. In addition, cf.
Jacob, Genesis, 331f.: "redder than wine"; von Rad,
Genesis, 415.
29. Cf. II, B.2.3, and nn. 5-11 thereof.
30. Gunkel, op. cit.
31. Jewish Enc., 175b.
32. C.H. Toy, Proverbs, ICC, 441.
33. Heller, 64 (for Gen. 49:12).
34. Jastrow, 411; Levy, I, 564, who cites Payne Smith, 1154,
for the Syriac.
244 NOTES TO PP. 131-133

35. Contra Pines, op. cit., 667.

36. For the MT VnnJi. and other suggestions, cf. Dahood, Psalms
II, AB, 146; BtP for this verse; Gray, JSS XXII, 15.
37. A discussion of Isa. 63:lff. and the imagery thereof was
undertaken in section II, B.I.4.
38. Gradwohl, 22-23; Pines, op. cit., and authorities quoted
by both.
39. Jastrow, /V'ttH, 478f.
M Knn*fl ,Kri:m'>n,457f.
r
40. cf. M.D. cassuto, n"o p^!?nti",' m3yi3 irnfl&i mtnpa tii-iao,
188-189.
41. Gradwohl, 27.
42. For the form mninn see N.H. Tur-Sinai, Job, 268; S.
Driver, G.B. Gray, Job II, ICC, 107-108.
43. Gradwohl, 16-18, 27, and material cited in the notes
thereof.
44. BOB, 330-331; KB, 312-313.
45. Cf. Moresheth, fryian Tlp>&3^, 63, text and n. 5.
46. Ibn Janah, tMnw, 160-161; Qimhi, P*Enu?, 217-218. 3
47. So Dt. 32:14; Isa. 27:2, and mainly Ps. 75:9, wher
proposes "inf} *p» for the ion 1?1 of the MT. In the last
instance, however, the words that follow - fPinB IK.. .*|0n Kfo-
point to the 'fermenting' element rather than the 'Colour*
element; and see Diet, and Gradwohl, op. cit., for the
classification of Ittt].
48. Contra Tur-Sinai, opl cit., who cites the Lam. instances
and explains them (together with our passage) by analogy
to the Arabic cognate as from 'boil, seeth'.
49. M. Pope, Job, AB, 115. Similarly NEB for this verse.
50. Cf. Driver-Gray, op. cit., 108.

D.2
1. BOB, 301; KB, 283f.; Gradwohl, 48f.; Jastrow, 438f.
2. Jastrow, op. cit., cf. modern Hebrew l^n = "pale, light
coloured".
3. Aboth 3:11, and more - cf. Ben Yehudah V, 2612; Ke
The Vulgate, 169.
4. E.Z. Melamed, "rPBIKn JiyDW, frrm IflD, 151-152.
5. Gradwohl, 49.
6. For a fuller treatment of lin and other lexemes listed in
the following paragraph, see below, E.2.3.2 and F.I.1.2.
7. BOB, 301; KB, 283f.
8. BOB, 301.
9. KB, 333.
10. Qimhi, tPiyw, 197, links the derived lexeme to the colour
term through the practice of wearing white clothes for
festive occasions (cf. Qoh. 9:8), obviously out of the
question for the underpriviledged, at least as a daily
practice.
11. For the phenomenon of temporary loss of original lexical
items, and their subsequent re-introduction into the lexis
through the secondary influence of a cognate language
(especially in connection with the parallel pairs of
poetic diction), cf. A. Hurvitzj 11gfr.fr 1.1 eft "P3, 27-28.
12. Ibid.
NOTES TO PP. 134-140 245

D.3.

1. Cf. section II, B.3 especially 3.0, 3.4 and 3.6.

2. Ibid.
3. Levias, Jewish Encyclopedia II, 175b, who adds the usage of
1/1KK1 in Ps. 69:24 - niKIB drPJ'y nJ^ntt - and in Lam. 5:17 -
imiy l^tj - to our instance. I "think" that the last two
are idiomatic; as such, they differ from Lam. 4:8. And see
below under MHD.
4. The form firj3, which appears as a syntagm with yAjlj] (vv. 6,
56) is classified by BDB, for example, as a feminine form
derived from MflS. This seems unsatisfactory, for )?AJ is
grammatically masc., and it would be a bit far-fetched to
assume that the more frequent employment of np±> with nifta
(sg. or pi.) has influenced the pointing of Mfl3 (y*5i\) .
Rather than classify )>AJ = "mark" as of indeterminate
gender (BDB, 619), we might assume that the lexeme was
pointed as if it were a verb in the Pi. theme - especially
since it denotes a process rather than a state. Furthermore
within the framework of the stipulation prescribed in our
chapter, the lesion/mark should not sprea.d (MKfO N^, 6, 56;
TJ>)?3 *Tfly, 5), but must change its appearance (MMD) in
order to be re-classified as ritually pure. Alternately,
perhaps there is a phonetic/alliterative influence carried
over from hbflCNtO to MH3.
5. In contradistinction to the preceding instances, this last
one refers to an organic change in materials such as cloth
and leather whose chief symptom is a change in colour,
and not to psoriasis or other (human) skin disorders.
6. For the second member of the parallelism in Job 17:7, which
is textually corrupt, see: Tur-Sinai, Job, 278-279; Pope,
Job, 127.
7. BH3: Vi*P.
8. BDB, 462 (with Syriac and Mandaic etymologies).
9. ibid., and therefore classified under I i/fifO. Admittedly,
it is closer in meaning to the Arabic etymology suggested
than all other occurrences.
10. Ibid.; KB, 424.
11. BDB, op. cit., Ibn Janah, CP£TW, 212-213.
12. In BDB, op. cit., the distinction is mentioned, but no
separate entry is accorded, although this seems to be
warranted by the lexical material.
13. Cf. section I, A.I.

i-
E.O

1. Gradwohl, 60-88.
2. L.B. Jensen, JNES XXII (1963), 108ff.
3. Abrahams et al., 183-191; in Y. Yadin. *D WP.. .tPKSQQn
K3D13, 278ff. (English).

E.I

1. For the archaeological evidence: Gradwohl, 60-61; Jensen,

JNES XXII, 104-118 (for the northern coastal strip and the
Phoenicians).
246 NOTES TO PP. 140-141

2. Cf.: Herszberg, >1Kn IT'gyrn AlKfl, 210£

3. Yadin, op. cit., 178ff.
4. Ex. 31:2, 35:30, 36:Iff.
5. For »;n3 (<Ci1{O3) cf. BDB, 80
6. Herszberg, op. cit., 210, n.
7. Herszberg, ibid., derives 1111 from 'TIT1, = 'pot, jar1
for boiling or immersing the dye/dyed cloth; see, howeve
BDB, 187, where the name is taken to mean 'his beloved
together with -IMVm ,11;T.
8. Although the activity of y^in is centred around 'Shamir
(within the nt!ttn-D*13K in territory), north-west to Shechem.
9. I think that Herszberg's view, 213-218, that the Northern
Israelite tribes competed with their Canaanite neighbours
and that they became almost as expert as the latter in
dyeing, to the point that they were even identified with
them in some passages - is a bit far-fetched. Herszberg
attempts to overcome the contradiction between 1 Ki. 7:
13-14 and 2 Ch. 2:12-13 concerning the craftsman 01*n and
his skills by accepting the 1 Ki. evidence for the man's
origin, that of 2 Ch. for his technical prowess (ibid.,
212, n. 1.) This is probably a tendentious interpretation,
dicatated by his conviction that the Israelites could
produce purple, and that linguistic gaps or lacks do not
necessarily indicate lack of perception/skill (220). Our
view is that the evidence of 1 Ki., for lack of any other,
should be accepted both for the man's origin and for his
skills; hence, he is excluded from the present discussion.
10. For example: A. Hurvitz, HThR 60 (1967), 117-121.
11. Herszberg, op. cit., 221.
12. Contra Gradwohl, 61 f.
13. Cf. BDB, I tfy$X and III ^ySX, 840 for the etymology,
albeit not for the meaning; ibid., I i/Jtttf (Aram.), 1109;
but mainly Tur-Sinai's treatment in Ijncft XIII, 21f. Cf.
also Blau's critique of Gradwohl's work, 1J3ia!? XXXII
(1968), 405.
14. The emendation of Job 38:14 - lasWl to jnttll or JttOStfn
(so BH3 and Gradwohl, 62), if accepted, should be
considered an Aramaism which is related to the tfyZX of
Dan. (ch. 4, 5) but not to Wy^y VlV. Cf., however,
Tur-Sinai's commentary in Jo2j,"T526.
15. 11»D^1 of Josh. 9:4 is an obvious misspelling for •II'DS'I -
cf. BH3; BDB, 851. Otherwise there is no occurrence of
V'llS (Jastrow, 1270, 1272) in one or more of the
specialized senses - 'embroider1, 'engrave', 'paint',
'sculpt' - which is a development from rm^/TlS that is
peculiar to MH and facilitates Midrashim and puns. See
Jastrow for 1-m ,N11»S and 1(*)(»¥ (1275-1276), with
examples. • -r •
16. Gradwohl, 63-65, following KB, 661; and Lb'w (cited ibid.
17. See M. Nidd. 9:6 -
miui inai o»!?;n >ni...!?£jn pi .-Dion ty •pvnya •paaao nyw
...ysx fit Hi isy K!n vanao nysv i>!?y i>nyn.... A^OKI K^jinp
.orj3 nr »in rime; IK lay
The contrast is between dttJ and JttS, wh'ile 1>J»OO are the
agents for clarifying the situation. Further, in M. Shabb.
12:4 DO appears in a (recurrent) list of writing materials,
with 171 ,K1f7*D ,Dltt1i7 .OlttlpJp. All in all, the term is
quite versatile and far from specific; cf. Jastrow, DO (998)
 247
NOTES TO PP. 141-145

KKOnO (1002); and III, D.5.

18. Blau, op. cit., 406.
19. Herszberg, op. cit., 217f., n. 3.
20. BOB, 460, 461.
21. Curiously enough, in MH the verb is always constructed
after the Pi. conjugation, while the noun - 'OH13' - is
in the Qal part, formation.
22. Cf. Levias, Jewish Enc. II, 174.
23. The distribution of flK*in in reference to 'bodily appearance1,
'shape', 'image', 'outline' is of course much wider.
24. See III, E. and provisionally Levias, op. cit.

E.2.

1. Although the word context of the latter - ntl>lb KT»J1 K^

tP3W 8Qi> nma ^O >3 tfm - favours the pointing 0*3B =
'two1 (layers of clothing). Otherwise, the mixed metaphor
could be seen as a free association based on the DIN/ "p^
T TT
contrast of Isa. 1:18.
2. Jensen, 'Royal Purple', JNES XXII (1963), 111; and Y. Feliks,
tm»EJn VW, 23. Feliks lists three kinds of the Shield
Louse which can be found in Syro-Palestine: Kermes biblicus;
Keriaes nahalali; Kermes greeni. (id., 15); further details
are supplied in his article "np'Jiyn Uniiaon "p-mm Ti^Kh",
>3*D XIX. Cf. also Gradwohl, 73f.; A. Lucas, Ancient
Egyptian Materials, 150-154; Y. Yisraeli, Enc. Miq., IV,
1009-1010.
3. Gradwohl, 77, following Bertholet and Baentsch.
4. But cf. Ben Sira, 45:17, A11K tvoyti Jiy!nn >3tp, in a
reference to Ex. 28, where the regular idiom is ">3KJ flyVlfl
(Segal's ed., 311).
5. Ill, D.2.1.
6. Temple Scroll, 10:10, 14; [Yadin's ed. vol. I.I
7. For the etymology of t»013 see C. Rabin, "nmin D»t>a",
Dyt> 133IB)!? XIV (1963), 241f. Rabin postulates Sanskrit
provenance together with a borrowing into Hebrew through
Persian. Similarly, also Pines, op. cit., 669.
8. Gradwohl, 73: BOB, »31U, 1040, for Arabic and Ethiopic
cognates; KB, 997f. 'r
9. Gradwohl, 74.
10. Blau, 1331^ XXXII, 406.
11. B. Landsberger, 'Uber Farben', 169.
12. Ibid.
13. K.R. Veenhof, Aspects of Old Assyrian Trade, in a list of
designations for coloured textiles, 186ff.
14. BOB, 1068f.; KB, 1021.
15. Feliks, op. cit., 22, 109.
16. Ibid.
17. Herszberg, op. cit., 253.
18. Feliks, 77.
19. 1flp"i ITOin n!?3D re-appears in Song. 6:7 as an independent
unit, which strengthens its non-colour interpretation
vis-a-vis '3BH Uin, 4:3.
20. Ex. 25:4; 26:1, 31, 36; 27:16; 28:5, 6, 8, 15, 33; 35:6, 23,
25, 35; 36:8, 35, 37; 38:18, 23; 39:1, 2, 3, 24, 29. Also:
2 Ch. 3:14.
21. But in 2 Ch. 3:14 - ym ^tTDi inVttO tfon, and similarly
in 2:13.
248 NOTES TO PP. 145-149

