A.

Teeuw
The history of the Malay language. A preliminary survey

In: Bijdragen tot de Taal-, Land- en Volkenkunde 115 (1959), no: 2, Leiden, 138-156

This PDF-file was downloaded from http://www.kitlv-journals.nl

Downloaded from Brill.com10/14/2020 03:39:39PM

via free access
THE HISTORY OF THE MALAY LANGUAGE.*
A PRELIMINARY SURVEY.
T
I have called this lecture a preliminary survey, because it is a curious
J ^ fact that in spite of the prolonged and periodically concentrated
attention that was bestowed on the Malay language by many investi-
gators, the history of this language was yet hardly ever the subject of
scientific research, and in a time too when linguistics in general was
predominantly a historical science. It is. true that ever since Werndly's
time, i.e. since the beginning of the eighteenth century, it has always
been felt as all but obligatory to precede any description of the Malay
language with an introduction of the history of those who speak
Malay, and of its spread as a means of communication over a more
extensive area. And generations of Malay scholars have no less faithfully
accounted it their duty to give an enumeration of the foreign languages
which have contributed towards the enrichment of the Malay lexicon,
with many examples of loan words from each of those languages. But
the history of Malay in its more proper sense, that is to say its origin
and evolution in all its various forms in the many and diverse commu-
nities in which in the course of history it served as a means of
communication (including the mutual relationship of those kinds of
Malay), has so far remained an almost completely unmapped territory.
Not only has not even so much as an outline of a historical grammar
of the Malay language ever been published — by which I do not mean
to suggest that the existence of a historical grammar is a guarantee
of an adequate description of the history of a language! — it is even
difficult to discover from publications devoted to Malay that the
history of that language can be discussed with any profit.
It is naturally possible to point to causes showing why Malay
has so long been studied without any regard to its history. It cannot
be denied that for centuries a uniform and rigid brand of literary
Malay has been very wide-spread. To those who confined themselves
to literary texts the idea of taking a historical view could easily remain
* Text of a lecture, delivered on Leiden University Day ("Universiteitsdag"),
February 7, 1959.

Downloaded from Brill.com10/14/2020 03:39:39PM

via free access
 THE HISTORY OF THE MALAY LANGUAGE. 139

a remote one. Nevertheless it seems to me that the idea of the

"remarkable invariability" of Malay, which Emeis recently recalled,1
was based on prejudice rather than on objective observation, even
taking written Malay only. The attitude of Dutch as well as of English
scholars with regard to Malay has often been highly normative; every
kind of Malay that they encountered was measured according to the
norm of a particular sort of literary Malay, and everything that was
not in accord with it was condemned as incorrect. No wonder .that
what remained was curiously uniform and invariable.
This normative evaluation resulted in all sorts of forms of Malay
being for many years almost entirely neglected, forms which might
have been of great importance to the study of its history. In the first
place there is spoken Malay, or rather all the spoken forms of Malay
which were or still are used in various parts of Malaya, Sumatra,
Borneo and even further afield, in Jakarta, the Moluccas and Western
New Guinea. Authorities on Malay were generally no less disapproving
of all written Malay which did not tally with the norm, because for
instance it was "spoiled" by foreign influence, such as Persian, Arabic,
Javanese and so forth; and disapproving meant uninterested. That it
was inconsistent faithfully to sum up loan words from all sorts of
divergent languages and to be silent on the possibility of those same
foreign languages influencing the grammatical structure of Malay
apparently escaped their notice. It almost goes without saying that from
this normative point of view all those forms of language, often also
called Malay, which serve as a means of intercourse between those
whose native tongue is Malay and those to whom it is a strange
language (non-Malay Indonesians or strangers), or among the latter,
were all too easily termed Low Malay or Pasar Malay, and usually
found no favour with scholars of the Malay language.
In the last few decades a number of circumstances have gradually
helped to change the situation we have just been describing, in which
the question of the history of Malay never so much as cropped up.
In the first place an astonishing historical prospect has been opened
up by the deciphering of a number of inscriptions from the seventh
century and later, which from the start have been recognized and
described as Old Malay. Investigators of Malay have lately evinced a
greater interest in various forms of Malay which fell outside the
1
M. G. Emeis, Bijdrage tot de vergelijking van het moderne Melaja-Maleis
en de Bahasa Indonesia, Bingkisan Budi, Feestbundel Van Ronkel, Leyden
1950, p. 117. ;

Downloaded from Brill.com10/14/2020 03:39:39PM

via free access
 140 A. TEEUW.

scope of previous investigators. The gradual spread of more modern

views among the scholars of Indonesian linguistics has naturally
contributed much towards the new state of affairs. It resulted in
normative thinking about language in general and Malay in particular
being replaced by a more unbiassed observation of the facts. Partly
on account of this interest was aroused in all sorts of texts which had
partially been preserved in old manuscripts written in a different
brand of Malay from the literary standard kind, namely older or locally
differentiated. Moreover the evolution of Bahasa Indonesia in Indo-
nesia, and, although at the moment to a lesser degree, of modern Malay
in Malaya, has forced people to take a fresh view of the relationship
between norm and reality and as a corollary to inquire into the history
of the Malay language.
Although it can certainly not be said that there is a superabundance
of new data, yet sufficient publications have recently appeared to form
a good opportunity for devoting our attention to the history of Malay.
It seemed to me especially attractive to do so in a lecture on 'University
Day', which in the first place is meant for alumni and for a wider
circle than for philologists only, and which aims at giving an idea of
the progress research has made in a given field of knowledge. Within
that framework recent literature of different kinds will come up for
discussion. But it will by now be clear that the present address can be
no'more than an exploration of a field hitherto but little investigated.
The time allotted me being short I cannot enter into the problems of
the development of Malay as one Indonesian language among numerous
others, but must confine myself to dealing with the history of Malay
in historical time. For the same reason I must pass over the relation-
ship between Malay and Bahasa Indonesia in silence.

