Read Anytime Anywhere Easy Ebook Downloads at ebookmeta.

com

The Explanation of Behaviour 1st Edition Charles

Taylor

https://ebookmeta.com/product/the-explanation-of-
behaviour-1st-edition-charles-taylor/

OR CLICK HERE

DOWLOAD EBOOK

Visit and Get More Ebook Downloads Instantly at https://ebookmeta.com

 The Explanation of Behaviour

“A vehemently interesting book.The philosophical part displays the most

remarkable grasp of the contemporary philosophical situation and its his-
torical roots… There is also a satisfactory absence of the tones and atti-
tudes of any particular philosophical school.”
Elizabeth Anscombe, The New Statesman

“A most valuable and systematic account of a major problem in the expla-

nation of behaviour. His arguments are both powerful and of the greatest
possible general interest.”
Times Literary Supplement

The Explanation of Behaviour was the frst book written by the renowned philoso-
pher Charles Taylor. A vitally important work of philosophical anthropology,
it is a devastating criticism of the theory of behaviourism, a powerful ex-
planatory approach in psychology and philosophy when Taylor’s book was
frst published. However, Taylor has far more to offer than a simple critique
of behaviourism. He argues that in order to properly understand human be-
ings, we must grasp that they are embodied, minded creatures with purposes,
plans and goals, something entirely lacking in reductionist, scientifc expla-
nations of human behaviour.

Taylor’s book is also prescient in according a central place to non-human

animals, which like human beings are subject to needs, desires and emotions.
However, because human beings have the unique ability to interpret and re-
fect on their own actions and purposes and declare them to others,Taylor ar-
gues that human experience differs to that of other animals. Furthermore, the
fact that human beings are often directed by their purposes has a fundamental
bearing on how we understand the social and moral world.

Taylor’s classic work is essential reading for those in philosophy and psychol-
ogy as well as related areas such as sociology and religion.

This Routledge Classics edition includes a new Preface by the author and a new
Foreword by Alva Noë, setting the book in philosophical and historical context.
 Charles Taylor is Professor Emeritus at McGill University,
Canada. The author of many books on social and political
philosophy, the philosophy of mind and language and the
history of philosophy, he is one of the best-known and widely
read philosophers in the world. He is also a prominent fg-
ure in Canadian politics and is a prominent voice in debates
about liberalism and multiculturalism.
Routledge Classics contains the very best of Routledge
publishing over the past century or so, books that have, by
popular consent, become established as classics in their
feld. Drawing on a fantastic heritage of innovative writing
published by Routledge and its associated imprints, this
series makes available in attractive, afordable form some of
the most important works of modern times.

For a complete list of titles visit:

https://www.routledge.com/Routledge-Classics/
book-series/SE0585
Charles
Taylor
The Explanation of Behaviour

With a new Preface by the author and

a new Foreword by Alva Noë

London and New York

First published in Routledge Classics 2021
by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
and by Routledge
605 Third Avenue, New York, NY 10158
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 1964, 2021 Charles Taylor
Preface © 2021 Charles Taylor
Foreword © 2021 Alva Noë
The right of Charles Taylor to be identifed as author of this work has
been asserted by them in accordance with sections 77 and 78 of the
Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this book may be reprinted or reproduced
or utilised in any form or by any electronic, mechanical, or other
means, now known or hereafter invented, including photocopying and
recording, or in any information storage or retrieval system, without
permission in writing from the publishers.
Trademark notice: Product or corporate names may be trademarks
or registered trademarks, and are used only for identifcation and
explanation without intent to infringe.
First published by Routledge & Kegan Paul Ltd 1964
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data
A catalog record has been requested for this book

ISBN: 978-0-367-70522-0 (hbk)

ISBN: 978-0-367-70521-3 (pbk)
ISBN: 978-1-003-14674-2 (ebk)
Typeset in Joanna
by codeMantra
TO ALBA
 Contents

FOREWORD TO THE ROUTLEDGE CLASSICS EDITION xi

PREFACE TO THE ROUTLEDGE CLASSICS EDITION xix
ACKNOWLEDGEMENTS xxvii

PART 1: EXPLANATION BY PURPOSE 1

1 Purpose and Teleology 3
2 Action and Desire 28
3 Intentionality 57
4 The Data Language 76
5 The Problem of Verifcation 104

PART 2: THEORY AND FACT 117

6 The Determinants of Learning 119
7 What Is Learned? 146
8 Spatial Orientation 172
9 The Direction of Behaviour 209
10 The Ends of Behaviour 234
11 Conclusion 286