22. See, for instance, Num. Rabb. 12, especially: Tfl^n K!>N
.na^ t0K'minis WK npiv KJK nan* WN it»yttt nmt? rv'ipn nxinu;
23. C£. Ezek. 27:7a, nnp.*):l «W, and below, for Tur-Sinai's
etymological equation: HOp/l = 1»>1N. The second part of
the same verse, however, has 113A1K1 n^Dfl.
24. See discussion of GDIS and 1-1 ft, F.2.3.
25. BOB, 1067; KB, 1028; Gradwohl, 66.
26. Cf. A. Goetze, 'Inventory1, JCS X, esp. 35ff. For thi
article, and the subject itself, cf. Landsberger, 'Uber
Farben', 163ff.
27. Ibid., 164.
28. Jensen, 'Royal Purple', 113f.; Gradwohl, op. cit.
29. Apart from 2 Ch. 2:13: . . .ttoSSQ 1&>11U
30. Jensen, op. cit., 114.
31. Cf. UfiTI 1DD in earlier sources vs. the reverse order SMT
HODI, more prevalent in post-exilic OT sources and
contemporary extra-biblical documents; see Hurvitz,
"''JIIDN'n DIOTK^", 11T>!> IflO, 248-251; 11^ licfr *p3, 146f
32. Jensen, op. cit., 111.
33. Ibid. Also Herszberg, op. cit., 266; and cf. Song, 7:6,
im-IK/Y^niX
34. Jensen, op. cit., 115.
35. Idem, 105f.; Feliks, op. cit., 22.
36. Unlike >3U; ,ytnn and ^tt"D.
37. Jensen's article (JNES XXII, 104-118) is perhaps the most
definitive; see also Gradwohl, 65ff.; M. Ellenbogen, Foreign
Words, 38f.; Loewenstamm, Enc. Miq. I, 529f.; Yisraeli,
Enc. Kid. IV. 1008-1010.
38. Ellenbogen, op. cit.; Rabin, "n'"Q)a rmnn CP^O", 156.
39. Loewenstamm, op. cit.; Rabin, op. cit., 156-157; idem,
"rm-Tin o^a", oy!? i33ie>!? xiv, 242.
40. For the same question - the etymology of 1H^1K/T1>*1K - cf.
also S.A. Kaufman, Akkadian Influence on Aramaic, 35f.,
and literature cited in n. 27 thereof.
41. Gordon, Textbook, Glossary, no. 340 (p. 365).
42. Goetze, op. cit., 32-38; Landsberger, op. cit., 155ff., and
summarized in the synoptic table on p. 164; F. Thureau-Dangi
Syria XV, 137ff. (Ugarit); Veenhof, Trade, 166ff.; Gradwohl,
65ff.; M. Elat, Klpan m^lK 1*3 tftt >1<yp, 87-97.
43. Rabin, op. cit., 157; Tur-Sinai, IJJia!? XIII, 19-23.
44. For additional notes on nnt7l, V/Qpl, see 2.4.2.
45. Ibn Janah, WVU, 54: XJUIN
TT;
rpniN:n 1U1K niMK 11ta!n 1J3A1K
.DTTK yni Kim
46. 10:12 (Yadin's Ed. II, 32).
47. Jensen, op. cit., 111.
48. Cf. II, B.2.6. and III, D.2.3.
49. A. Hurvitz, Rev. Bib. 81 (1974), 33ff.; HThR 60 (1967),
117-121. More recently see also Grintz, 1331K>t? XXXIX (1975),
179-180.
50. Grintz, op. cit.; Ellenbogen, Foreign Words, 164.
51. Hurvitz, op. cit.
52. Cf. KB3, I, lllb; T.O. Lambdin, JAOS 73 (1953), 14
53. BOB, 101; KB, 114; Gradwohl, 49f.
54. Landsberger, "liber Farben', 141ff.
55. Veenhof, Trade, 189.
56. Grintz, op. cit., 13-15.
57. Thus the LXX has the rendering byssos. For the V, the
Aram., and some etymological notes cf. Gradwohl, 49.
NOTES TO PP. 149-151 249

58. See under Indirect Colour Allusions, I, F.I.1.2.

59. Noth (Personennamen, 221} does not relate the proper name
to our /TIH; cf. also KB3, I, 287a for other interpretations.
60. Ellenbogen, Foreign Words, 94; Rabin, op. cit., 240;
Kutscher, irPniTtnni tP^Q 98-99, 117. M. FrankeJ^on the
other hand, attempts to derive from the Hebrew ^033, /b!l!5,
with a dissimilatory 1: 'Bemerkungen', HUCA XXXI, 84f.;
his explanation is not convincing, since the practice of
borrowing the term referring to a new product together with
the introduction of the foreign product cannot be ignored.
61. Kutscher, op. cit., 98.
62. Rabin, op. cit.
63. For the dropping of the lexeme in MH, the homonymic clash
with 0013 - 'green herbs', and the penetration of \/ktn
derived terms referring to this material into Eastern and
European languages cf. Kutscher, op. cit., 98-100, 117.
64. According to the Arabic cognate - "barama, twist a rope of
two strands; i>arimun - rope (or fabric) of two shades or
colours" (BOB, 140).
65. Veenhof, op. cit.,186.
66. Landsberger, "Farben", 160f., and especially the critique
of the entry in CAD (B, 257) in note 106 thereof.
67. Cf. the instance from Nuzi cited in Landsberger, op. cit.,
157, col. II, where birim is mentioned again in conjunction
with 'purple'.
68. BOB, 955; KB, 909f.; G.W. Thatcher, DB, 458.
69. Thatcher, op. cit.
70. If the basic reference of v'QP'i is to the notion '(be of)
many colours', there is no need to define the usage of
nnm in Ezek. 17:3, 1 Ch. 29:2 as figurative (thus BOB
ad loc.) just because in the OTfiOft'l,t}j7S usually modify
'cloth of various kinds and hues'.
71. Di?;i: a. Midianite King (Num. 31:8, Josh. 13:21); b. son
of'Hebron (1 Ch. 2:43f.); c. of Gilead (1 Ch. 7:16).
72. In Benjamin (Josh. 18:27); cf. KB, op. cit., for location.
73. Once also a Pu. (Ps. 139:15) in the sense of 'I was woven'?
74. Tur-Sinai, 1J31U)^ XIII, 19-23, ties up n&fp and in>1K
(cf. section E.2.1 above). The suggestion,' though
attractive, is far from certain. Further, it has no
bearing upon the understanding of biblical passages
containing Ofcl and MOpp, - apart from, possibly, Ezek. 26:16,
where MBj7*l >*T>3 are the garments of 0'fi 'K^WJ - royal
purple garments?; and Ps. 45:15 (a 'royal' circumstantial
context, and see v. 13, IS D3).

E.3.

1. Both tvlfln lip (Job 42:14) and rpfl ^N (Isa. 54:11,

1 Ch.*29:2) are assigned to Ilj F. ("Colour Allusions")
below. Cf. also Rabin, Bhc. tyig. VI, 442,.
2. Gradwohl, 80f. and 88.
3. Rabin, op. cit.
4. Lucas, Ancient Egyptian Materials, 80ff.
5. BDB, 806, for 1-19, but with no satisfactory etymology or
cognates.
6. Rabin, op. cit., following Thompson.
7. KB, 754, YI3.
8. Gradwohl, 82; BDB, 471; KB, 430; Rabin, op. cit.
250 NOTES TO PP. 151-156

9. Ibn Janah, D">¥THa, 397; Qimhi, D*BTHg, 577.

10. BDB, 974; Ibn Janah, 534; Qimhi, 808.
11. Gradwohl, 84.
12. Cf. Ill, D.3.1.
13. Cf. Gradwohl, op. cit.} Jastrow, 986.
14. Lucas, op. cit., 84.

E.4.

1. Herszberg, op. cit., 251ff.; Yisraeli, Enc. .Mia. IV, 1010.1.

2. Gradwohl, 78ff.; Lucas, 150-154.
3. Cf. II, E.I.
4. BDB, 499; KB, 453; Gradwohl, 79; Guillaume, Lexicography,
19; M. Zohary, Enc. Mig. IV, 230-231.
5. Ibid.
6. Ibid., 312.313.
7. KB, 455; Ellenbogen, Foreign Words, 93; Zohary, ibid.; Rabin,
"D'^tt", 239. Contra the definition as a loan, see Fraenkel,
'Bemerkungen', 96f. The latter's view, however, is not
convincing.
8. Yadin, D^KSna, 181; contra Forbes, IV, 122 (quoted by
Gradwohl, 80).
9. Gradwohl, 85; BDB, 1059; KB, 1014.
10. Lucas, op. cit., 343.
11. Herszberg, op. cit., 261f.

E.5.

1. Lucas, op. cit., 338-359. For Mesopotamian practices see

H. Frankfort, Art and Architecture, 95.
2. Lucas, op. cit., 362ff.
3. Gradwohl, 83; BDB, 188.
4. Gradwohl, 85.
5. KB, 930.
6. Gradwohl, 85-86.

E.6.

1. Cf. M. Neg. 1:1: ^DTin T>0.

2. D. Parnas, Enc. Mig. II, 484-485.
3. See III, D.6.
4. Cf. previous note; and Jastrow, 961, 978.
5. Contra Gradwohl, 86-87.

£.

F.O

1. Cf. II, A.5.

F.I

2. Gradwohl, 15.
3. Y. Zakovitch, DB *E?Tia >03 (Unpublished dissertation,
Jerusalem, 1971), 42, 72-73.
4. BDB, 10.
5. Ugaritic °bd.'dm.
NOTES TO PP. 156-158 251

6. See Eusebius, Onomastikon, no. 521. (all references to the

Onomastikon given forthwith will follow E.Z. Melamed's
Hebrew edition 01 *at)3K> Tl|7*DDaiSIKH 100, Jerusalem, 1966.
7. II, C.I.3.1.
8. II, D.I.3.
9. BDB, 314.
10. For |/"mn - Gradwohl, 17f.; II, D.I. 5.
11. Thus Ullendorff; cf, F.2. 1.2 below.
12. Zakovitch, op. cit., 28f.
13. Gradwohl, 73ff.
14. 'Uber Farben1, 169; cf. section II, E.2.1.3.
15. Gen. 36:13, 17; 1 Ch. 1:37.
16. Gen. 36:33; 1 Ch. 1:44.
17. Num. 26:13; 1 Ch. 4:24.
18. 1 Ch. 6:6, 26.
19. 2 Ch. 14:8.
20. Num. 26:13, 20; Josh. 7:17; 1 Ch. 27:11, 13, and probably
v. 8 (MT; *mnn)
21. I. Ezr. 7:4; 1 Ch. 5:32 (twice), 6:36, II. Ezr. 8:4
(rpmr in ^yiM^K).
22. I. 1 Ch. 7:3 (twice). II. Neh. 12:42.
23. Gen. 2:12; Ex. 25:7, 28:20.
24. BDB, 995; KB, 950; Landsberger, 'Farben', 151f.
25. II, E . I . I and 4.1.
26. Gen. 23:8-17, 25:9, 49:29, 50:13.
27. Josh. 15:19, perhaps also 2 Ch. 13:19 (K Iliay, Q l>13y),
2 Sam. 13:23 (MT: 0*-|2K) .
28. I Gen. 25:4; 1 Ch. 1:33. II. 1 Ch. 4:17. III. 1 Ch. 5:24.
29. Cf. BDB, 780.
30. KB, 723f.
31. But not JVlOiy = 'lead' - MH "QN, Aram. K*nN, Akk. abaru -
which is probably a foreign culture word (BDB, 780).
32. KB, 472; Gradwohl, 45f.; Loewenstamm, Enc. ?M±q. IV, 421-.
33. Zakovitch, 67.
34. Gen. 6:17; Num. 3:18; 1 Ch. 6:2, 5, 14; \J^iT_ - Num. 3: 21,
26:58. ':'
35. J. Liver, rPKIptt S3K IV, 430.
36. I Gen. 23:8, 25:9. II. Gen. 46:10; Ex. 6:15. III. 1 Ch.
4:7 (Q).
37. II, C.2.1.
38. BDB, 301.
39. 1 Ch. 5:14.
40. 1 Ch. 11:32.
41. Gen. 38:1, 2.
42. II, D.2.1.
43. Jud. 7:25, 8:3; Isa. 10:26; Ps. 83:12.
44. BDB, 788; KB, 733.
45. BDB, 871, for a list and literature.
46. For /YTp, 1-119 and 1-llKfl, see section F.2.3.2.
47. I. Num. 31:8;TJosh. 13:31. II. 1 Ch. 2:43, 44. III. 1 Ch.
7:16.
48. Josh. 18:27.
49. Cf. II, E.2.4.2-3.
50. Eusebius, Onomastikon, has two entries: under A for
Josh. 15:7, under E for 18:17. St. Jerome adds a "Midrash"
based on the equation DT*$ - tn, that is, one that is more
suitable to the Arabic name (with a reference to the Good
Samaritan story, Luke 10:30). Cf. Onomastikon, no. 70 and
252 NOTES TO PP. 158-161

431, pp. 10 and 42 respectively in Melamed's edition.

51. Onomastikon, no. 861, 869.
52. Isa. 5:2; Jer. 2:21.
53. Gradwohl, 21, n. 146; Eshel, DyV IJmw!? V- (1954), 5; and
p'lb*, II, C . I . 4 .
54. Eshel, op.cit.
55. Gen. 36:36 = 1 Ch. 1:47; Onomastikon, no. 652.
56. Onomastikon, no. 446; BOB, 314. Identification: Hirbet
Keila, east of Beit Gubrin.
57. ii, D.I.3 (^fcjn.mWan) and F.I.i.i (^ton).
58. See n.pr.m. Iliay, l i l . l above. "*: :
59. I Josh. 18:23 = 1113)?, I Sam. 13:17, II. Jud. 6:11, 24;
8:27, 32; 9:5. Cf. Y. Aharoni, n*Tlt>pn rpaiA1K>;\,226f.
60. Josh. 10:29ff., 21:10, and more; Onomastikon, no. 630;
Gradwohl, 47.
61. Num. 33:20, 21; Onomastikon, no. 627, 628; Aharoni, op. cit.,
37 and 135, identifies with Sheih a-Zuweid.
62. Jud. 21:19. Cf. BOB, 526; Gradwohl, 46f.
63. Onomastikon, 646; Gradwohl, 46.
64. BOB, 1009.
65. Aharoni, op. cit.f 222f.
66. A. Cohen, "...JDttl DTIO'IS D">1f1K H1tt12>", KIpQ n>3 (1973), 420.
67. Cf. Gradwohl, 46f., for the above mentioned place names.
68. Gradwohl, 24.
69. Onomastikon, no. 712: Miletos in Asia Minor; cf. note
pertaining to the same entry, ibid.
70. Dt. 11:29, 27:4, 13; Josh. 8:30, 33.
71. Gen. 10:28 (^liy) = 1 Ch. 1:22; Gen. 36:23 = 1 Ch. 1:40.
72. P. Fronzaroli, 'West Semitic Toponymy in Northern Syria
in the Third Millenium B.C.', JSS XXII (1977), 145-146.
73. Aharoni, op. cit., 14, 26, and passim.
74. BDB, 301.
75. 2 Sam. 15:23; 1 Ki. 2:37, 15:13 (= 1 Ch. 15:16); 2 Ki. 23:4,
6, 12; 2 Ch. 29:16, 30:14; Jer. 31:40.
76. 1.1.3.
77. BDB, 871.
78. Aharoni, 92; for a summary of location and occurrences,
cf. ibid., 37, 57, 104 and passim, (see Index, 367).
79. II, B.4.2. Support for this hypothesis may be derived
from the fact that another 'colour' sequence - l^!? - is
the ultimate origin of the god-name Labban. Therefore,
even if the d^ is the theophoric element, it has little
bearing on our interpretation.
80. Num. 31:8; Josh. 13:21, 18:27; 1 Ch. 2:43-44, 7:16.
81. Aharoni, 210; Onomastikon, 771.
82. Num. 32:3 (rtlBa), 36 (mm ms); Josh. 13:27 (''J IPS).
83. Isa. 15:6; Jer.' 48:34.
84. II, G.2.3.
85. BDB, 649, and also Aharoni, 90; see the latter for the
history of the names and the identification of the various
places, 77, 100, 105, 178.
86. BDB, ibid.; KB, different for maa and D*im, 618f.
T' ' ' ' '

F.2

1. See also n»TK = 'red clay1, M. Shabb. 8:5.

2. For a summary of this subject see Zakovitch, DE> *B>Tra, 27,
69f., and notes to the latter, 220f. Also M. Ben Yashar,
NOTES TO PP. 161-164 253

"nttnNI dIK DY', Dn 1DQ, 112-119.