As has already been said chronologically speaking the most important

new prospect that has been opened to the study of Malay is the
discovery and decipherment of a number of inscriptions. They can
with certainty be dated back to the second half of the seyenth century,
and their language has from the beginning been termed Old Malay.
The result is that the period known to us as having Malay sources has
now become at least twice as long as heretofore. High lights in the
history of the discovery of Old Malay have so far been an excellent
publication of four inscriptions by Coedes in 1930,2 and the brilliant
2
G. C°edes, Les inscriptions Malaises de Qrivijaya, BEFEO 30 (1930),
pp. 29—SO.

Downloaded from Brill.com10/14/2020 03:39:39PM

via free access
 THE HISTORY OF THE MALAY LANGUAGE. 141

decipherment of a big stone by De Casparis in 1956,3 which had

previously been considered wholly illegible. In addition a few Old
Malay inscriptions of that same period or of a bit later date have come
to light, curiously enough originating in part from Java.
The name Old Malay has without more ado been accepted for the
language of these old inscriptions, but for the rest very little attention
has been paid to the relationship between Old Malay and Malay. In
his Grammatik der Bahasa Indonesia which appeared in 1956 Kahler
inserted a chapter (not unoptimistically called Abrisz der Grammatik
des Altmalaiischen),4 in which he summarizes a number of morpho-
logical and syntactic peculiarities in the language used in the inscript-
ions. But the relationship to Malay remains undiscussed, even though
he gives seemingly comparable Malay forms by the side of the Old
Malay ones. So far it is really only Aichele 5 who has gone deeper into
the problems of this Old Malay and its relationship to other Indonesian
languages, especially Old Javanese, the hypothetical Old Batak, and
Malay. The name of the language of the inscriptions "die man mit
Recht als 'Altmalaiisch' bezeichnet hat" 6 presents no problems to him
either. But he does' go into the differences between this language and
later Malay, and into their explanation. One of the questions he
propounds is how to explain that a number of affixes which occur in
the inscriptions do not appear in later Malay, or at least not in the
same form. Interesting examples are the causative maka-, which is
not known in Sumatran languages, but is known in Old Javanese and
other IN languages, the suffix -a, which seems to correspond in function
with the Old Javanese -a, still to be found in Javanese too, and also
well known from the Philippine languages; and then the prefixes
nt- and mar-, which are familiar in similar forms and functions in
various IN languages, but which in later Malay correspond to di- and
bar-. I am going a bit further into the relationship ni-/di- and mar-/bar-
because they raise fairly important questions of principle. Aichele sees
two possibilities for explaining the difference between Old Malay and
Malay: "(es) g i l t . . . . zu prufen, ob es sich dabei um Entlehnungen

3
J. G. de Casparis, Selected inscriptions from the 7th to the 9th century A.D.
Prasasti Indonesia II, Bandung 1956, pp. 1—46 and pp. 344—353, and cp.
pp. 207—211.
4
H. Kahler, Grammatik der Bahasa Indonesia, Wiesbaden 1956, pp. 22—29.
s
W. Aichele, Die altmalaiische Literatursprache und • ihr Einfluss auf das
Altjavanische, Zeitschrift fur Eingeborenen-Sprachen, XXXIII (1942—1943),
pp. 37—66.
6
Aichele, p. 39. . :.

Downloaded from Brill.com10/14/2020 03:39:39PM

via free access
 142 A. TEEUW.

handelt oder um Sprachgut, das dem Malaiischen im Laufe seiner

Entwicklung verloren gegangen ist". 7
A third possibility for explaining the difference between Old Malay
and Malay suggests itself, at least at a first glance, when Old Malay
ni- and mar- are compared to Malay di- and bar-t for it would only
be natural in a formal sense to interpret it as a not unusual type of
sound change in the IN languages, viz. a nasal turning into a homor-
ganic occlusive. De Casparis, who did not know Aichele's article or who
did not have it at his disposal in Indonesia, apparently thought of that
explanation and even believed that he recognized the same correspon-
dence in other cases as well.8 His lexical identifications of muah with
Malay buah and of malun with Malay belum seem to me unacceptable
for semantic reasons, however.9 In any case they are inadequate as an
argument for proposing a sound change nasal > voiced occlusive.
Aichele apparently rejects this view of a sound change and gives
different explanations regarding the two seemingly corresponding cases
ni- oo di- and mar- ~oo bar-. In- the former case he believes that we
have to do with two elements, namely the well-known prefix ni-, which
is usually historically identified with the infix -in- and the preposition
di-, which in historical time took over the function of that old prefix. 10
In the latter case he believes that a mar-Jbar- isogloss runs through
Western Indonesia, which must be of an early date as appears from
the distribution of those forms among the languages. Therefore "Wir
ko'nnen nicht annehmen, dasz mal. ber- im Laufe der Sprachentwick-
lung aus mar- entstanden sei" 1 X : the only alternative is to suppose

1
Aichele, p. 40.
8
' D e Casparis, p. 24 f.
8
De'Casparis himself already points to Coedes' statement that the occurrence
of vuah (=: buah) in the same inscription makes the identification of muah
with Malay buah doubtful, to_ say the least. The many places where muah
occurs make the explanation given by Coedes and Aichele (p. 61) seem more
likely. (:=Minangkabau muah, etc.). Malun, the identification of which with
Malay belum De Casparis considers 'hardly doubtful' (p. 24) or 'probable'
(p. 40, note 30), is translated by De Casparis himself in one of the four places
in which it occurs (line 11, 'in order to') in a way which is very far removed
from Malay belum, while the other places are doubtful too. The translations
of kadaci kamu mati malun. mamrurua (line 10) by "if you die (?) before
having succeeded in destroying (my palace)1 (?)", and of manalit mas mani
malun mamrurua kadatumku (line 11) by "spend gold and jewels in order to
destroy my keraton" are not very probable, especially when considered side
by side.
10
Aichele, pp. 46—52.
" Aichele, p. 45. . . . .

Downloaded from Brill.com10/14/2020 03:39:39PM

via free access
 THE HISTORY OF THE MALAY LANGUAGE. 143

that borrowing took place, and for various reasons Aichele considers
Batak the original source.
I must say that I think his whole argument unconvincing. In the
first place I seriously object to his too easy explanation of the Malay
prefix di- coming from the preposition di, and in the second place to
the very sharp distinction he' makes between the prefixes di- and ni-.
On investigating other languages it appears that by the side' of the
prefix ni- there also often occurs in closely related languages or
dialects a corresponding' prefix di-; this is especially the case in
Batak, but it is a variation also to be found in languages on Celebes.12
Moreover, in the light of comparative linguistics it is difficult to isolate
the preposition di so strictly from the whole group ni-/ i-/ di-/ -in-
as Aichele does. In other words it seems hardly likely that the Malay
prefix di- in the conjugated form should historically go straight back
to the preposition di (which is assuredly connected with ni-, etc.), and
should show no connection with for instance di- as prefix in the con-
jugated forms in Toba and' Mandailing Batak. No, if mar- should be
a. borrowing from Batak in Old Malay then little can be said against
explaining ni- in the same way and a great deal is to be said for "it.
By all this I do not at allmean to say that I am of opinion that in
the case of mar- Aichele rightly speaks of a borrowing from Batak,
and that the same explanation is also preferable for ni-. The basic
mistake he makes in his explanation of mar- and also in the whole of
his survey of Old Malay in my opinion lies in the fact that he tries to
explain it from the point of view of Malay. It is a mistake which can
already be noticed in his otherwise important.article about the study
of Old Javanese in 1929.13 It is highly curious that-in considering the
difference between seventh and seventeenth century Malay he does
indeed think of the possibility of Old Malay having borrowed (appa-
rently assuming that there was some other language before that time
which corresponded to classical Malay on the points in question), but
that he does not so much as mention the logically and historically much
more probable possibility of a renewal after 700. One wonders whether
even so critical, shrewd and independent-minded an investigator as
13
H. N. van der Tuuk, Tobasche Spraakkunst, Second Part, Amsterdam 1867,
§ 159, 3°, note 318 f. and the places mentioned there. N. "Adriani, Spraakkunst
der Bare'e-taal,' VBG 70 (-1931), p. 95'note:- See also R. Haaksma, Inleidihg
t o t d e studie der vervoegde vormen in de Indonesische talen, thesis Leydeh
1933, p. 6 ff. - • . . . . . .
13
W. Aichele, Grundsatzliches zur . Kawi-Interpretation, Feestbundel Koninklijk
Bataviaasch Genootschap Volume I, Weltevreden 1929, pp. 1—21. -

Downloaded from Brill.com10/14/2020 03:39:39PM

via free access
144 A. TEEUW.

Aichele was yet unconsciously so bound to a norm, to the invariability

of classic Malay, that he could only look upon differences in seventh
century Malay as deviations from the classical norm, and whether the
idea of the later forms being rejuvenations of the older ones really
never occurred to him.
This preliminary assumption makes Aichele propound his theory of
borrowings amongst those seventh century languages, but I believe that
one cannot be too careful in such a matter. Morphologically for instance
the combination of two facts such as the absence of an Old Malay
element in Malay and its presence in present-day Batak is no sufficient
reason for concluding that the element was therefore borrowed in Old
Malay from Old Batak. And all this quite apart from the fact that the
borrowing of affixes, although not quite impossible or unknown, is yet
hot a very usual thing.
But my main objection is a question of principle and of method.
As long as there are no compelling indications that point in another
direction it is better to consider morphological differences between Old
Malay and Malay as an indication that seventeenth century Malay is
not directly descended from seventh century Old Malay apparently. All
sorts of complications are possible. Classical Malay may date directly
back to some other Malay dialect already in existence in the seventh
century (geographically it is of importance that the inscriptions come
from South Sumatra, whilst classical Malay seems to have a much
more Northern cradle).Or there may have been continuity in the literary
tradition of Malay, but far-reaching influences on the part of other
spoken Malay dialects may yet have made themselves felt in the course
of those ten centuries.
Lexically too Aichele believed he could point to distinct traces of
Batak influence in the Old Malay of the inscriptions. On the one hand
one may take it that the borrowing of a word occurs more easily than
borrowing an affix. But on the other hand the single fact that an
Old Malay word which is missing in present-day Malay occurs in
present-day Batak is by no means a conclusive reason for assuming
it to have been a loan word in Old Malay that was taken from Old
Batak. It would only be right to come to such a conclusion if specific
characteristics of a phonetic, morphological or semantic kind distinctly
stamp it as a non-Malay and markedly Batak word. Only one of all
Aichele's supposed loan words started by giving me the impression of
fulfilling the specific conditions mentioned above. I thought Aichele's
identification of parban in parbanda, "dieser crux interpretum", with