3. Gradwohl, 4£.
4. KB, 13.
5. F.I. 1.1 (may).
6. Bauer-Leander, Grawmatik, 577. Ben Yehudah, I, 66.
7. Ullendorff, Ethiopia, 126; cf. II, B.I.6 (to Song. 5:10,
onto ns nit).
8. C£. Gradwohl, 21f. for the form, which contains a double
reading.
9. F.I.2.1, and notes thereto.
10. Cf. Iflinn*, II, D.I.5.
11. Gradwohl, 18; BDB, 330; KB, 312.
12. Ibn Janah, D^BHO, 160.
13. Qimhi, EPETlE), 217.
14. Jewish Enc. II, 176 (for Ps. 75:9, Job 16:16).
15. Ullendorff, 'Is Biblical Hebrew a Language1, 186.
16. Gradwohl, 19f.
17. Isa. 24:23, 30:26; Song. 6:10. Gradwohl, 38f., tries to
substitute ru^/nmn for nan^/nan in the first instance
(following LXXJ. THis attempt is not convincing.
18. For details cf. Gradwohl, 38-45, and dictionaries ad. loc.
For M33^ and M335 add cognates given in Leslau, Ethiopia
and South Arabic Contributions, 28.
19. Dt. 22:6 (twice); Isa. 10:14; Isa. 59:5 (twice); Job 39:14.
20. II, E.2.3.1; Gradwohl, 49f.
21. II, C.2.2.
22. 1 Ki. 7:19, 22, 26; Hos. 14:7; Song. 2:2, 16; 4:5; 5:13;
6:2, 3; 7:3; Zech. 4:5 and in titles of psalms 45, 60, 61,
and 80.
23. song. 2:1, -pion rtonn -ON
.tPpnyn ruwiiy
24. BDB, 1004.
.25. II, C.2.1.
26. Feliks, D'Ton Tty, 28-30; *K1j7ttn hOISM Dtny, 234-244.
27. BDB, 966.
28. II, B.3.0.1-2.
29. II, B . 3 . 0 . 4 , 3.4.
30. The proposed BH3 emendation, tljiwril. for pWfllll, limits the
metaphor and the picture it conveys unnecessarily.
31. Jer. 8:21, 14:2; Job 5:11.
32. BH: non:i for nnh; cf. two verses infra, v. 30 - 1hK> »1iy
'_^)W. If we accept the emendation, the field becomes
considerably poorer.
33. For y"inty, y"liyn, VYT17, cf. also II, B.3.0 above.
34. BDB, 485. Alternately, the lexeme should be analysed as
5 + 110; so Prof. Barr in a letter. His article 'Questions
Disputees d'A.T.' which deals with this lexeme (and niQ^if,
see below) was not available to me.
35. Job 3:5; 10:21, 22; 12:22; 16:16; 24:17; 28:3; 34:22; 38:17.
36. Ps. 23:4; 44:20; 107:10, 14.
37. Isa. 9:1; Jer. 2:6; 13:16; Am. 5:8.
38. Barker, 'The Value of Ugaritic', Bibl. Sacra 133 (1976), 122.
39. S.M. Paul, "nitt^", gftc. Miff. VI, 73S-736; but cf. n. 33
above.
40. Gradwohl, 25-27.
41. II, B.4.1-3.
42. Dt. 28:22; 8:37 = 1 Ch. 6t28. Am. 4;9.; Hag. 2:17.
254 NOTES TO PP. 164-166

43. Jer. 30:6 - llpl^ D'Jfl ft IDOittl. Cf. MH d>;j3 +

pmn,D»:»fl + q*DDn, III, B . 2 . 2 ; 4.2; C . 4 . 2 .
44. Tal.Bab.Taan. 19a, 21a.
45. No colour terra; rather, it refers to 'blasting, burn' —>
'dryness 1 , 'emptiness' of the ears of grain, cf. tfc\1V>
Qal, MS-TO - Gradwohl, llpl*, 31f.; BOB, 995; KB, 951;
Ibn Janah, DUJTHg, 506. T
46. E. Hare'ubeiii, ".. .tnOif fllttM dnpntt", jai^ II (1929) 176-183.
47. Gradwohl, 33, n. 50, counts 18 times and does not mention
the additional biblical Aramaic occurrence, Dan. 4:1.
48. Ibid.: "13yi ... ist keine Farbbezeichnung (gegen Galling
... "grtin"; Brockelmann ... GB") . Cf. there for the
prevalent contrary view.
49. S. Morag. "njn hlTKD tnyttrt", V^ltl XLI (1972), 17-19.

F.3

1. General bibliography: Lucas, Ancient Egyptian Materials,

231ff. (minerals), 386-405 (precious and semi-precious
stones); Landsberger, 'Farben', 150-155 ("Farbige Steine");
Ellenbogen, Foreign Words (various entries); N. Shalem,
"maibn Q^nK.i maw!?", lajicfr in, 291-229; idem,
"n»inyn d'ynsn niau^", "mngy iv, 61-67; c. Rabin,
"nmjtt m»fin d^»", fit) ISO. 1965, 151-179; idem,
"nn:Wl nmifl D»t>n", Dy!> 13J1U>!? XIV, 232-245; S. Loewenstamm,
"mp» PK", Enc. Miq. ~I, 48-49 and bibliography cited
there.
2. Loewenstamm, op. cit., 48; Shalem, "d^SKM mow", 291f.
3. Loewenstamm, op. cit.
4. Cited with no critical treatment, in Shalem, "tpyssn nittKft?",
66f.
5. Shalem, 'Q'J^Kn nine;1. 294. On the next page he tries to
collate the material in each source together with that of
the Aramaic Targum to Song., and to produce unified lists
for each source and across sources.
6. The list is reproduced verbatim, with the Greek column as in
the original 1331U)t> text.
7. 3 times - Ex. 28:17, 39:10; Ezek. 28:13.
8. And so the Aramaic T (from /pttd).
9. Gradwohl, 15.
10. Gen. 2:12; Ex. 25:7; 28:9, 20; 35:9, 27; 39:6, 13;
Ezek. 28:13; Job 28:16; 1 Ch. 29:2.
11. BDB, 995, with a question mark; Landsberger, 150f.
12. BDB, 986; Grintz, "d^ttlTi? d»rmtt", IJJUgb XXXIX (1975), 9.
13. Ezek. 27:16; Isa. 54:12.
14. Cf. KniDTS as the rendering for 1£JJ, KJn313 for 13», TDTD
for d^n> - (cf. Qimhi, d?Eni!>, 347, for the latter), and
see Shalem1s table (p. 258).
15. BDB, 461.
16. GBi KB, ad. loc.; TutvSinai, "IDTD", Sac. Miq. IV, 12-13.
17. Rabin, "Hl»nn d^tt", 152.
18. Ex. 28:19, 39:12 (P).
19. Ellenbogen, Foreign Words, 22; KB, 31; and Grintz, "d^ruia",
8.
20. For a summary, see Loewenstamm, "ftt^riK", Enc. Mig.. 1, 228.
21. Ex. 28:17, 39:10 (P); Ezek. 28:13; Job 28:19.
22. KB, 758; Ellenbogen, op. cit., 133. For a summary see
Rabin, ",YTU3", Enc. Miq. VI, 453f.
NOTES TO PP. 167-169 255

23. Grintz, op. cit., cf. Rabin, op. cit., and in his article
"ni'Tin O^O", 240 (in a ref. to Tur-Sinai's reservation).
24. Ex. 28:17, 39:10 (P); Ezek. 28:13.
25. For /;m: Qimhi, tPKHW, 98; Grintz, op. cit., 9. In
favour of the loan from Sanskrit: KB, 156; Rabin, op. cit.
For a summary see Loewenstamm, "fit?1"!!!", Enc._ Mig. II, 265f.
26. Ex. 28:18; Ezek. 28:13, 27:16 (the latter has n»|7ll IttYlN) .
27. BOB, 656; KB, 624.
28. Rabin, "119", Enc. Mig. IV, 441,
29. Qimhi, D*ETH!>, 444.
30. Jastrow, 1262: antimony. Thus also the Aramaic Targum to
Heb. 113 (Jer. 4:30, 2 Ki. 9:30) as well as to 113 stones
(Isa. 54:11).
31. Lucas, op. cit.
32. Ex. 24:10, 28:18, 39:11; Isa. 54:11; Ezek. 1:26, 10:1,
28:13; Job. 28:6, 16; Lam. 4:7; Song. 5:14.
33. Ex. 28:20, 39:13; Ezek. 1:16, 10:9, 28:13; Song. 5:14;
Dan. 10:6.
34. Cf. below, III, D.I.1.3 (MH). Thus also St. Jerome, who
understood KPEJlft as 'sea' - cf. J. Barr, Bull, of the
J. Rylands Library 49 (1966-67), 291-2.
35. Qimhi, WVTW, 830.
36. BOB) KB, for T>30; Grintz, op. cit., 10.
37. Cf. Landsberger, 'Farben', 154, for a similar usage of
'lapis lazuli' for describing the colour of the sky.
38. 1 Ch. 29:2 (KPKJ); Song. 5:15; Esth. 1:6 (twice).
39. II, E.2.3.1 and dictionaries ad Joe.
40. E. Ben Dor, "t>m", Enc; Mig. II, 36f.
41. Lucas, 233ff. (Egypt); Landsberger, op. cit., 144.
42. II, B.5 (lh^ and C.5 (SniM).
43. BOB, 359; KB, 332; Kutsofter, D^Q, 9.
44. Jastrow, 502; III, D.4.2.
45. II, C.4.1.1.
46. For the etymology of DID: KB, 461, proposes a Nubian
origin; compare, however, Ellenbogen's reservations in
Foreign Words, 95. The latter suggests the Egyptian
ktm.t, a loan in Egyptian itself.
47. BOB, 494, after Gesenius's Thesaurus.
48. Jastrow, 655.
49. I am grateful to Professor Barr for reminding me of the
Akk. cognate and the Arabic inter-linguistic equivalent.

G.

G.I

1. Cf. II, A.6 for categorization and arrangement.

2. BOB, 666; KB, 632; Qimhi, D^1K>, 450; Ibn Janah, D>erw f
316; Albeck, K13Q, 192'(for IpJ); BOB, 785; KBi 73U1
Qimhi, 552f.; Ibn Janalj, 382 (for *ftjp; Cf. also
commentaries, and D. Ashbel, "D^Tmi QmiJj tP'Tlpy",
Klptt tTO X, 48-52.
G.2

3. BOB, 378; KB, 352; Ibn Janah, Q'nm, 180; Qimhi, Q*ETlE),
254£.
4. Qimhi, op. cit.
256 NOTES TO PP. 169-173

5. For instance, BOB, III y/W, 742; y'W, 905; and

commentaries ad Joe.
6. BOB, 742, II N/qOy.
7. Dillmann, Lexicon, 1018b - although no verb form is
attested for Ge'ez.
8. D. Sedan, "pKa", Dy!? IJJIBft II (1951), 20-24; Y. Abineri,
"tim^n *miy by", U3iu>5~xvii (1966), 78f.
9. BOB, 649 - from 'shine', 'gleam1, 'be glossy'; KB, 618f.
Levias, Jewish Encyclopedia, 177; Qimhi, d^unw, 438,
.D'rniD D^mo mUM FVPyt>> n*n. For a summary of data and
literature, see: E. Bilik, Enc. Miq. V, 870-872.
10. Jastrow, 914; Moresheth, byiDH 11??03!>, 119.
11. For 'multi-coloured textiles' see under E.2.4.
12. Cf. Gradwohl for a survey of the VSS. he uses, 54ff.;
Thatcher, DB, I, 458; and literature cited above, nn. 1, 2
and 3.
13. Gradwohl, 56.
14. Cf. W.C. Libby, Color and Structural Sense; "Colour
Contrast", 63ff., and especially 73ff.

H.

H.I

1. G.-K., 84h.; Bauer-Leander, Graimatik, 466-468.

2. D.J. Kamhi, 'The Term To'ar in Hebrew', Bulletin of the
School of Oriental and African Studies XXXIV (1971), 271.
3. Thus Rashi; cf. J. Pereira Mendoza, Rashi as Philologist,
35. For a survey of other commentators and grammarians -
Ibn Janah, Ibn Ezra, Saadia, Qimhi, E. Levitas - cf. Kamhi,
op. cit.', 257-262; and 263-268 for attempts of modern
scholars and lexicographers to grapple with the question,
.followed (269ff.) by a proposed solution to the problem,
viz. a, tripartite division into substantive nouns, epithet
nouns, and proper adjectives, with the admission that
there are regular shifts from class to class. It must be
notedj however, that Kamhi defines tHK and the like as
T
"pure adjectives" (271).
4. For a survey and general remarks see Lyons, Introduction
to Theoretical Linguistics, 323-325 (Adj. vs. verb); 327
(Parts of speech); and 435ff. ('lexical' and 'grammatical'
meaning).
5. Thus Ben Yehuda. - cf. R. Sivan, 1Ty>!M ^ tP!?an »t31T>n",
'Win* r^py!? laaitft xn (1961), 66.
6. Lyons, op ~cit., 323-325.
7. Cf. Bauer Leander, op. cit., for incomplete lists. Here
we record mainly lexemes exhibiting identity of series in
all forms - m. and f., sg. and pi., where all members of
the series are attested (excluding basic phonetic influences,
such as in ril-A ,r»!nl->, etc., and original gatal forms).
H.2 -r T .

8. Hos. 6:8; Jer. 17:9.

9. Isa. 40:4.
10. In Gen. 2:25, perhaps paranpmasia to resemble d-1iy = 'sly',
'shrewd', Gen. 3:1; cf. t^tt'l^, 3:7. ^
11. F. pi. and suffixed forms constructed from the alternative
W.
TT
NOTES TO PP. 173-180 257

12. As in D^aOl^n, 1 Ch. 26:15, 17; Neh. 12:25.

is. mac* t'tyz'; Qoh. 12:11.
14. Ex. 28:8, 39:5; Isa. 30:
15. Ezek. 23:41, from TUS?; Jud. 18:21, 'riches
16. = 'swarm', Ex. 8:17, passim.
17. Ace. to BDB, also phO (608), %iy (772), inj) (800), 101 (o
301, 936); cf. list £., supra.
"T
H.4

18. Perhaps because of the homonymic v/1^ = 'make bricks'

(Gen. 11:3; Ex. 5:7, 14), although this is not frequent
enough or as widely distributed as to cause the elision of
an identical verb form from the vernacular. At any rate,
whether this Qal v"n!7 is connected to the 11^ = colour
notion through M^ - 'brick', 'tile' or is genetically
distinct from the latter, it should be considered a
homonym (semantic, through polysemy; or etymological/
semantic).
19. Basic Color Terms, 15ff.
20. Kamhi, op. cit., 269ff.
21. Grammatik, 461-463, although not an exhaustive list.
22. Although not necessarily 'milk', as claimed here; cf. II,
B.2.0.1.
23. Op. cit., 462.