Downloaded from Brill.com10/14/2020 03:39:39PM

via free access
 THE HISTORY OF THE MALAY LANGUAGE. 145

Karo-Batak perban ingenious,14 even though the form looked rather

anachronistic. In view of the fact that the uncontracted form bahan still
seems to be normal in Karo-Batak it is difficult to take for granted that
thirteen centuries ago Malay had already taken over this word in its
contracted form. To crown all the decipherment of the serpent inscrip-
tion by De Casparis has added three new parbanda instances to our
repertoire, at least two of which make Aichele's identification with
Karo-Batak perban an impossibility.15 There is nothing much for it
but to go back to the very oldest interpretation of the word as given
by Bosch, who sees an indication of a title in it, although the etymology
is not clear.16 To my mind at least the interpretation suggested by
De Casparis is riot very probable.17
Summarizing I believe the linguistic data of the seventh century
inscriptions which have thus far become known can be recapitulated
as follows: the language of these inscriptions in many very characteris-
tic respects resembles Malay: phonologically it is striking that (as far
as the data go in the non-Sanskrit words) the language only has three
vowel phonemes besides the e, namely a, i, and u (e and o only occur
in Sanskrit words). The consonant system does not show any impor-
tant differences with Malay, unless the occurrence of a series of cacu-
minal consonants by the side of dental, palatal, and velar consonants
should count as such. De Casparis, however, considers it highly probable
that it is merely a peculiarity in spelling. Moreover the b is lacking —
another difference of which it will be difficult to prove that it it more
than a question of spelling. The correspondences of van der Tuuk's
rt and r 2 are also typically Malay. Morphologically there are all sorts
of things that correspond to classical Malay: nominal derivations with
par-an, ka-an, pa-n&s-an, -an; sa-; and verbal ones with -i and -kan,
and ma- with nasalization (though with a formal peculiarity in mamawa
which is considered irregular in classical Malay but which is very
frequent in old Malay mss.). Among the well-known Malay formatives
we do not come across ter-. Besides the affixes ni- and mar-, and maka-
and -a, already mentioned above, the prefixes um- and mi- are unknown
in Malay, although the occurrence as affix of the three mentioned

14
W. Aichele, Die altm. Lit. spr., pp. 53—55.
18
De Casparis, p. 34 line 15, and p. 36 line 26.
16
In Ph. S. van Ronkel, A preliminary Notice concerning two Old Malay
inscriptions in Palembang (Sumatra), Acta Orientalia II (1924), pp. 14—15.
1T
De Casparis, p. 46 note 83: Sanskrit parva{n) + -nda.
Dl. 115 10

Downloaded from Brill.com10/14/2020 03:39:39PM

via free access
 146 A. TEEUW.

last is not wholly certain.is it is striking that all four of these affixes
are found in Old Javanese but that three of them only occur in a petri-
fied form in Javanese.
From a lexical point of view also this language presents all sorts of
peculiarities, which it is true are not one and all only Malay, but which
taken together do definitely make a very Malay impression, as for
instance the pronouns aku, kamu, dia, and kita (for "you" — but that
is also found e.g. in the Hikayat Acheh 1 9 and their short forms -ku
(but also -nku, which is not Malay but which is Old Javanese for
instance), -(m)amu (later -MM), -na and -ta (not Malay, but again Old
Javanese). The prepositions di, dari, ke, denan (not pada), are very
striking, and equally remarkable are yan, ada, tida, janan, and ini (but
inan instead of itu). The numerals that have been found so far show
some deviations: telu and sapulu dua instead of duabelas, which is still
Karo-Batak 2 0 but which may well have had a wider circulation'in the
seventh century. — All in all it must be admitted that this language '
rightly bears the name Old Malay, at least in the sense that it is not
related to any other present-day language so closely as to Malay.
The next source for the history of Malay is an inscription' in Kedu
on Java which had already been known for some time but which was
not available in a reliable form until De Casparis took it in hand. 21
Unfortunately the inscription is not only fairly short but a number of
possible or probable place names occur in it, and Javanese at that,
which is not exactly the easiest material from a linguistic point of view.
That the language is just as Malay as that of the previously men-
tioned inscriptions is almost as well-established a fact as that it goes
back to the middle of the ninth century. Phonologically it strikes us in
the first place that sapopo occurs twice, a word which etymologically
tallies with Malay sepupu. How to explain the 6 remains the question,
but that here we clearly have to do with an IN word containing an o
presents an important difference with the South. Sumatran material.
Another noteworthy point is that there is a form ampa for four; "a
dialectal form for ampat" De Casparis calls it; 2 2 but is that so certain?-
18
De Casparis (p. 352) does not consider umamgap to be an um-iorm. He does
mention the prefix mi- in miayu-ayit (p. 348), however.
19
T. Iskandar, De Hikajat Atjeh, thesis Leyden = VKI 26, The Hague 1959,
p. 197.
20
Here again to my mind Aichele's conclusion goes too far and too fast
(pp. 52—53).
21
J. G. de Casparis; Inscripties uit de Cailendra-tijd, Prasasti Indonesia I =
thesis Djakarta 1950, pp. 50—73.
82
De Casparis, p. 71.

Downloaded from Brill.com10/14/2020 03:39:39PM

via free access
 THE HISTORY OF THE MALAY LANGUAGE. 147

As far as I can see it is the only word where the final consonant one
would expect is lacking. De Casparis is apparently thinking of an early,
trace of the change from -t into -q, later to become familiar, which can be
seen in Malay dialects (Kelantan and Trengganu) and in Minangkabau.
But is it not rather an abbreviated form of a numeral, to be compared
to abbreviations of the names of months, coins and the like in OJd
Javanese inscriptions {po for pon, ma for masa) ? It should furthermore
be noted that both the b and the w occur here, just as in Malay,;whereas
in the seventh century only the w is found. In an Indian alphabet it is
difficult to make out, however, whether such a distinction is more than
a spelling variation only. Morphologically also this ninth century.mate-
rial is limited. In any case tnaka-, par-an, and ma- nas. occur. Curious^
but also dubious is tarkalaut, a combination of tar- and ka-- as De
Casparis suggests ? Particularly important in connection with what has,
been discussed above is the occurrence of a di- form which" should
probably be understood as a classical verbal di- form, 23 whereas ni- is
lacking. No less remarkable is war- (or, = ? bar-) in waranak, which
occurs three times, besides which marhyan is also to be found, however,
probably as a title. Particularly these last peculiarities give the impres-
sion that this language is closer to classical Malay than is seventh
century Old Malay. De Casparis suspects that some other Malay dialect
formed the basis of this inscription. I also am inclined to" look for its
cradle nearer to the later centre of the Malay world. It is naturally
impossible for me to enter further into the question of the origin of
this particular brand of Malay in the interior of Java within the frame-
work of this'lecture. De Casparis has at least made it probable that
the inscription was promulgated at a time that radical changes were
apparently taking place, and he suspects that "the use'of Old Malay in
this charter may be understood as a kind of demonstration manifesting
the origin of the vamga to which the Rakarayan Patapan belonged".24
This Malay inscription is not an isolated case in Java. De Casparis

28
It says: ycmg rajya dirakfa iya sabanakna ycmg dega itas = tatah.... (line 8).
De Casparis translates it as: ,Het rijk dat door hem beschermd wordt—
zoveel streken als er zijn in alle richtingen ..'..' (The kingdom that is
protected by him — as many regions as there are in every direction. ...')•
Apparently he takes iya to be the 'agens' going with the di-iorm: I am more
inclined to think it refers back to the kingdom: 'The kingdom, it is protected,-
etc'; I do so chiefly because iya as a non-honorific pronoun can hardly refer
to the sovereign; for other highly placed personages the inscription uses sida,
whilst iya in other places in this inscription is apparently used in a non-
honorific way.
94
De Casparis, p. 200.

Downloaded from Brill.com10/14/2020 03:39:39PM

via free access
 148 A. TEEUW.

mentions another one of one line, and Bosch published yet another of
three lines from Western Java dating back to 942, which for more than
one reason is very interesting.^ Although it contains hardly as much as
five Malay words it nevertheless furnishes us with both a mar- and a
bar- form, and therefore almost seems to have been preserved for the
direct purpose of tempting philologists to advance hypotheses. Aichele
certainly could hardly withstand the temptation to put forward these
curious forms in support of his hypothesis and to surmise that "das
volkssprachliche bar- habe zu dieser Zeit begonnen, neben dem Fremd-
ling mar- in die Schriftsprache einzudringen und literaturfahig zu
werden, bis es schlieszlich die war-Form vollig verdrangte". 26 Mean-
while bar- (war-) already appeared to be "literaturfahig" in Central
Java as early as 842, so that for the time being, as long as no fresh
data are forthcoming, there seems to be little else to do than to record
the curious fact of two variants of a prefix appearing side by side in
one form of language which otherwise only occur complementally in
IN languages.
The next remnants of Malay are of much later date, and this time
they come from the heart of Minangkabau, from Pagarruyung, where
right in the middle of a collection of Sanskrit verses there is an inscrip-
tion of the year 1356 dedicated to the well-known Adityawarman, which
contains a piece of prose entirely Malay in structure, even though the
number of Sanskrit words is again particularly great. This piece of
Malay has never yet formed the object of careful research, 27 but it
seems certain that linguistically it will present very little that is new.
Phonologically a word with e (rentak) by the side of uran is striking;
morphologically two bar- forms are important as well as the word
diparbuatkan, entirely like later Malay. Kopadrazva may be a form with
the Old Malay prefix ka-, traces of which are still to be found in later
Malay. A question which arises is whether in this Malay from Minang-
kabau characteristics already occur which distinguish later Minang-
kabau from Malay. It is not very clear — there are some particulars
which might be interpreted thus (inan, rabut, handak), but others are
definitely not Minangkabau (tyada, lemah).
The most interesting fact about this inscription is really that it lends
a5
F. D. K. Bosch, Een Maleische inscriptie in het Buitenzorgsche, BKI 100
(1941), p. 49-53.
88
Aichele, p. 46.
87
The stone has been transcribed by N. J. Krom in het Oudheidkundig
Verslag van de Oudheidkundige Dienst in Nederlandsch-Indie, Weltevreden—
's-Gravenhage 1912, p. 51f.

Downloaded from Brill.com10/14/2020 03:39:39PM

via free access
 THE HISTORY OF THE MALAY LANGUAGE. 149

support to the obvious surmise that Old Malay remained in use as a

written language right up to the end of the Hindu Sumatrari period,
just as in Java it was the case with Old Javanese. For we have to do
here with the period in which the Islam had long ago penetrated
Sumatra. This is confirmed by an inscription from Pasai, in Acheh,
from 1380, a curiosity because it comprises a Malay poem in Indian
writing and metre with various Arabic words and Moslem ideas. 28
The text teaches us next to nothing about the so-called internal history
of Malay; but it is important proof, together with other data, that in
1380 in Pasai, an area in which at the present day Achehnese is the
local language, Malay was apparently considered the right literary
language for a commemorative inscription on a dead queen, and that
Indian writing was the vehicle.
No less interesting is an inscription that comes from Trengganu
dating from that same century, but this time in Arabic characters, which
seems to proclaim Islam as the State religion.29 What is most interes-
ting to us in connection with the present subject in this rather mutilated
inscription is that the language is largely identical with classical Malay,
and that where it differs it seems to correspond to the present Treng-
ganu dialect, "although we should speak with the greatest caution with
a view to the scanty data.
We are gradually approaching the period in which literary Malay
comes into our field of vision, even though there is still an annoying
gap in our knowledge. It is true that we assume that some of the oldest
texts that have been preserved go back to the fifteenth century, perhaps
even to the fourteenth, but the oldest manuscripts are from as late as
the second half of the sixteenth century, at least cannot be proved to
be earlier, while the manuscripts of some of the most interesting texts
are of an even later date. And, unlike the Balinese copyists of the Old
Javanese literature, who preserved the original texts "with surprising
fidelity and accuracy century after century, the copyists of Malay
literature not only set to work rather carelessly, they often even seemed
to deem it their duty and even their honour to purify their material,
adapting it to the requirements of the day and smoothing out everything

28
W . F. Stutterheim, A. Malay sha'ir in Old-Sumatran characters of 1380 A.D.
(with 4 plates), Acta Orientalia 14 (1936), pp. 268—279; and
G. E. Marrison, A Malay Poem in Old Sumatran Characters, J M B R A S 24
(1951), pp. 162—165.
29
H . S. Paterson, An early Malay Inscription, from Trengganu, arid C. O.
Blagden, A Note on the Trengganu Inscription, J M B R A S 2 (1924), pp. 252—
258 and pp. 258—263.