Part III

A.

A.I
1. For a general list of MH terms (albeit a loosely
structured one), cf. Herszberg, AlKn, 224-227.
2. For the purpose of this study - where MH is compared to
biblical Hebrew, which is the subject of investigation -
no distinction is made between the various dialects of
MH; see Kutscher, Enc. Jud. XVI, 1590ff. Rather,
linguistic corpus is treated en bloc as a single entity.
A. 2
3. Kutscher, op. cit., 1603,
4. In the following sections only a few references to
standard works (dictionaries, concordances) are supplied,
because of the concise manner in which the notes are
presented. The works used, apart from the texts themselves,
are: Levy, Jastrow, Ben Yehuda, (dictionaries), Kassovsky
(concordances), Kohut (Aruch).
5. As in Jewish literature, not as a dialectical designation.

!•
B.2
1. Kedar, The Vulgate, 169.
258 NOTES TO PP. 181-198

B.3
1. Cf. dictionaries ad. loc.
B.5
1. Cf. I, A, and literature cited there.
C.
C.I
1. Albeck, Kiaa, 166; Moresheth, XV'D^, 144, and n. 8 ibid.
2. Albeck, op. cit.
3. Jastrow, 17, "especially with reference to hair".
4. Ibid., 457-458.
5. Ibid., 681.
6. Ibid., 235ff.
7. As summarized in the dictionaries ad. loc.
C.3
1. And cf. Jastrow and Kohut ad. loc.
2. And cf. Jastrow and Kohut ad. loc.
3. A well organized list is to be found in Herszberg, A1KM,
230f.
C.4
1. Ibid., 239-241.
2. Ibid.
2-
D.I
1. Herszberg, 226; Jastrow, 60, 665, speculates that the
lexeme is an allomorph of the Hebrew d11p, which does not
sound convincing.
2. BOB, 702.
3. Jastrow, 670.
4. Ibid., 1169f.; from Greek 'seaweed'; with no colour
property apparent in the source language.
5. ibid., 1229.
6. Jastrow, 1140 Moresheth, T»j7">D3b, 163.
7. Cf. Jastrow, 1152f.
D.2
1. Yadin, tjTTpnn fl^O II, 23.
2. Jastrow, 1148.
3. Landsberger, 'Farben', 169 - contra Gradwohl, 73ff., who
derives ITTlflT from v'lnT, 'shine1.
4. Landsberger, op. cit., 164f.
5. Kutscher, D>»a, 98, 117; see also: II, E.2.3.3. above.
6. Jastrow, 60.
7. Herszberg, 251ff.
8. Yadin, D>Kyaa, 178ff. Especially interesting for this
matter are the colour photographs of dyed cloth supplied.
D.3
1. Jastrow, 1169.
2. Herszberg, 257.
3. Ibid., 261-262.
4. Landsberger, 'Farben', 145, n. 28.
NOTES TO PP. 199-213 259

D.4
1. Herszberg, 251ff.
2. Yadin, op. cit., 180ff.
3. Ibid.
4. Jastrow, 1480.
5. But cf. Herszberg, 254.
6. Ibid.
7. Jastrow, 710; Herszberg, 255.
8. Herszberg, 256f.
9. Cf. Kohut ad. loc.
10. Jastrow, 21.
11. Yadin, op. cit.
D.5
1. Albeck's commentary ad. loc.
2. Jastrow, 774f.
D.6
1. Ibid., 241; Herszberg, 263.

Part V

1. Berlin-Kay, Basic Colour Terms, 4.

2. Jastrow, 618, 629.
3. R. sivan, ">inyn imn mnni mm* 11 "iry^N", oyi> I33iefr
XXIV, 88f.
4. Jastrow, 375.
5. R. Sivan, "D^nn "lltt", 133Hg!7 XXIV, 209.
6. op. cit., 210. This particular suggestion dates, according
to Sivan, from 1898 and was proposed by Yudilevitz. On the
other hand, according to R. Weiss, "mrliT> 13 IT^N WPM >y",
13311!>5 XII, 144, 5l>0 was first introduced into the language
by Y.M. Pines in 1894.
7. Ben Yehuda, Diet., XI, 5367.
8. Defined (ii>id.) as 'reddish black' or 'blackish-brown1 .
9., R. sivan, "o^nn *«m»n'*; py5 I3am5 xn, 41.
10. JMd., 66.
11. Founded in 1890.
12. Founded in 1953.
13. A certain insight into the energetic activities of llU^n iy\
might be gleaned from a cursory glance in any of the first
volumes of 1331195. For instance, within the randomly picked
1944/5 volume, there are lists of terms relevant to the
following fields: Psychology (1944: 166-177, 252-258; 1945:
220-233); Music (1945: 62-64); Statistics: (ibid., 234-250);
Anatomy (ibid., 251-277); Israeli plants (1944: 106-129).
Later the Academy collected most lists and published them,
sometimes with additions and corrections, in special
booklets. The lists are up-dated, revised, and re-issued
periodically.
14. See, among others, the arguments presented by Shalem,
"rp-nyn a»3nxn nittl!>5", 1331E>5 IV, 16-66; and also the
discussion between Har Zahav and Zlotnik in 1331W5 I (1928),
127-144 and 443-447 respectively, in which both attempted to
explain biblical d*T)3*TK (and its productive potential), and
dealt with the wider notion of adapting old forms and
patterns to modern usage as well. For statements on the
principles which governed the renewal of the language, see
260 NOTES TO PP. 213-218

the Hebrew Language Academy, THgfrVryl fllTfr'infr niliyn 0?5,

ym - Tin nnayn iie>X> rpqipKni, 1970, 27-35, (on the
foundation and the tasks of the Committee for the Hebrew
Language), and other articles there, like the one on 125-156
(on the enlargement of Hebrew vocabulary). For general
remarks upon the subject see R. Sivan, 13W TIEfr ">3aiK 5y,
especially ch. 2 (13-24), 3 (25-28), 6 (53-56), 9 (77-95),
and II (109-114).
15. The following remarks relate only to those sectors (2-9)
which contain words for colours, pigments and dyes, and not
to section 1. (General concepts; concepts for colouring,
painting and drawing). As section 1. covers a semantic
sector that is almost non-existent in biblical Hebrew -
although the position is different in MH - there is no point
in positing the suggestions catalogued there vis-a-vis the
material that belongs to the linguistic strata described in
the present work.
16. See the entry in Ben Yehuda's Dictionary, XVI, 7752-7753,
and Tur-Sinai's notes there.
17. And see objections to some of the proposed forms in
Y.E. zeidman, "D^ntti >njin ntt'Kn!? niTtiN miyn", i;me>t> vi,
262f. .
18. ^ on.3£2«» - DTK: Y. Kena'ani 11 Efrn 1>nK I, 32. It must
be noted,' 'however, that Kena'ani substantiates his definition,
which must perforce be valid for modern Hebrew, by recording
biblical material - msflti M13 (Num. 19:2) and DJ3 0*)3*T£ d»»n
(2 Ki. 3:22). Of these'two, the former does not confo'rm to
the above quoted definition, whereas the latter is open to
other interpretations because of its circumstantial context.
19. Ben Yehudah, VIII, 4118; IK/IK/ 13N, IV, 1817.
20. Rabin-Raday, ti»»afT nSIK II, 867-871, 3>3#.
21. In Ben Yehuda:, XVI, 7748 5bfl is translated Hiwxnelblau,
azure and described as used in speech and in modern
literature (following a certain interpretation of biblical
fl^n, ibid., n. 2). There are entries for the term as
such in Itm -pN (1951), IV, 1794; HyiU/ "pN (1966), VII,
2854, together with quotations from modern literary sources;
and Rabin-Raday, II, 869. However, when Hebrew speakers
attempt to define tOfl, they usually employ the syntagm
"Pna Vh^ and/or CCPI'toM) tPBWn y^X3 (thus 1KH0 "ON). This
practice might be indicative of the restricted usage of %Ml
(and its derived relatives) to literary or poetic contexts,
while 1*M3 ^H3 is the more usual term in everyday speech.
22. D>!?nn 1S1K Ilj 867-871.
23. Jastrow, 17.
VII BIBLIOGRAPHY
This page intentionally left blank
BIBLIOGRAPHY 263

BIBLIOGRAPHY

1. For the sake of convenience the following list is divided into

two parts:
a. Literature originally written or edited in European
languages.
b. Literature originally written in Hebrew, or that was
available to me in a Hebrew edition.
2. Items included are either those that have been extensively
used throughout this study, cited in the Notes and/or
discussed; or items which have influenced my thinking
considerably, although they may not be cited directly.

3. Additional literature - mostly in German and of earlier

dates - is available in R. Gradwohl, Die Farben im alten
Testament (= BZAW 83).

W.F. Albright, Studies in the History of Culture 1942.

C.J. Ball, The Book of Genesis: a Critical Edition
of the Hebrew Text 1896.
K.L. Barker, "The Value of Ugaritic for Old Testament
Studies", Bibliotheca Sacra 133 (1976),
119-129.
J. Barr, "Etymology and the Text of the Old
Testament", Oudtestamentische Studien
XIX (1974), 1-28.
Comparative Philology and the Text of the
Old Testament Oxford, 1968.
J. Earth, ffurzeluntersuchungen zum hebraischen und
aramaischen Lexicon Leipzig,1902.
H. Bauer-P. Leander, Historische Grammatik der hebraischen
Sprache des alten Testamentes Halle, 1922.
W. Baumgartner et al, HebrSisches und aramaisches Lexicon zum
alten Testament (= KB3) Leiden, 1967, I-II.
W. Baumgartner- Lexicon in Veteri Testament! Libros (= KB)
L. Kbhler, Leiden, 1958.
W.H. Bennett, Genesis (Century Bible), 1904.
B. Berlin-P. Kay, Basic Colour Terms: Their Universality and
Evolution California, 1969.
J. Bewer, The Book of Obadiah [in: Micah ... etc.],
(ICC), Edinburgh, 1911.
The Book of Joel, ibid.
F. Birren, Color Psychology and Color Therapy
New York, 1961.
C.A. Briggs, The Book of Psalms (ICC), Edinburgh, 1907,
I-II.
L. Brink, "Semantic Boundary Lines in Languages and
Their Influence on our Cognition of the
Surrounding World", Acta Linguistica
Hafniensia XIII, 1971, 45-74.
264 BIBLIOGRAPHY

H.B. Broch, "A Note on the Hare Indian Color

Terminology", Anthropological Linguistics
XVI,1974, 192-196.
2
K. Brockelmann, Lexicon Syriacum Halle, 1928.
F. Brown-S.R. Driver- Hebrew and English Lexicon of the Old
C.A. Briggs, Testament (= BOB), 1939.
R.W. Brown, "A Study in Language and Cognition",
E.H. Lenneberg, Journal of Abnormal and Social Psychology
49, 1959, 454-462.
C.F. Burney, The Book of Judges 1918 (= Ktav, New York,
1970).
M. Buttenwieser, The Psalms 1938 (= Ktav, N.Y., 1969).
(N. Sarna, ed.),
J.B. Carroll, Language and Thought New York, 1964.
R.H. Charles, The Revelation of St. John (ICC) Edinburgh,
1920, I-II.
T. Cheyne, The Book of Isaiah Leipzig, 1899.
Chicago Assyrian Dictionary (= CAD): see under I.J. Gelb, et al.
G.A. Collier, "Basic Color Terms - a Review", Language
49, 1973, 245-248.
H.C. Conklin, "Hanuo'o Color Categories", Southwestern
Journal of Anthropology XI,1955, 339-344.
G.A. Cooke, The Book of Ezekiel (ICC) Edinburgh, 1936.
2
A.E. Cowley (trans.), Gesenius' Hebrew Grammar (E. Kautzsch,
ed.; 1966 reprint).
M. Dagut, A Linguistic Analysis of some Semantic
Problems of Hebrew-English Translation
(unpublished Ph.D. thesis, Hebrew
University, Jerusalem, 1971).
M. Dahood, Psalms 51-100 (Anchor Bible) New York,
1970.
G. Dalman, Arbeit und Sitte in Palastina GUtersloh,
1928-1942, I-VII.
D. Daube, Studies in Biblical Law Oxford, 1947.
G. Henton-Davies, Genesis (Broadman Bible Commentary), 1969.
J. Deeze, Psychological Linguistics Boston, 1970.
F. Delitzsch, Assyrisches Handw&rterbuch Leipzig, 1896.
A. Demsky, "Dark Wine from Judah", IEJ XXII, 1972,
233-234.
A. Diez Macho, (ed.), Neophyti I (for Genesis - Numbers; Madrid,
1968), I-IV.
A. Dillmann, Lexicon Lingua Aethiopicae cum Indice
Latino (1969 reprint).
Die Genesis Leipzig, 1892.
BIBLIOGRAPHY 265

T. Donald, "The Semantic Field of 'Rich' and 'Poor'

in the Wisdom Literature of Hebrew and
Akkadian", Oriens Antiquus III, 1964,
27-41.
G.R. Driver, Canaanite Myths and Legends Edinburgh, 1956.
"Hebrew Notes on Song of Songs and
Lamentations", in Festschrift A. Bertholet
1950, 134-146.
S.R. Driver, The Book of Job (ICC) Edinburgh, 1921, with
G.B. Gray.
Genesis (Westminster Bible Commentary)
Edinburgh, 1904.
H. Dtlrbeck, "Zur Methode des Semasiologen bei der
Bedeutung-bestimmung von Farbenzeichnungen",
ZDMG 118, 1968, 22-28.
C.B. Ebeling, Linguistic Units The Hague, 1960.
W. Eichrodt Ezekiel (OTL) 1970.
(C. Quin, trans.),
M. Ellenbogen, Foreign Words in the Old Testament: Their
Origin and Etymology London, 1962.
S. Erwin, "Semantic Shifts in Bilingualism",
American Journal of Psychology 74, 1961,
223-241.
R.M. Evans, An Introduction to Color New York, 1948.
M.H. Farbridge, Studies in Biblical and Semitic Symbolism
London, 1923.
"Symbolism (Semitic)"; in: Hastings,
Encyclopaedia of Religion and Ethics, XII,
150.
J. Fellman, "Concerning the 'Revival' of the Hebrew
Language", Anthropological Linguistics XV
1973, 250-257.
S. Fillenbaum, Structures in the Subjective Lexicon
New York, 1971, 41-57.
W.D. Fischer, Farb- und Formbezeichnung in der Sprache
der altarabische Dichtung Wiesbaden, 1965.
R.J. Forbes, Studies in Ancient Terminology Leiden,
1955-6.
H. Frankfort, The Art and Architecture of the Ancient
Orient Harmondsworth, 1954.
M. FrSnkel, "Bemerkungen zum hebrSischen Wortschatz",
HUCA XXXI, 1960, 55-102.
J.G. Frazer, The Golden Bough London, 1950.
J.A. Frisch, "Mohawk Color Terms", Anthropological
Linguistics XIV, 1972, 306-310.
P. Fronzaroli, "I Cavalli del Proto-Zaccaria", Accademia
Nazionale dei Lincei, Series III, Vol.
XXVI, 593-602.
266 BIBLIOGRAPHY

P. Fronzaroli, (Cont'd) "Sulla Struttura del Color! in Ebraico

Biblico" Studi Linguistic! in Onore di
Vittore Pisani I, 1969, 377-389.
"West Semitic Toponyms in Northern Syria
in the Third Millenium B.C.", JSS XXII,
1977, 145-166.
2
A. Gardiner, The Theory of Proper Names 1954.
T.H. Caster, Myth, Legend, and Custom in the Old
Testament New York, 1969.
L. Geiger, Contributions to the History of the
Development of the Human Race London, 1880.
I.J. Gelb, Th. Jacobsen, The Assyrian Dictionary of the Oriental
B. Landsberger, Institute of Chicago (= CAD), GlUckstadt,
A.L. Oppenheim, 1956.
Gesenius1 Hebrew Grammar - see under Cowley.
W. Gesenius, F. Buhl, Hebraisches und aramaisches Handworterbuch
uber das Alte Testament Leipzig, 1921.
Gesenius' Hebrew and Chaldee Lexicon to the Old
Testament Scriptures (S.P. Tregeller,
trans, and ed.) London, 1870.
C.D. Ginsburg, The Song of Songs and Coheleth 1861 (=
Ktav, New York, 1970).
W.E. Gladstone, Studies on Homer and the Homeric Age
London, 1848, III, 457-499.
H.A. Gleason, An Introduction to Descriptive
Linguistics New York, 1961.
A. Goetze, "The Inventory Ibot I 31", JCS X, 1956,
32-38.
C.H. Gordon, Ugaritic Textbook Rome, 1965.
M.H. Goshen-Gottstein Aramaic Bible Versions Jerusalem,. 1963.
(ed.),
'"Linguistic Structure and Tradition i
the Qumran Documents", Scripta Hieroso-
limitana IV, 1958, 101-137.
R. Gradwohl, Die Farben im Alten Testament (BZAW 83)
Berlin, 1963.
G.B. Gray, The Book of Isaiah I - XXXIX (ICC)
Edinburgh, 1912 = 1956.
The Book of Numbers (ICC) Edinburgh, 1903.
Job (see under S.R. Driver).
J. Gray, "A Contata of the Autumn Festival: Psalm
LXVIII", JSS XXII, 1977, 1-26.
J. Gray, The Legacy of Canaan (VTS) Leiden, 1957.
F. Grbndahl, Die Personennamen der Texte aus Ugarit
Rome, 1967.
A. Guillaume, A Hebrew and Arabic Lexicography: A
Comparative Study (Abr Nahraim I-IV,
1959-1965; Leiden, 1965).
BIBLIOGRAPHY 267

A. Guillaumont, "La designation des couleurs en Hebreu et

Arameen" in I. Meyersohn, Problemes de la
couleur Paris, 1957.
H. Gunkel, Genesis3 GOttingen, 1910 (= 1964).;
(W.H. Corruth, trans.), The Legends of
Genesis 1901 (= 1964).
A.C. Heinrich, "A Non-European System of Color Classifi-
cation", Anthropological Linguistics XIV
1972, 220-227.
J.H. Hill, K.C. Hill, "A Note on the Uto~Atztecan Color
Terminologies", Anthropological
Linguistics XII, 1970, 231-238.
W.L. Holladay, A Concise Hebrew and Aramaic Lexicon of
the Old Testament Leiden, 1971.
H.M. Hbningswald, Language Change and Linguistic Reconstruc-
tion Chicago, 1960.
E.V. Hulse, "The Nature of Biblical 'Leprosy' and the
Use of Alternative Medical Terms in
Modern Translations of the Bible", PEQ
107, 1975, 87-105.
A. Hurvitz, "The Evidence of Language in Dating the
Priestly Code: A Linguistic Study in
Technical Idioms and Terminology", RB 81,
1974, 24-56.
"The Usage of W and VIS in the Bible
and Its Implication for the Date of P",
HThR 60, 1967, 117-121.
B. Jacob (E.I. Jacob, The First Book of the Bible - Genesis
W. Jacob, ed. and Ktav, New York, 1974.
trans.),
M. Jastrow, A Dictionary of the Targumim, the Talmud
Babli and Yerushalmi and the Midrashic
Literature 1903 (= New York, 1975).
L.B. Jensen, "Royal Purple of Tyre", JNES XXII, 1963,
104-118.
S.A. Kaufman, The Akkadian Influence on Aramaic
Chicago, 1974.
P. Kay: see under B. Berlin.
B. Kedar-Kopfstein, The Vulgate as a Translation (unpublished
Ph.D. Thesis, Hebrew University,
Jerusalem, 1968).
D.J. Kamhi, "The Term To'ar in Hebrew and Its Status
as a Grammatical Category", Bulletin of
the School of Oriental and African
Studies (University of London) XXXIV,
1971, 256-272.
R. Kittel, Biblia Hebraica (= BH3) Stuttgart, 1962.
268 BIBLIOGRAPHY

L. KOhler, "Alttestamentliche Wortforschung",

Theologische Zeitschrift 1947, 314-18,
ibid., 1948, 233f.; ibid., 1949, 314f.
"HebrSische Vokabeln", ZAW 54, 1937, 161-167.
Lexicon; see under W. Baumgartner.
E.Y. Kutscher, Mishnaic Hebrew in Encyclopedia Judaica,
XVI, 1590-1607.
T.O. Lambdin, "Egyptian Loan Words in the OT", JAOS 73,
1953, 145-155.
H.J. Landar, S.M. Ervin "Navaho Colour Categories", Language 36,
A.E. Horowitz, 1960, 368-382.
B. Landsberger, "Akkadisch-hebrSische Wortgleichungen",
VTS XVI, 1967, 176-204.
"Uber Farben in sumerisch-akkadischen",
JCS XXI, 1967, 139-173).
E.W. Lane, Arabic-English Lexicon London, 1865.
P. Leander: see under H. Bauer.
G.M. Lee, "Song of Songs V:10, "My Beloved is White
and Ruddy"", VT XXI, 1971, 609.
de Lee Lantz, "Language and Cognition Revisited",
V. Stefflre, Journal of Abnormal and Social Psychology
69, 1964, 172-181.
de Lee Lantz, "Verbal Communication and Colour Memory
E.H. Lenneberg, in the Deaf and Hearing", child
Development 37, 1966, 765-779.
E.H. Lenneberg: see previous entry.
"Cognition in Ethnolinguistics", Language
29, 1953, 403-471.
W. Leslau, Ethiopic and South Arabic Contributions to
the Hebrew Lexicon Berkeley and Los
Angeles, 1958.
E.A. Leslie, The Psalms New York, 1959.
C. Levias, Color; in The Jewish Encyclopedia, II,
174-178.
2
J. Levy, Wbrterbuch Uber die Talmudim und Midrashim
Berlin-Vienna 1924 - 1963, I-IVv
C.T. Lewis, C. Short, Latin Dictionary Oxford, 1958.
W.C. Libby, Color and Structural Sense New York, 1974.
I. Lbw, Die Flora der Juden Vienna-Leipzig, 1928-
1934, I-IV.
"Semitische Farbenpflanzen", Zeitschrift
fur Semistik und Verwandte Gebiete 1922,
97-162.
A. Lucas (J.R. Harris, Ancient Egyptian Materials and
rev.), Industries^ London, 1962.
J.R. Lumby, The Second Book of Kings Cambridge, 1904.
BIBLIOGRAPHY 269

J. Lyons, Introduction to Theoretical Linguistics

Cambridge, 1968.
A. Kaerz, M.R. Paul, A Dictionary of Color New York, 1950.
Y. Malkiel, Essays on Linguistic Themes Oxford, 1968,
175-256 (chapters on etymology).
B. Maisler (= Mazar), "Canaan and the Canaanites", BASOR 102,
1946, 7ff.
G. Currie Martin, Proverbs, Ecclesiastes, and the Song of
Songs (Century Bible), 1908.
W. McKane, Proverbs (OIL), London, 1970.
H.G. Mitchell, Haggai, Zechariah, Malachi, and
J.M.P. Smith, Jonah (ICC), Edinburgh, 1912.
J.A. Bewer,
J.A. Montgomery, The Book of Daniel (ICC), Edinburgh, 1927.
G.F. Moore, The Book of Judges (ICC), Edinburgh, 1895.
S. Moscati et al., An Introduction to the Comparative
Grammar of the Semitic Languages
Wiesbaden, 1964.
A. Murtonen, "On the Interpretation of the Matres
Lectionis in Biblical Hebrew: Abr Naharain
XIV, 1973-1974, 66-121.
E.A. Nida, "Principles of Translation as Exemplified
by Bible Translating", in R.A. Brower (ed.)
On Translation Harvard, 1959.
"A System for the Description of Semantic
Elements", Word VII, 1951, 2-3.
M. Moth, Die Israelitischen Personennamen im
Rahmen der gemein-semitischen Namengebung
Stuttgart, 1928.
(J.E. Anderson, trans.), Leviticus (OTL),
London, 1965.
S. Ohman, "Theories of the Linguistic Field", Word
IX (1953), 123-134.
W.O.E. Oesterly, The Psalms London, 1939, I.
P. Ole*ron, "Conceptual Thinking of the Deaf",
American Annals of the Deaf 98, 1953,
394-310.
A.L. Oppenheim, "Towards a History of Glass in the
Ancient Near East", JAOS 93, 1973,
259-266.
Optical Society of America, Committee on Colorimetry:
The Science of Color New York, 1953.
J.L. Palache (R.J.Z. Semantic Notes on the Hebrew Lexicon
Werblowsky, trans. Leiden, 1959.
and ed.)
F.R. Palmer, Semantics: a New Outline Cambridge, 1976.
270 BIBLIOGRAPHY

C.H. Parker, The Tyrian Oracles in Ezekiel: a Study of

Ezekiel 26:1 - 28:19 (Ph.D. Thesis,
Columbia University, 1970 = University
Microfilms, Ann Arbor, Michigan).
S.M. Paul, "Classification of Wine in Mesopotamia
and Rabbinic Sources", IEJ XXV, 1975, 42-44.
J. Pereira-Mendoza, Rashi as Philologist Manchester, 1940.
R. Polzin, Late Biblical Hebrew: Towards an
Historical Typology of Biblical Hebrew
Prose (HSM), Missoula, 1976.
J.D. Prince, "Note on Leprosy in the Old Testament",
JBL 38, 1919, 30.
M.H. Pope, Job (Anchor Bible), 1965.
N.W. Porteus, Daniel (OTL), London, 1965 (= 1972).
G. von Rad Genesis (OTL), London, 1961.
(J.H. Marks, trans.),
V.F. Ray, "Techniques and Problems in the Study of
Human Colour Perception", South-Eastern
Journal of Anthropology VIII, 1952, 252.
G.H. Reichard, "Language and Synesthesia", Word 5, 1949,
R. Jacobson, E. Werth, 224-233.
L. Sabourin, The Psalms: Their Origin and Meaning 1969,
II.
E. Sapir, Language: An Introduction to the Science
of Speech London, 1921 (= 1971), 192-220.
F. de Saussure, A Course in General Linguistics (W. Baskin,
trans.), revised ed., 1974.
J.F.A. Sawyer, "A Note on the Etymology of sara'at", VT
XXVI, 1976, 241-245.
"Root Meanings in Hebrew", JSS XII 1967,
37-50.
Septuagint: A. Rahlfs, Stuttgart, 1935.
H.B. Swete, The Old Testament in Greek.
D. Simpson, Cassell's Latin Dictionary London, 1964.
S. Skard, "The Use of Color in Literature: A
Survey of Research", Proceedings of the
American Philosophical Society 90, 1946,
163-249.
J. Skinner, The Book of Genesis (ICC), Edinburgh, 1912.
Isaiah XL-LXVI (Cambridge Bible), 1898.
G.A. Smith, The Book of the Twelve Prophets London,
1928, I-II.
W. Robertson-Smith, "A Journey in the Hedjaz" in J.S. Black-
G. Crystal (eds.), Lectures and Essays
London, 1912, 583-590.
N.H. Snaith, Leviticus and Numbers (Century Bible),
1967.
BIBLIOGRAPHY 271

D.L. Snow, "Samoan Color Terminology: A Note on the

Universality and Evolutionary Ordering of
Color Terms", Anthropological Linguistics
XIII, 1971, 385-390.
V. von Soden, Akkadisches Handworterbuch Wiesbaden, 1959.
W. Sommer, "The Meaning of 113 in the Old Testament",
JBL 62, 1943, 33-35.
E.A. Speiser, "The Name Phoinikes", Language 6, 1936,
121-126.
A. Sperber, The Bible in Aramaic Leiden, 1959, I-IV.
2
C.J. Spurrell, Notes on the Text of the Book of Genesis
Oxford, 1895.
A.W. Streane, Jeremiah and Lamentations (Cambridge
Bible), 1903.
G.W. Thatcher, 'Colours' in Hastings' Dictionary of the
Bible, I, 456-458.
F. Thureau Dangin, "Un Comptoir de la lain pourpre a Ugarit
d'apres une tablette de Ras Shamra",
Syria XV, 1934, 137ff.
W.C. Till, "Die Farbenzeichnungen in Koptischen",
Analecta Biblica 1959, 331-342.
C.H. Toy, The Book of Proverbs (ICC), Edinburgh,
1899).
W.V. Turner, "Colour Classification in Ndembu Ritual";
in: M. Banton (ed.), Anthropological
Approaches to the Study of Religion
London, 1966 = 1969).
N.H. Tur-Sinai, The Book of Job Jerusalem, 1967.
E. Ullendorff, "The Contribution of South Semitics to
Hebrew Lexicography", VT VI3 1956, 190-198.
Ethiopia and the Bible, London, 1968.
"Is Biblical Hebrew a Language?" Bulletin
of the School of Oriental and African
Studies University of London), XXXIV, 1971,
241-255.
2
S. Ullmann, The Principles of Semantics Oxford, 1959.
Semantics: An Introduction to the Science
of Meaning Oxford, 1972.
K.R. Veenhof, Aspects of Old Assyrian Trade and Its
Terminology, Leiden, 1972.
2
M.D. Vernon, The Psychology of Perception
Harmondsworth, 1973.
The Vulgate: ed. P.M. Hetzenauer, Rome, 1914.
V. Weinreich, "On the Semantic Structure of Language"
in I. Greenberg (ed.), Universals of
Language, Cambridge, Mass., 1963, 114-171.
A. Weiser, The Psalms, (OTL), 1962.
272 BIBLIOGRAPHY