Downloaded from Brill.com10/14/2020 03:39:39PM

via free access
 150 A. TEEUW.

they considered to be an inequality. That means that for the investig-

ation of the history of Malay it is the age of the manuscripts that
happen to have been preserved, which primarily determines the value
of a given literary work and not the age of the text. To name a few
examples: a Malay translation of an Arabic panegyric, although it can
only be used with the greatest circumspection for linguistic studies in
view of the peculiarities of such translations, is nevertheless an impor-
tant source for our knowledge of sixteenth century Malay, owing to
the fact that it has been preserved in a manuscript which was in any
case written before 1600.30 But an original history of Pasai, which in
all likelihood was written as early as the fifteenth century, is only
known to us in a nineteenth century manuscript 3 1 and, except for a
few lexical relicts (kutaha 32 ) which slipped through the meshes of the
purifying net, has been entirely made to conform with the classical
ideal as far as language is concerned.
• For there is such a thing as classical Malay which serves as standard
and ideal for Malay literature. As tradition will have it is always
identified with Malay as spoken in Johor and the Riau-Lingga Archi-
pelago. 33 . To my mind that identity is only relative. Very little is
known about the Malay spoken at Johor or the islands facing it. And
the little we do know creates the impression that the distance between
the spoken language of those parts and written literary Malay is no
smaller than between West European cultural languages and their
local, spoken counterparts.
This identification should be interpreted in such a way that the
fixation of literary Malay, in the form in which we know it to-day took
place at the Malay courts of Johor, Riau, etc. Naturally local Malay
made its influence felt on literary Malay in the same way as official
Malay in the Netherlands Indies came about in the twentieth century
via the pens of authors from Minangkabau, giving it a Minangkabau

30
G. W . J. Drewes, Een 16de eeuwse Maleise vertaling van de Burda van
Al-Buslrl (Arabisch lofdicht op Mohammed), V K I 18 (1955).
31
Hikajat radja-radja Pasai, edited by E. Dulaurier in the Collection des
principales chroniques Malayes, Paris 1849. J. P . Mead's romanized edition
in J S B R A S 66 (1914), pp. 1—55, cannot be profitably used without keeping
a close check on the text.
32
The word that is spelled k-t-alif-h (cp. Drewes, register op. cit. sub ketaha)
seems to me most probably t o be kutahu, a worn-out form of 'how-do-I-know',
Malay = gerangan after interrogatives. Van Ophuysen I believe already
took it to mean that.
33
E.g. Ch. A. van Ophuysen, Maleische Spraakkunst 2 , Leyden, 1915, pp. ,3—4.
R. O. Winstedt, Malay Grammar 2, Oxford 1927, p. 30, cp. p. 75.

Downloaded from Brill.com10/14/2020 03:39:39PM

via free access
 THE HISTORY OF THE MALAY LANGUAGE. 151

tinge, but not enough to entitle us to say that modern Malay is entirely
similar to Minangkabau. We can continue this comparison- and say
that Minangkabau authors did not create literary Malay out of thin
air any more than the authors at the courts of Johor and Riau created
a literary Malay tradition straight off. We are reasonably sure-that
literary traditions existed at the older Malay courts. We know this
for certain of Acheh in the sixteenth and seventeenth centuries, of
Malacca in the fifteenth century, and of Pasai in the fourteenth and
fifteenth centuries. Both on the ground of internal evidence furnished
by Malay literature and on the ground of general considerations and
comparison with Java we may assume that the Malay literary tradition
must be much older, and that even in pre-Moslerh times there must
have been an important Malay literature.
The question in connection with our subject now is whether anything
is left of that older literary Malay. From what has been said above it
follows that we can only expect to find traces of it in the not very
numerous old manuscripts. Van Ronkel, who was the first to concern
himself with such manuscripts, answered the question in the affirmative.
On the strength of all sorts of peculiarities in a number of manuscripts
dating from not later than 1600 he concludes that "the language of the
manuscripts presents another, in this case older, period. than that of
the common Malay". 34 Recently Drewes raised the question anew, on
the occasion of his publication of one of the manuscripts dealt with by
Van Ronkel. He remarks that all sorts of deviations from classical
Malay in the language of the manuscript can now readily be explained
on account of the advance in Malay lexicography, and through the in-
crease in our knowledge of Malay dialects. He consequently terms Van
Ronkel's conclusion of 1896 no longer acceptable.?5 As can be seen
from his comments further on in the book on the language of the
manuscript he believes that most of the deviations from standard Malay
are to be ascribed to the Perak dialect,3^ in itself not an incongruous
supposition when we think of the close relations which existed between
Acheh and Perak in the sixteenth and seventeenth centuries.
In his review of Drewes' book Voorhoeve has rightly pointed out that
the fact that dialectal peculiarities are to be found in the Malay Burda
text by no means precludes our having to do with an older literary

34
Ph. S. van Ronkel, Account of six Malay. manuscripts of the Cambridge
University Library, BKI 46 (1896), p. 25. .
35
Drewes, op. cit. p. 10.
38
Drewes, op. cit. pp. 40—41.