R.W. Wescott, "Bin! Color Terms", Anthropological

Linguistics XII, ,1970, 349-360.
C. Westermann, Genesis (BKAT), 1974.
(D.M.G. Stalker, trans.), Isaiah 40-66
(OIL), 1969.
B.L. Whorf Language, Thought, and Reality M.I.T.,
(J.B. Carroll, ed.), 1956.
L.M. Wilson, Ancient Textiles from Egypt in the
University of Michigan Collection,
University of Michigan Studies, 1933.
D. Winton Thomas, "Some Observations on the Hebrew Word
13)71", VTS XVI, 1967, 387-391.
 273
BIBLIOGRAPHY

.87-35 ,(1966) xvii dy» U3ie>» ,"d.*3i!»n -»rniy ty" ,'IJ^K „ »

.1970-1966 ,d»^env ,Knn?T_vi^ari , lenty in* O K
.(1978) ii ,(1977) i ,T'3n!7 nmri fpyinipaip
.iv-i ,1960 jd^tyfp ,p>:nrDi d*K*33 "yy oiva ,^N3313N .*
,nmpon iVaiAiK^ :Kipnn naipna frncp VIK ,>3iinK .•>
1962 2 ,d^K7lV
.1969 9 ,d^Bm> = K"tnn ,1^13 ,-IDI^!? Kii7a ,i>t?inK ^"N
,d>!?»yii^ ,np^oa^^i7ioKb riKiajn ia :n^iat n u i r n ,io»^njaiK .^
.1961
, m^Ki n^a »n>a K-ipan ni^iK T>3 nt>^p ^iwj? ,n!?>K .n
.1977 ,d'»^TT'
,n i >nK !7n-d'^iT> ,n3m^ xina
.1974 ,P^K ,n
.vii-i ,--1950 ,d^wn> ,n">Kipa rpisi^m
nmyn IIK>^ n^anpKm n^n iyi nn!?in!7 nmyn opt? iiu;!?!? n*mpNn
.1970 ,Q'>''W\V ,!?"wn-Vin ,n>inyn
.52-48 ,x Kipn n»a ,"d'^^'^n'l dmpj d'Tipy" ,^BK ,n
v
Py^ iJJi^ ,"nt>'N TV d'^wn^n :!?Kiu;'n runty" ,!?U;K ^"^
.10-3 3 ,(1954)
= VJ'in , T»!?i3 ,ruo':\ I^K M3i> "i^ d^aian lap ,iyDK3 T"n
.1966 jd^Bll^
"ynn pipi>^ n>nn^Ki qoM nanj rm1* >aiD" ,i^>^ .T
.227 ,(1973) 53 = II KIpQ TP3
.872-870, (1963) V n^Klptt .S3K ,"naj" ,P'!"n .K
1
.in-i ,--1957 .d'^n ? ,Kipnn in;;!? I^IK ,iNto .•»
,dKDty3v^ K"W
,(1968) xxxn i3jiat> ,"Kipm d^y^fi ^y ipno" ,iKt?n .K
.406-402
.(1972) tn*mni n>nn tiiiti
.39-36 ,11 fl^KlpO .^JK ,"Dn3" ,111 13 .y
.77-75 ,(1941) xii y^nin ,"d^a npn" ,d>-»n in .r
.1955 ,d#> ninnn-nerrnm ruu>^n n^inyn iit>o ,min» in .K
Kipnn ipnt? ninnn) d>i ia& , n nmKi diK di" ,IBI> n .0
.119-112 ,1958 ,a^e>1"P ,t?K1E)'^
.1972 ,Q>!>B>II> ,nnnyn n^n pnpn loio^in .>
(n»K p .a rdAin)
.1968 ^^env = 1884 ,1^1^ ,Ol!?pJ1K dUltl ,1J''!?m .K
byi iTTinn t>y >"t UKA n^iyo• u^ai un~T>o ,(Tiiy) i> .'
.1963 ,iii:n5 ,111 ,nt>np
.1903 ,T>t?in ,rviinn by ^Kmy p i m i ^ d u i n ,im:m»> .0
xxxix 1331^ ,"d'3ni rmrn d^aiip d^njia" ,VJ'IA o">
.32-5 ,(1976) XL 13310^7 ;181-163 ,20-5 ,(1975)
,*a»D"i:i IPO ,"i>>ep dro ^D ^^ UK »by hi? dro" ,'j3inn .w
.1970 jd^un-p ,!?KIU)*^ Kipnn ipn^ niann
274 BIBLIOGRAPHY

.1959 ,d*bwiv ,min"> xnn ni^na , ttmn n"x

.1972 ,d^K>iT> ,ing!?!? iiefr "pa ,y>^iin .K
ninn ,"n>Kipnn rr»iiyn "on^n p-mtK-o'"
,"111^ spy* tw ITJT!? tPiftnn - frnep nn^ini Kipon
255-248 ,1972 ,^-»^K-^n
.(1966) ,"maa - >iap ,v»^mn p"n
,(Yiijfl
1927 , •ptro .o^^Kia iao ^niy ^3t> pnyia Ku>B;a ,i'^n .n
.1970 ,^'DK-^n-
.183-176 ,(1929) ii iJ3iwt> ,"*"K ^nn^ moBi d^ipno" .•'nniKin .K
442-126 ,(1930) in ijji^ ,"!?N-)ty* ^ii m»w ii7nV
.144-127 ,(1928) i 13J1K;!? ,'fi!mnm naopnn mi.ix" ,nnr in 0 ^
,D»^II> ,>j^t3 110 lap , Mi »ittnn •'J^un AiTnn iiDvynn" ,lin .n
.42-27 ,1960
.200-183 ,1973 ;^^K-^n ,Kipttn nnoiQi niaipn
n"y>K?yni >iNn :i ,Tm5nm ruemn naij?n^ nimnn ^»n ,;n:n2nn WK
.1924,rH3111 , AIKM
, yi>3*ni7 lap ,"nKiinn iim dty!? HIK <»n^e;" ,1^*1 .K
.209-171 ,1964 ,d'^TT>
.ii-i ,1964 jd^wT-p 2 ,nxj^ i^K nji^ "fo ryayiniao ^PPJ^^I .0
, ' N yng -wo ,M»jn!nan1i">'nD^)3ttn Typi - KIIWD luwnn" ,ibaj>i .n
.131-122 ,1973
f
,(1961) xii D#> 1331^ , «nTin» in 'K »Kn"Ph t»y" ,p>i .1
.148-143
xii dyt? i33ie>t> ,"'Pi!nT>rQ nmn^ 12 'K bw ino'iy"
.240-232 ,206-199 ,(1961)
.231-230 ,iv n^Kipa .^3K ,"iaD" .•'inr .n
.313-312 ,DW ,"d31D"
vi ujiefr , fl ory^n »njin n)3'»K;Ni!? nnnK imyn" ,IO*T>T K"*
.263-262 ,(1934)
.447-443 ,(1928) I 1331^ ."DTttlK ,p1|7T>" ,p'»3Dl!7r .'
,1971 ,d'»!7unii» ,dt>> ^aim !?aa .viipt .•»
.1954,1951 ,p^B11» ,iapm 11K>!?n,'3*P-11D n"3
.2-1 ,(1947) XIX 1331E>5 ."nan KBtt"
.13-12 ,IV n">K1j7Q .^3K ,"TDTD"
xiii i33iu)!> ,"iYnn-> -]i iry*t»K ^w I3it>»a ^iK)i" (=i31>i2n!;iiu n"3)
.27-15 ,(1941)
.1957 ,D»!?i2nv 2 ,lgih ^3^ IIK ?33 nnnt>tt ri!>^AO , VT* .»
.d^Ni^o TID+III-I ,1977 ,d'!7\yiT ,OTjm n'y^n
.1963 ,d»^ii^ ,niiAKn iiiyaa K^aia i^ ^o^a d>Kaann
.D^^K; .K : "i ,pi»^a Piapi>
.1010-992 ,iv tT>K"ij7tt .^3K ,"n-»^n ni^tt" ^^KIW .•>
BIBLIOGRAPHY 275

in Kipa n^ /'ra^on jnsn diio»w d»inN rnnt;" , iro O K

.420 ,(1973)
.xiv-i ,-1960 ,d^B>ii» tPinyn nefoi isix , ">3y3D .*
.xxin-i ,-1927 ,pii> i1^ ,na^a nun groin ,iiy:> n"o
,114-101 ,(1933) v i33iufr ,"iiiy!7n men d^i'o" ,w*i!? .p
.220-205
rirn /MOIIK ma jr>tyy3 T^D :Q^!«>ii» ni»3imp" ,N>II!? *":i
.329-323 ,(1976) XXI K1J7Q
.430 0 ,IV n^K1i?n .^3K ,"»32!?n ,111»!? .»
.ui-i ,-1957 ,n>t>wii» ,Kipan 11 ^ I^IK ,DKD^II»!? K"U;
abcdefghij
.228 ,i iPKipn O^JK ,"na^nK" * °*
.49-48 ,QB» ,"in 1DK"
.530-529 ,QW ,"in>1K"
.366-365 ,IItT>K1j7n ,^3K ,"nt7l3"
.1957 ,D^t>B>ii' ,iiiiii» lain tii!?>>aH ni^iinn ^">>a ,o^!> „•»
oT'Jn^n^jiiipjip .iip^uxa „»
XLI y>ain ,"n"!7 ,T"!? o^nn :'i3jn nitKD niymi"1 ,Aiin .B
.23-1 1 =(1972)
oi973 ,Q»^B»II» ^Jtt^n-^nn i* an:?-- ^K^T IDD
.1972 ,D»!«?II» ,D^K3tin n^a u>inn3K> t>yian 'iip^o^ ,ne;iin O o
,188-172 ,(1934) vi i:mt!>!> ,"^M ni^A •'ihK n^inyn" ,inD nnun .»
.348-338
,ypn iao ,"iii>)w> ^ i3i»t> "yy mniKn nyawn" ,Tnt>n ^"y
.164-142 ,1962 ,d^U>11»
.1966 ,o>^B>ii» .Di^aoaK^ ii{7>obai3iKn ia&
I^IK nitr>3i ,.T>s3Tiii73ii7 ,iit7»n ;KI->D p iao ii^ttii t»yaa
.1973 ,D»^BII^ ,p>ban IIB»W »noo»nn
,ji'in)?n
.1973 ,D->^11-> = 1525,n^31 ,ni^11> HIKIpQ
.166-162 ,(1963) xiv pyt> 1331^ ,"iiT»ninwi nibno" ,n'b>in .T
.1936 ,^^K-t7n,n3<!ff3n iiab pnpi ,^o ^"n
.1972 2 ,D^iyi1> ,D!?t!>n KI^D 13 lap
.24-20 ,(1951) II Dy5 1331^ ,"p^" , 110 .1
,"n>n»i>iK miyittiiruaKmoen D^awn nn^wi -»t;ie;n" jp^pio .•>
.128-125 ,(1947) XVIII V^Din (T>T> .•»)
pyt> I33ie>^ ,"'i^3?ii ii^^in n-»nni nun^ in iry^N" .in^o .1
.116-67 ,(1973) XXIV
xii py5 i33ia!> ,'Miin' in ity^N "?w Q^^on ^wii^n"
.77-37 ,(1961)
XXI ;12-5 ,(1969) XX Dy5 1331K/!? ,"D»t«3i1 »»ntt"
.212-209 ,(1973) XXIV ;249-245 ,(1970)
.1976 ,d»!«j»ii» ,i3^a^ my!? •>3niK "?y
276 BIBLIOGRAPHY

,(1932-1931) iv 13:11^ ,"di>'>3y3:j a^-isy dnpna" ,vwit>t> .3

,124-109
.23-21 ,(1959) x dy5 13311;;!? ,"i3An >h3i»" ^ino .a
.736-735 ,vi iT>KTpa ,S3K ,"nifl^" ,!>IK£> a"e>
xii dy!? I33ia5 ,"mirp in »wmn ty rpSTyn nyaw ,nt)3ajO3 .a
.158-150 ,(1961)
.672-663 ,VI rPKlpa .S3K ,"Q'»)?^" ,03>D 01

.1970 ,i'»^K-!?n=i926 ,n3-»i ,mag> 100!? KiTy PK iao ,iyw^o T'>

2
01968 jd^K/iT ,?K"ii?»'i naiyn d!?iy ^p^^a .»
.1974 jd^Knv ,ii>ii>">»Ki n!?^y yao - D^van i>a
.485-484 ,11 n->K1pa .S3K ,"1^" ,0313 .T
.1937 ,d^tyiT» ,Knaoin ,^T3Knipi^ K;lf)3
. I X - I ,1950 ,p11'>1''3 = 1892-1878 ,i13>1 ,d!?K>n ^liy ,Uinip 0 K
.iv-i ,1967 ,n1>nK-^n 2 ,h3Knan nu>!? I^IK .•'poiiKOp .n
oVi-i ,1961-1941 .d^BiT' ,Knaoinn IIK?!> I^IK
0-1954 ,o»twi» ,nia!?nn nwb I^IK
,1965 ,d^wiT 4 ,n3y n!?Kn ,IOIOKP vn
.1972 ,d->^ii^ ,tr>3y33 tniaoi mtnpo miao
,"Kipon IIK;!?I niirn nmni nynnn niynuan n" •'•np s"n
.46-40 ,(1975) ix n >K>an rrnay ni3^a
niy^Aon nn^n in^ya^ n^nn 'TV ->3iu>!?n ypini iiB>!?n ,iwoip .•>
a i959 ,p^t>e>n» ,nt?arrdT
.1965 ,d>^K>TP , in>ni^ini d^^o
.1962 .d^BT-p ,d»oaie> lap ,i»a>ip .•>
.iv-1 ,1964 ,n^nx-!?n ^^^KI^^M n3iaKn ni"T!?in
.889 ,111 n^Klptt .^3K , n 11pT>" ,'K^p .T
.1948 ,PIT T»3=T"in ,T>^in ,d>a;ie)n lap ^nnp TIT 'T
(P"T-0
.186-182 ,(1975) XXXIX 1331E>5 ,"mDfl1 D3^K" , "P^T .h
.11-1 ,1973-1969 ,d^W1T^ ,d^Qn I^IK , (>T1 .^1)
.1070-1066 ,IV n">K1pfl .^3K ,"niTTn d^fl"
,(1963) xiv dy5 1331^ /'JT'inyi nmin d^o"
.245-232
.1080-1070 ,iv twipn .^3K ,"nnT d»&n"
,1965 ,d>^K)iT> ,!?AP lap ."n^ii^ ni^nn d^n"
.179-151
,(1963) xiv dy» 1331E>!7 ,"'^ nuit>in»DKn ^y"
.202-197
.442-441 ,VI tl^KTpa .^3K ,"1ia"
11
Qy5 iJJig^ ,"d!?iyiT ni3nj>^ nm:iy niyawn" ,ITIT ^"n
.24-21 ,(1950)
.1962 ,d^u>iT-> ,niin n3tg» :d"3nT ,(iTiy)i»>i .a
BIBLIOGRAPHY 277