Downloaded from Brill.com10/14/2020 03:39:39PM

via free access
 152 A. TEEUW* :

Malay, whether it be the direct predecessor of classical Malay from

Johor or not. 37 The very fact that the particulars Drewes describes
are peculiar to a large number of manuscripts is an argument in favour
of considering this language on its own merits, and for not taking the
later literary Malay as a starting point for describing this older literary
Malay language. A general characteristic that strikes us in studying
this written Malay is that the link with the pre-Moslem world is still
so clearly visible. It already appears from the spelling, e.g. the use of
the Arabic duplication sign with a consonant to indicate a preceding
pepet can only be explained as a continuation of a similar spelling in
Indian writing. 38 As has already been observed by various scholars, 39
all sorts of Sanskrit loan words are closer to their counterpart in the
language of origin than in classical Malay, both as regards form (it is
irrelevant in this connection whether it is only a spelling tradition or
a difference in pronunciation) as well as meaning. That this should be
so in late sixteenth and seventeenth century Acheh is all the more
striking when we remember that Moslem influence had prevailed there
for centuries on end. Morphologically also these texts show peculiarities
of which the most interesting, considered from the classical Malay point
of view, is the irregular nasalization and pre-nasalization. As Winstedt
remarks "These rules are fixed only in literary or Riau-Johor Malay
and even there with some few variants and exceptions". 40 In the pic-
ture displayed by the Malay dialects and languages geographically
contiguous to Malay the view that the fixed rules in literary Malay
are a peculiarity rather than that the whole of the rest are an exception
to a fixed rule seems more fitting. The variation of be- and her- in
contrast to the fairly fixed rules in classical Malay also deserve mention
in this connection.. Moreover in these texts other derivations occur
which are unknown in classical Malay. Drewes points to curious Re-
forms. 41
Taken all together there is reason enough therefore to assume that
clear traces of an older literary Malay have been preserved. A closer
and completer description cannot be given here. Further research will

37
P. Voorhoeve, in Museum, Tijdschrift voor Filologie en Geschiedenis 62
(1957), p. 38.
38
T. Iskandar, De Hikajat Atjeh, p. 11.
39
Van Ronkel, Drewes, Iskandar. Cp. also W. G. Shellabear's articles, An
account of some of the oldest Malay mss. now extant, JSBRAS 31 (1898)
pp. 107—151, and The evolution of Malay spelling, id. 36 (1901), pp. 75—135.
40
R. O. Winstedt, Malay Grammar, p. 75.
41
Drewes, Burda, p. 40.

Downloaded from Brill.com10/14/2020 03:39:39PM

via free access
 THE HISTORY OF THE MALAY LANGUAGE. 153

have to devote itself to the question as to whether these older manu-

scripts have all been written in an identical, in its turn rigidly fixed
literary Malay, or whether there are more relevant differences (for
instance between manuscripts from Acheh and other regions). A com-
parison of the vocubularies of the texts will certainly also have to be
undertaken. Another thing deserving of attention is the fact pointed
out by Voorhoeve that a man like Nurud-Din ar-Raniri when he
arrived in Acheh in 1637 had already learnt Malay elsewhere, and had
written texts in Malay, the language of which, except for the unavoid-
able Arabic influence, was classical Malay. 42 This again is an indication
that two forms of literary Malay are to be found side by side, and
partly in the same period of time, for are not the manuscripts from
Acheh with the peculiarities mentioned above coincident with Niirud-
Dln's apprenticeship in classical Malay? At least as far as can be
judged to-day Nurud-Din's language nowhere shows the pre-classical
peculiarities, although he was later reproached that his Malay contained
words from Bahasa Acheh which were not understood on Borneo. 43
Should this be looked upon as evidence of a concealed rivalry between
two literary forms of the Malay of the first half of the seventeenth
century, a struggle which in the end was naturally won by the Johor-
Riau party? For in Johor-Riau Malay as the mother tongue of the
population had always had deeper roots than in Acheh, where it was
only the cultured language of an elite — even though it remained so
right into the twentieth century.
Another important question is closely bound up with the afore-going,
but through lack of data it is not yet ripe enough to be answered, hardly
to be studied even, namely the relation between diverse other forms of
local literary Malay and standard Malay. There are texts from Palem-
bang, Bandjermasin, Kutai, Amboina, and many other regions in In-
donesia. To W. Kern especially we owe important material concerning
the Malay of Bandjermasin and Kutai'.44 But when he says of the
Salasilah of Kutai "De taal der kroniek kan men ruwweg karakteriseren
als litterair Maleis met Koetaise inslag", (the language of the chronicle
can roughly be characterized as literary Malay with a Kutai tinge to

42
V. Voorhoeve, Van en over Nuruddin ar-Raniri, B K I 107 (1951), p. 357, cp.
G. W . J. Drewes, De herkomst van Nuruddin ar-Rariiri, B K I 111 (1955),
p. ISO.
43
P. Voorhoeve, ib. p. 360.
44
W . Kern, Aantekeningen op de Sja'ir Hemop (Sja'ir Kompeni Welanda
berperang dengan Tjina), T B G 82. 2—4 (1948), p. 211—257.
W . Kern t, Commentaar op de Salasilah van Koetai, V K I 19 (1956).

Downloaded from Brill.com10/14/2020 03:39:39PM

via free access
 154 A. TEEUW.

it), 4 5 the interesting question remains unsolved as to whether that

literary Malay is identical with the standard Malay of Riau-Johor
after having undergone a secondary Kutai metamorphosis, or whether
Kutai always had its own brand of literary Malay, or at least whether
some other literary Malay had-already existed there which in this text
has partly been adapted to seventeenth century literary Malay. And the
same question may well be asked about all the other written regional
forms of Malay. An answer will only be possible after systematic and
intensive research. And here again it seems to me entirely in place to
give warning that it is not certain without examination that seventeenth
century literary Malay should be the starting point for explaining all
the peculiarities, and the possibility of different and possibly older
Malay traditions should be taken into account.
In mentioning Nurud-Dln just now another fundamental problem
cropped up for scholars in,the history of Malay as a literary language,
namely the problem of the Arabic influence. We know something about
this influence since Van Ronkel's article appeared in 1899.46 In Bing-
kisan Budi, the volume of studies devoted to Van Ronkel, Drewes again
drew attention to it. Moreover he himself has made a fresh contribution
to the knowledge of Kitab-Malay, 47 and particularly through his public-
ation of the Malay translation of the Burda he has furnished fresh
material towards the study of this problem. 48
In spite of all this not much more than a start has been made. The
question most interesting to our subject, namely how far literary Malay
was influenced by Arabic idiom, has really not yet been systematically
examined. For Winstedt's continual testing of the language of Malay
literary products with his idiomatic standards cannot be considered as
a scientific contribution to the evolution of literary Malay. When he
remarks that most of the Moslem teachers who translated foreign texts
into Malay were themselves foreigners, "who in their difficult task of
translation murdered Malay idiom", 49 one might just as well argue
that these people "created Malay idiom" — naturally a different Malay
from the kind that was dear to Winstedt's heart, but a literary Malay