.87-83 ,(1934) vi i33iBfr .o^y^n >n3in flaxen

,D»Tifr>_n ni^jimp idn^nna p qo'P) DIDDI? oi*s!?a ,o>tny O K
. C'3-'K) I I I - I ,1973-1944 ,d^U>TT>
.299-291 ,C1931) III 1331^ ,"imiOn Q>3!lKn H10^" ,D^\y .3
.66-16 ,(1932) iv 1331^ /'iT'inyn o^nsn ninwV
,"0'»u3non d'jpnynni nniyn n")n 'w wmna nio'3t?" ,^ii?''3!7i|7e; ,v
.271-254 ,(1971) XXXV 1331^
This page intentionally left blank
ABBREVIATIONS
This page intentionally left blank
ABBREVIATIONS 281

ABBREVIATIONS

1. Old Testament Books*

Gen. Ex. Lev. Num. Dt. Josh. Judg. 1 Sam. 2 Sam. 1 Ki. 2 Ki. Isa.
Jer. Ezek. Hos. Jo. Am. Ob. Jon. Nah. Hab. Zeph. Hag. Zech. Mai.
Ps. Job. Prov. Song. Ru. Qoh. Lam. Esth. Dan. Ezf. Neh. 1 Ch. 2 Ch.
3
*Following the order of the BH

2. New Testament Books cited

Matt. Matthew
Luke
Rev. Revelation of St. John.

3. Tractates Cited from the Mishna, Talmudim, and Tosefta'*

Ab . Zar . Abodah Zarah Meg. Megillah

Bab . Kam . Baba Kamma Men. Kenahoth
Bab. Mets. (B.M.) Baba Metsi'a Midd. Middoth
Bekh. Bekhoroth Ned. Nedarim
Ber. Berakhoth Neg. Nega' im
Bets. Betsah Nidd. Niddah
Ed. Eduyoth Par. Parah
Erub. Erubin Shabb. Shabath
Gitt. Gittin Sheb. Shebi ' ith
Hag. Hagigah Snh . , Sanh . Sanhedrin
Hull. Hull in Sot. Sot ah
Kel. Kelim Sukk. Sukkah
m. Kat. Mo ' ed Kat an Tann. Ta'anith
Maas. Ma ' asroth Terum . Terumoth
Maas . Shen . Ma'aser Seni

a. Passages from the Mishna and the Tosefta' are cited by

tractate, passage and section.

b. Passages from the Babylonian Talmud are cited by tractate,

folio and page.
c. Passages from the Palestinian Talmud are cited by tractate,
chapter, folio and column.
d. Passages from the Midrashim are mostly cited by section
(Parashah), but sometimes according to the biblical verse
they refer to.
282 ABBREVIATIONS

4. Bibliographical Abbreviations

BASOR Bulletin of the American School for Oriental Research

BOB Brown, Driver, and Briggs, Hebrew and English Lexicon
to the Old Testament
3
BH Kittel - Kahle, Biblia Hebraica
CAD Gelb et al., Chicago Assyrian Dictionary
DB Hastings' Dictionary of the Bible
Enc. Jud. Encyclopaedia Judaica
Enc. Miq. Encyclopaedia Migra'ith (Biblical Encyclopaedia, in
Hebrew)
GB Gesenius and Buhl, Hebraisches und aramaisches
Handworterbuch uber das Rite Testament
G.-K. Gesenius' Hebrew Grammar (ed. E. Kautzsch, trans.
Cowley)
HSM Harvard Semitic Monographs
HThR Harvard Theological Review
HUCA Hebrew Union College Annual
ICC International Critical Commentary
IEJ Israel Exploration Journal
JAOS Journal of the American Oriental Society
JBL Journal of Biblical Literature
JCS Journal of Cuneiform Studies
JNES Journal of Near Eastern Studies
JSS Journal of Semitic Studies
KB Kfthler and Baumgartner, Lexicon in Veteris
Testament! Libros (1958)
3 n «»
KB Baumgarner et al., Hebraisches und aramaisches
Lexicon zum Alten Testament (1967)
NEB New English Bible (Oxford - Cambridge, 1970-1976)
OIL Old Testament Library
PEQ Palestine Exploration Quarterly
RB
(Rev. Bib.) Revue Bibligue
VT Vetus Testamentum
VTS Supplements to VT
ZAW Zeitschrift fur die alttestamentliche Wissenschaft
ZDMG Zeitschrift der Deutschen Morgenl'andischen
Gesellschaft
ABBREVIATIONS 283

5. Abbreviations - sources and general

ace. according to
adj . adjective
Af. Aph el
Akk. Akkadian
Ant. Flavius, Antiquities of the Jews
Aram. Aramaic
Bab,, B. Babylonian (Talmud)
bib. biblical
cf. confer, compare
ch(s). chapter (s)
denom. denominative verb
diet. dictionary
f. feminine
gent. gentilic (name)
Heb. Hebrew
Hif. Hiph°il
Hithp. Hithpa el
Q
Hof. Hoph al
intrans . intransitive (verb)
£
Ithp. Ithpe el
Jon. Pseudo Jonathan
K Ketib
M(asc) . masculine
MH Mishnaic Hebrew (Hebrew of the Mishna, Talmudim,
Midrash, and Qumran)
Mid. Midrash
Mish., M. Mishna
MS(S) Manuscript (s)
MT Masoretic Text
n(n). note(s)
n(om) . Noun, nominal form
Q
Nif. Niph al
Nithp. Nithpa°el
n.pr. proper noun
NT New Testament
0, T° Onkelos, Targum Onkelos
opp. in opposition to, as opposed to
284 ABBREVIATIONS

OT Old Testament
P(p). page(s)
Pa. Pacel
part. participle
pass. passive verb form
Pesik. R. Pesikta Rabbati
Pi. Picel
PI. plural
Phoen. Phoenician
Pu. Pucal
Q Qere
1Q Isaa Isaiah: first scroll from Qumran, Cave I
1Q H Hymns: the Hymns scroll from Qumran, Cave I
Rab(b). Rabba (Gen. Rabb., Ex. Rabb., Lev. Rabb., Num. Rabb.)
Sam. Samaritan
sg. singular
Sif. Sifra
Syr. Syriac
T Aramaic Targum
Tal. Talmud
Tos. Tosefta1
trans. transitive (verb)
Ug. Ugaritic
V Vulgate
v(v). verse(s)
verb. verb formation
VS(S) Version(s)
Yer. Talmud Yerushalmi (Palestinian Talmud)
LXX Septuagint

6. Sigla

reconstructed base form

generation, shift
o n a n t a l root sequence
shift (semantic or phonetic)
INDEX 285

I. INDEX OF BIBLICAL PASSAGES

1. Old Testament

(ch.) (v.) (PP-) (ch.) (v.) (PP-)

GENESIS (GEN. cont'd.)
1; 30 101, 237 (n. 15) 30: 35 39, 83-5,
2: 5 101 121-3, 169
7 161 37 85-6, 92, 162
12 251 (n. 23), 254 231 (n. 4)
(n. 10) 40 39, 121-3
19 161 41-42 169-70
25 256 (n. 10) 31: 4-18 169-70, 113
3: 1, 7 256 (n. 10) 10,12 113
19 161 33: 19 157
6: 17 251 (n. 34) 34: 157
9: 3 101 36: 13, 17 251 (n. 15)
6 161 23 252 (n. 71)
10: 28 252 (n. 71) 33 251 (n. 16)
11: 3 257 (n. 18) 36 114, 252
14: 2 158 (n. 55)
15: 15 162 38: 1-2 251 (n. 41)
22: 9 83 28-30 143-4, 157
23: 8-17 251 (n. 26) 40: 16 40, 133
8 44, 116, 251 (n. 36) 41: 42 90, 234
25: 4 251 (n. 28) (n. 89)
8 162 42: 38 163
9 44, 116, 251 (n. 26, 44: 29 163
36). 46: 10 44,116, 251
25: 25 39, 53, 127-9 (n. 36)
29 60 13 140, 143
30 35, 52, 58-62, 80, 49: 11 76, 82, 86-7,
156, 172, 227 (n. 9) 114, 141, 162
34 60 12 49, 72, 82,
27: 1 134 86-8, 119-20,
30: 29-43 169-170 130-31, 168,
32 39, 121-3 233 (n. 59),
33 39, 121-3, 169 240 (n. 61),
34-37 157 243 (n. 33)
286 INDEX

(GEN. cont'd.) (EX. cont'dO

49: 29 251 (n. 26) 28: 18 255 (n. 26,
50: 13 251 (n. 26) 32)
20 251 (n. 23),
EXODUS
254 (n. 10)
4: 6 82, 89, 168
31 146
5: 7, 14 257 (n. 18)
33 247 (n. 20)
6: 15 44, 116, 251
37 146
(n. 36)
30: 34 119
8: 17 257 (n. 16)
31: 2 246 (n. 4)
9: 22 101
35: 6 247 (n. 20)
10: 15 101, 163
7 110-12
16: 11 168
9 254 (n. 10)
14 201
23 110-12
31 88, 188, 225
25 247 (n. 20)
(n. 3)
27 254 (n. 10)
24: 10 166, 168 , 236
30 246 (n. 4)
(n. 120) , 255
35 247 (n. 20)
(n. 32)
36: Iff. 246 (n. 4)
25: 4 145, 247 (n. 20)
8 247 (n. 20)
5 110-12. 239 (n. 30)
19 110-12
7 251 (n. 23), 254
35, 37 247 (n. 20)
(n. 10)
38: 18, 23 247 (n. 20)
26: 1 23, 247 (n. 20)
39: 1, 2, 3 247 (n. 20)
4 146
5 257 (n. 14)
14 110-12
6 254 (n. 10)
28 146
10 254 (n. 7,21)
31 247 (n. 20)
11 255 (n. 32)
33 146
12 254 (n. 18)
36 150, 247 (n. 20)
13 254 (n. 10),
27: 16 150, 241 (n. 8),
255 (n. 33)
247 (n. 20)
24 247 (n. 20)
20 119
28 90
28: 5-6 247 (n. 20)
29 247 (n. 20)
8 247 (n. 20), 257
34 110-12
(n. 14)
9 254 (n. 10) LEVITICUS
15 247 On. 20) 1: 14 192
17 254 On. 7, 21), 13-44: 188-9
255 (n. 24) 13: 3 38, 88, 90
INDEX 287

(LEV. cont'dO (NUM. cont'd.)

13: 6 134-6 11: 7 82, 88, 141,
10 53, 234 (n. 74) 168, 225 (n. 3),
19 14, 44, 93, 106, 233 (n. 67)
128-130, 225 8 163
(n. 2), 234 12: 10 82, 89, 168
(n. 76) 19: 1-2 62ff.
21 134-6 2 23, 62-5, 69,
24 44, 106-7, 129-30, 71, 80, 260
225 (n. 2), 234 (n. 18)
(n. 76) 1-10 64
25 53 4, 5 65
26, 28 134-6 22: 4 101
30 7, 53, 97, 103-4 26: 13 240 (n. 1),
31 7, 53, 95, 97 251 (n. 17, 20)
32, 36 7, 53, 97, 103-4 20 251 (n. 20)
37 7, 53, 95, 97 23 140, 143
38 106 58 251 (n. 34)
39 88, 93, 126, 31: 8 157, 249 (n. 71),
134-6, 170, 225 251 (n. 47),
(n. 2), 238 (n. 6) 252 (n. 80)
42 44, 106, 128, 32: 3, 36 170, 252 (n. 82)
129-30, 225 (n. 2), 33: 20-21 158, 252 (n. 60)
234 (n. 76)
43 14, 44, 106, DEUTERONOMY
129-30, 142, 225 11: 10 101
(n. 2), 234 (n. 76) 29 252 (n. 70)
49 43, 55, 100, 106, 14: 5 162
124 22: 6 253 (n. 19)
56 134-6 24: 8 156
14: 4, 6 143 27: 2-4 45, 155
37 43, 55, 100, 106, 4, 13 252 (n. 70)
124, 128 28: 22 253 (n. 42)
49, 51-2 143 32: 14 76, 86, 162,
24: 2, 7 119 244 (n. 47)
34: 7 134
NUMBERS
3: 18, 21 251 (n. 34) JOSHUA
4: 6-7, 12 146 2: 18, 21 143
13 145 7: 17 251 (n. 20)
288 INDEX

(JOSH, cont'd.) (JUDG. cont'd.)

8: 30, 33 252 (n. 70) 21: 19 252 (n. 62)
9: 4 246 (n. 15)
5 169 1 SAMUEL
10: 29ff. 252 (n. 60) 2: 14 163
13: 21 157, 249 (n. 71) 3: 2, 13 135
251 (n. 47) 13 : 17 252 (n. 59)
27 170, 252 (n. 82) 17 : 42 39, 53, 127-9
29 252 (n. 80) 23 : 19 86, 158
15: 7 158, 251 (n. 50) 26 : 1 158
19 251 (n. 27) 3 86, 158
18: 17 158, 251 (n. 50)
23 252 (n. 59) 2 SAMUEL
27 159, 249 (n. 72), 1: 24 143-4, 229 (n. 71)
251 (n. 48), 252 6: 10 156
(n. 80) 13 : 15 78
19: 13 158 23 251 (n. 27)
26 159 15 : 23 252 (n. 75)
33 158 23 : 11 60
46 159
21: 10 252 (n. 60) 1 KINGS
2: 6, 9 163
JUDGES
37 252 (n. 75)
5: 6 108
5: 3 162
10 44, 116-18
7: 13-14 246 (n. 9)
30 140, 147
19, 22, 26 253 (n. 22)
6: 11 252 (n. 59)
10 : 23 78
19 163
11 : 1, 17 156
24 252 (n. 59)
15 : 13 252 (n. 75)
37 232 (n. 26)
21 : 2 101
7: 25 251 (n. 43)
8: 3 251 (n. 43) 2 KINGS
26 145, 146 3: 16-19 65
27 252 (n. 59) 22 23, 65-6, 80, 161,
32 162, 252 (n. 59) 168, 227 (n. 9),
9: 5 252 (n. 59) 260 (n. 18)
10: 1-2 140, 143 23 65-6
16: 4 114, 158 4: 38-39 60
18: 21 257 (n. 21) 5: 27 47, 82, 89, 168
INDEX 289

(2 KINGS cont'd.) (ISAIAH cont'd.)

9: 30 141, 151 , 255 19: 9 40, 114, 126, 133
(n. 30)
149, 180
12: 22 112
24: 23 253 (n. 17)
14: 1 112
27: 2 162, 244 (n. 47)
19: 26 101
9 155
23: 4, 6, 12 252 (n. 75)
29: 22 40, 126, 133, 180
30: 22 257 (n. 14)
ISAIAH
26 253 (n. 17)
1: 1 112, 157
32: 4 120
11-15 75-76, 91
33: 12 155
16-17 76
37: 27 101
18 30, 39, 41,
40: 4 256 (n. 9)
45-7 , 49, 57,
42: 3-4 135
65, 75-6 , 78,
44: 13 154
80, 82-5 , 90,
49: 2 235 (n. 111)
91-2 , 119, 138,
50: 3 163
143, 145 , 168,
54: 11 116, 249 (n. 1),
180, 229 (n. 71),
255 (n. 30, 32)
240 (n. 61), 247
12 254 (n. 13)
(n. 1) 55: 1 87
18-20 76, 91
58: 11 120
2: 1 112
59: 5 253 (n. 19)
3: 16 151- 2, 186, 198
61: 3 135
5: 2 114, 252 (n. 52)
63: 1-3 66-9 , 77, 230
6: 2 162
(n. 90)
13 241 (n. 35) 1 71, 131, 239 (n. 27)
7: 22 233 (n. 57) 2 65, 66-9 , 80, 196
9: 1 253 (n. 37) 4 66
10: 14 253 (n. 19)
26 251 (n. 43) JEREMIAH
11: 6 170-1 2: 6 253 (n. 37)
13: 12 167 21 114, 162 , 252 (n. 52)
15: 6 101, 252 (n. 83) 4: 11 29, 120, 241 (n. 35)
16: 6 170
28 163
8 114, 162 , 240
30 143, 151 , 229 (n. 71)
(n. 62) 255 (n. 30)
18: 4 29, 120, 241 5: 6 170-1
(n. 35)
8: 21 253 (n. 31)
290 INDEX

(JER. cont'd.) (EZEK. cont'd.)

10: 9 145 26: 4, 14 120
12: 14 101 16 150 , 249 (n. 74)
13: 16 253 (n. 37) 27: 7 146 , 147, 150, 248,
23 47, 150, 168, (n. 23)
170-1 13 79
14: 2 253 (n. 31) 16 145 , 150, 254
17: 9 256 (n. 8) (n. 13), 255 (n. 26)
22: 14 141, 153-4 18 44, 116-7, 159
23: 12 108 24 146 , 149-50
30: 6 55, 254 (n. 43) 28: 13 254 (n. 7, 21)
31: 40 252 (n. 75) 255 (n. 24, 26, 32,
35: 230 (n. 95) 33)
36: 18 154 18 255 (n. 26, 32)
46: 20 108 20 161
48: 34 252 (n. 83) 32: 7 163
40: 3 142
EZEKIEL 47: 16, 18 159
1: 142
9-10 167 HOSEA
16 168, 255 (n. 33) 4: 13 231 (n. 4)
26 166, 168 6: 8 256 (n. 8)
8: 2 142 7: 9 162
10: 142 11: 4 239 (n. 33)
1 166- 8, 255 13: 7 170
(n. 32) 14: 8 253 (n. 22)
9 168, 255 (n. 33)
16: 10, 13 150 JOEL
16 169 1: 5 233 (n. 58)
18 150 7 41, 91-2, 232
17: 3 147, 150, 249 (n. 42)
(n. 70) 2: 2 95, 97, 134, 163,
21: 12 135 236 (n. 2)
23: 13 254 (n. 10, 21) 6 163
14 153-4 4: 13 228 (n. 47)
40 130, 151, 190, 18 87
211
41 257 (n. 15) AMOS
24: 7-8 120 2: 1 155
INDEX 291

(AMOS cont'd.) (ZECHARIAH cont'd.)

4: 7 164 6: 3 42, 69, 71, 83, 85,
9 164, 253 (n. 42) 112-14, 169, 228
5: 8 253 (n. 37) (n. 60)
7: 10, 12, 14 112 6-7 228 (n. 60)
9: 13 233 (n. 58f.) 6 83, 85, 95, 98,
169
NAHUM 7 42, 85, 112-14
2: 4-5 Ill 11: 17 135
4 67, 110-12, 138,
143, 168, 225 MALACHI
(n. 2) 3: 14 163
11 163
3: 12 243 (n. 2) PSALMS
19 135 6: 8 135
23: 4 253 (n. 36)
HABAKKUK 35: 10 231 (n. 96)
1: 8 170 37: 2 101
44: 20 253 (n. 36)
ZEPHANIAH 45: 1 253 (n. 22)
1: 2-3 161 15 147, 150, 249
4: 5 253 (n. 22) (n. 74)
12-14 168 51: 9 30, 41, 82-3,
91-2, 119, 168,
HAGGAI 180, 226 (n. 6)
2: 12 60 19-21 92
17 253 (n. 42) 60: 1 253 (n. 22)
61: 1 253 (n. 22)
ZECHARIAH 65: 14 169
1: 1-17 112-14, 228 68: 7 120
(n. 57ff.) 11 131
8 23, 42,68, 69-71, 14 43, 55, 100, 109-10
80, 83, 85, 113, 124-25, 167, 183
114-15, 240 24 126, 131
(n. 62) 28 147
6: 1-8 70 69: 24 135, 245 (n. 3)
2 23, 69-71, 80, 85, 73: 6 169
95, 98, 112-14 7 118-20, 142
292 INDEX

(PSALMS cont'd.) (JOB cont'd.)

75: 9 126, 162, 244 39: 14 253 (n. 19)
(n. 47), 253
(n. 14) PROVERBS
83: 12 251 (n. 43) 8: 19 167
104:15 192 IS: 17 101
107:10, 14 253 (n. 36) 30 231 (n. 96)
119:127 167 16: 24 231 (n. 96)
139:15 249 (n. 73) 20: 29 162
147:16 84 21: 8 108
22: 29 236 (n. 8)
JOB 23: 29 44, 86, 130-31,
3: 5 163, 253 (n. 35) 168
5: 11 253 (n. 31) 31 76-8, 80, 86
10: 21-2 253 (n. 35) 31: 21 143, 229 (n. 71)
12: 22 253 (n. 35) 22 145, 229 (n. 71)
13: 27 141
16: 16 126, 131-2, 253, SONG OF SONGS
(n. 14, 35) 1: 5 35, 73, 79, 95-6,
17: 7 135 97, 98, 108, 110,
24: 17 253 (n. 35) 121, 158, 168, 181,
28: 3 16, 253 (n. 35) 231 (n. 101)
6 255 (n. 32), 6 39, 73, 79, 95, 97,
16 254 (n. 10), 98, 108, 121, 181,
255 (n. 32) 190-91
17 162 14 153
19 254 (n. 21) 2: 1 253 (n. 23)
30: 28 163 2, 16 253 (n. 22)
30 35, 95-6, 99, 3: 10 145
121, 164, 181, 4: 3 46, 138, 143, 145,
231 (n. 98), 247 (n. 19)
236 (n. 15) 5 253 (n. 22)
33: 11 141 8 170
34: 22 253 (n. 35) 13,14 153
37: 21 188, 234 (n. 71) 5: 9-16 71-73
38: 14 246 (n. 14) 10 29-30, 71-4, 78,
17 253 (n. 35) 80, 97, 118-20,
39: 8 19, 40, 54, 253 (n. 7)
100-102, 172 11 95, 97-8, 167-8
INDEX 293

(.SONG OF SONG:S cont'd.) ESTHER

5: 12 168 1: 6. 40, 133, 145-6,
13 253 (n. 22) 149, 167, 255
14 255 (n. 32, 33) (n. 38)
15 255 (n. 38) 8: 15 40, 90, 133, 145-6,
6: 2, 3 253 (n. 22) 149, 234 (n. 89)
7 247, (n. 19)
10 253 (n. 17) DANIEL
7: 3 253 (n. 22) 4: 12, 20,
6 23-5, 46, 138, 22, 30 141, 193
143-5 5: 5 155
10 230 (n. 86) 7, 16 145
12 153 21 141, 193
8: 2 233 (n. 58) 29 145
7: 9 91, 133, 168, 180,
QOHELETH 231 (n. 13)
9: 8 90-91, 119, 147, 10: 6 142, 255 (n. 33)
196, 244 (n. 10) 11: 35 41, 92-3, 119, 180,
11: 10 95, 163 226 (n. 6)
12: 11 257 (n. 13) 12: 10 92-3, 112, 226
(n. 6)
LAMENTATIONS
1: 15 228 (n. 47) EZRA
20 132 2: 45 157
2: 11 132 7: 4 251 (n. 21)
4: 1 236 (n. 23) 8: 4 251 (n. 21)
5 143 27 7, 54, 56, 103-4,
7-8 14, 29-30, 42-3, 125, 167, 168, 184
49, 72-3, 78-9,
80, 82, 95-6, 97, NEHEMIAH
98-9, 118-20, 121, 1: 1 157
126-8, 134, 142, 4: 7 120
163, 168, 225 7: 48 157
(n. 2), 236 10: 2 157
(n. 15, 20), 245 11: 12 112
(n. 3), 255 12: 25 257 (n. 12)
(n. 32) 42 251 (n. 22)
5: 17 135, 245 (n. 3)
294 INDEX

1 CHRONICLES (1 CHRON. cont •d.)

1: 22 252 (n. 71) 29 : 2 150, 166 , 249
33 251 (n. 28) (n. 70), 250 (n. 1)
37 251 (n. 15) 254 (n. 10), 255
40 252 (n. 71) (n. 38)
44 251 (n. 16) 28 162
47 114, 252 (n. 55)
2: 43-44 249 (n. 71), 251 2 CHRONICLES
(n. 47), 252 2: 6 '143
(n. 80) 12-13 246 (n. 9)
4: 7 44, 116, 240 13 143, 145 , 247
(n. 1), 251 (n. 21), 248
(n. 36) (n. 29)
17 251 (n. 28) 3: 14 143, 145 , 247
21 140 (n. 20, 21)
24 240 (n. 1), 251 9: 22 78
(n. 17) 13 : 19 251 (n. 27)
5: 14 251 (n. 39) 14 : 8 251 (n. 19)
24 251 (n. 28) 25 : 14 156
32 251 (n. 21) 29 : 16 252 (n. 75)
6: 2, 5 251 (n. 34) 30 : 14 252 (n. 75)
6 251 (n. 18)
14 251 (n. 34) 2. New Testament
28 253 (n. 42)
31 112 MATTHEW
36 251 (n. 21) 17 : 2 235 (n. 114)
7: 1, 2 140, 143
3 251 (n. 22) LUKE
16 249 (n. 71), 10 : 30 251 (n. 50)
251 (n. 47), 252
(n. 80) REVELATION
11: 13 60 6: 85, 113, 232
32 251 (n. 40) (n. 35)
15: 16 252 (n. 75) 6: 2 239 (n. 47-49)
27 90 14 : 19 228 (n. 47)
24: 27 157 19 : 15 228 (n. 47)
26: 15, 17 257 (n. 12)
27: 8, 11, 13 251 (n. 20)
INDEX 295

Jr. INDEX OF RABBINIC LITERATURE PASSAGES

1. Mishna

(Tractate) (Passage) (PP.) (Tractate) (Passage) (PP-)

Ber. 1:2 185, 191 Neg. l:2f. 180, 238
Sheb. 6:1 238 (n. 2)
(n. 5) ^1:5 180
6:3 183 2:7 136, 188,
7: Iff. 193, 199 190
Terum. 4:8 181 4:7 193
Maas. 1:3 181 Par. 2:5 64
Shabb. 8:4 198 3:5 62
8:5 252 Nidd. 2:6-7 187
(n. 1) 2:7 200
12:4 200, 246 6:7 179
(n. 17) 9:6 193, 246
Yoma 3:7 197 (n. 17)
Sukk. 3:6 20, 190
4:9 155 2 . Tosefta '
Ned. 3:8 181
Sot. 3:3-4 183 Maas.Shen. 5:13 195
3:4 20 Neg. 1:2 221 (n. 19)
Gitt 2:3 154 1:4 190
Bab . Kara . 9:4 196 1:5 180
Ed. 5:6 183
Hull. 1:5 192 3. Talmud Babli
3:3 20
7:2 7 Ber. 31b 117
11:2 190 36b 190
Bekh. 7:6 187, 190 Shabb. 30a 164
Midd. 2:3 184, 192 64a 198
Kel. 15:2 191, 194 107b 150
Neg. 1:1-2 83-4, 89, Erub. 53b 193
106-7, 54b 116
110, 129, Yoma 7a 90
136, 186, 75a 188, 233
188, 201 (n. 67)
296 INDEX

Bets. 15a 201 4 . Tal . Yerushalmi

Taan. 19a, 21 a 254
(n. 44) Sukk. III, 53d 129 , 191
Meg. 13a 124, 191 238
M. Kat. 24b 7 (n. 5)
Bab. Mets. 29b 90 Shabb. VIII, lOc 179
59a 180 Hag. II, 77d 164 , 190
Snh. 14a 198
105a 7
Sheb. 6a ff. 189 5 . Midrashim
Ab. Zar. 38b 227
(n. 23) (Book) (Passage) (PP .)
Men. 86a 241
(n. 29) Gen. Rabb 1 194 , 200
Hull. 7b 184 12 181
22b 192 36 190
47b 185 Ex. Rabb. 42 190
88b 201 Lev . Rabb 5 192
89a 146, 185 12 187
135b 190 Num . Rabb 2 165 , 193
Bekh. 45b 187, 190 12 248
Sot. 17a 146, 185 (n. 22)
Gitt- 19a 20.0 27 179
Nidd. 19a ff. 187 Pesik. R. 20 192
19b 191 Sifra 14:2 238
20a 200 (n. 5)
24b 193