45
W . Kern, Commentaar, p. 13.
46
Ph. S. van Ronkel, Over invloed der Arabische syntaxis op de Maleise, T B G 41
(1899), pp. 498—528.
47
G. W. J. Drewes, De herkomst van het voegwoord bahwasanja. Bijdrage tot
de kennis van het Kitab-Maleis. Bingkisan Budi, Leyden 1950, pp. 104—116.
48
See note 30.
« R. O. Winstedt, A History of Malay Literature, J M B R A S 17. 3 (1940), p. 93.
Cp. also his Preface, and passim.

Downloaded from Brill.com10/14/2020 03:39:39PM

via free access
 THE HISTORY OF THE MALAY LANGUAGE. 155

which answered a need that was widely felt, and which also strongly
influenced non-religious Malay in untranslated texts. It is naturally
the right of every student of literature-to feel a preference for some
particular use of language, or for some particular style, and to object
to foreign influence. But I do not believe it can be denied that in ap-
plying such a literary and esthetic standard the linguistic investigation
into the evolution of Malay has been and still is "considerably hampered.
In Dutch studies of Malay this normative attitude with regard to Malay
has been taken up by Van Ophuijsen, and his influence, through the
Minangkabau guru bahasa Melayu, still makes itself felt in Indonesia
to the present day. Not that any Indonesian will deny the reality of
a strongly altering language, but up till how in Indonesian educational
circles there has been a strong tendency to judge every change from
the point of view of the old norm. To my mind the influence of the
normative viewpoint of Winstedt and other older English scholars on
the attitude towards developments in modern Malay in Malacca is
quite as strong. Up to the present day the practice of pointing out
so-called non-Malay idiom in older texts forms an important part in
the teaching of Malay. It is a pastime that is all the more pointless
since it is often done without any knowledge of Arabic or Persian, but
simply intuitively, according to fixed shibboleths. Only an open-minded
attitude in observing and interpreting facts in the Malay language
which at present are considered inferior or uninteresting will make it
possible to gain a subtler insight into the history of the language.
The same problems which crop up in studying foreign influence on
Malay also arise in considering the question of the relationship between
official written Malay and its many spoken forms, particularly the
dialects which geographically lie further off; even though the relation
of Riau and Johor Malay to literary Malay is hot without its problems
either, as has already been said..A history of Malay should take these
relationships into account. But here again the strongly normative way
of thinking has impeded Malay scholars from gaining a correct
insight into the nature of the problems, and has hindered the progress
of research. Brown's recent republication of a number of dialogues in
three Malay dialects from the Malay Peninsula furnishes a striking
example of this attitude. 50 It is undoubtedly a useful and valuable book,
because it gives a great deal of interesting material. But its presentation
and elucidation is characteristic: these Malay dialects are described and
60
C. C. Brown, Studies in Country Malay, London 1956. Cp. my review in
BKI 113 (1957), pp. 293—297. '

Downloaded from Brill.com10/14/2020 03:39:39PM

via free access
 156 A. TEEUW.

phonetically rendered from the stand-point of Malay as written by the

English — and in addition they are judged according to those
standards, they are even given marks "judged by Sejarah Melayu
standards" and so there is a difference in quality among those different
dialects.
Not until such a normative approach has been definitely abandoned
will the investigation of Malay dialects be really fruitful, also as regards
the study of the history of Malay. It will probably even become the
most important source for that purpose. For in Western Europe too
it was dialect research after the manner of language geography which
brought great increase and refinement into our knowledge of the history
of language. At least part of the Malay area satisfies a few of the
important conditions necessary to the success of an investigation into
the geography of dialects; there is sufficient historical continuity, and
the unifying influence of the modern Malay of Singapore has not yet
gone too far. Besides quite a fair amount is known about the history
of these regions. Difficult problems will undoubtedly arise if one should
want to extend such an investigation to Malay dialects in Sumatra and
Borneo. Quite recently Voorhoeve 5 1 on the one hand and Cense and
Uhlenbeck 5 2 on the other have repeatedly had to point to the difficulty
in defining the boundaries of Malay in their linguistic bibliographies
of those islands, and only detailed research on the spot will be able
to answer all sorts of hitherto unsolved problems. Certainly in the
coastal area of Borneo it will probably often be impossible to catch up
with the historical development on account of the numberless immi-
grations and shiftings that have taken place there since time immemorial
up to the present day. But the data which notably Brown's material
has already provided on the one hand and the experience I myself
gained in the investigation of the geography of the dialects of Lombok
on the other 5 3 together cause us to expect that a similar investigation
in Malaya and the Malay region of Sumatra, providing it is carefully
prepared and carried out on a somewhat larger scale, will be able to
shed a great deal of light on the history of Malay.

A. T E E U W

51
P. Voorhoeve, Critical survey of studies on the languages of Sumatra,
's-Gravenhage 1957, esp. pp. IS—20.
52
A. A. Cense and E. M. Uhlenbeck, Critical survey of studies on the languages
of Borneo. 's-Gravenhage 1958, esp. pp. 7—13.
33
A. Teeuw, Lombok. Een dialect-geografiche studie. VKI 25 (1958).

Downloaded from Brill.com10/14/2020 03:39:39PM

via free access