INDEX 293
 Foreword to the Routledge
Classics Edition

Philosophy, like art, when it is serious and vital, does not go out of
date; but as the situation changes, the sources of its signifcance may
vary too. This applies to the book you are reading now. When The Ex-
planation of Behaviour was published in 1964, behaviorism was still the
dominant approach in academic psychology (especially in the United
States), and this notwithstanding Chomsky’s negative review of Skin-
ner’s Verbal Behavior some fve years earlier (Chomsky 1959). Assault
is not too harsh a word to describe the pile-driver-like intensity of
Taylor’s dismantling of behaviorism’s absurdities (in the book’s sec-
ond half); this is concealed not at all by the restrained style of High
Analytic Philosophy in which the book is written.
But philosophical readers uninterested in exhuming behaviorism’s
remains, or coming better to understand its rise and fall, will need
to look beyond the book’s surface preoccupation with behaviorism,
to glimpse its truer concern. And this a topic no less important today
than it was then, namely, the question whether it is possible to do
this thing we call science when it is we ourselves that would be its
xii FOREWORD TO THE ROUTLEDGE CLASSICS EDITION

object of study. It would be unfair to say that the proposition that sci-
entifc psychology, or neuroscience, or cognitive science, are, at least,
possible, is a dogma of the scientifc world view. It would be better to
say that this claim is a basic upshot of a prior commitment to ration-
ality itself. If we can ask questions, well then, we can use the meth-
ods of science—the gathering and sifting of evidence, the testing of
hypotheses—to work out answers. And yet the project of a science of
the human, or that of a rigorous anthropology, faces immense and all
too well-known obstacles. These have to do with consciousness and
subjectivity, value, freedom, and culture, all of which refuse, till now
anyway, to let themselves be regimented by what is often thought to be
demanded of a scientifc world view, namely, Reductive Materialism.
This is what Taylor’s The Explanation of Behaviour concerns itself with;
this is what Taylor’s recent books (Dreyfus and Taylor 2015; Taylor
2016), written nearly half a century later, on realism in science, and
on language, continue to concern themselves with. And it is thanks to
its engagement with this theme, second to none in fnal value, that The
Explanation of Behaviour recommends itself to contemporary readers.
But there are other considerations one might mention as well.
Philosophy books bashing science are a dime a dozen. Books that
manage to do so in ways that scientists themselves can be expected,
indeed demanded, to pay attention to, are much rarer.Two striking ex-
amples of this are Dennett’s Consciousness Explained (1991), and of course,
Merleau-Ponty’s The Phenomenology of Perception (1945/1964). Now Taylor’s
is a book of this kind. I shudder to think what it must have been like
to be one of the thinkers that came under his unfinching and untiring
examination in the book’s Part Two. But there are other connections be-
tween Taylor and this pair of authors that should be noticed.
Consider frst that the outlook and ideas of the Oxford philosopher
Gilbert Ryle, whose Concept of Mind was published in 1949, are never
far from the surface of The Explanation of Behaviour; and Ryle, whose infu-
ence over the whole feld was enormous, was to be Dennett’s Oxford
mentor in the years immediately following the publication of Taylor’s
book. Taylor’s own remarks (64) about “intentional systems,” that is,
about systems whose intelligibility requires viewing them with an eye
 FOREWORD TO THE ROUTLEDGE CLASSICS EDITION xiii
to their purposes, comes close to anticipating not only Dennett’s ter-
minology, but also central elements of his theory.
Taylor’s attack on behaviorism is strikingly Rylean. For example, it
zeros in on the tacit reliance of behaviorism on a conception of the
mental as occult and interior, and so as beyond the reach of empiri-
cal science. Science, so the behaviorist believes, on Taylor’s analysis, is
confned to the study of “colorless movement” because the mind itself
is supposed to be inaccessible to direct observation and controlled
experiment. It is illuminating to be reminded, and ironic, that behav-
iorism leaves intact and indeed relies on just the way of thinking about
our mental lives which it so earnestly aims to discredit. And in all this
it is hard not to recall Ryle’s famous attack on Cartesianism and the
“Ghost in the Machine” (Ryle 1949).
As for Merleau-Ponty, consider this: The Phenomenology of Perception is
not cited, by my count anyway, in any of the major works on language,
perception, or mind published by philosophers or other scientists
writing in English in the four or fve decades following its publica-
tion in France at the end of World War II. To name only a few of the
authors whose books or papers do not mention Merleau-Ponty: Ryle,
Austin, Quine, Putnam, Davidson, Kenny, Chomsky, Strawson, Marr,
Fodor, Gibson, Dennett, and Cavell. We might throw in Wittgenstein
and Carnap for good measure, although their most important writings
were not in English.
The exception to this general neglect is the book you have before
you now, which, as it happens, was published the very same year
that Merleau-Ponty’s great book was fnally translated into English.
The Explanation of Behaviour, while in no way about Merleau-Ponty—he
is only mentioned a couple of times, in footnotes—is, one can say,
thoroughly informed by the latter’s insight and spirit.There are echoes
of Merleau-Ponty in Taylor’s excavation of the underlying empiricism
that motivates behaviorism, and that convinced so many that only be-
haviorism, with its reductive refusal to countenance goal, purpose,
intention, let alone consciousness or desire, had any chance of rising
to the level of rigorous science. Like Merleau-Ponty, Taylor argues that
we cannot explain what matters to people or animals, and why they
xiv FOREWORD TO THE ROUTLEDGE CLASSICS EDITION

do what they do, in terms of the regular law-like association of mere

elements, whether these are conceived of as physical stimuli and bod-
ily movements, or as sensations. And for the simple reason that it is
only against the background of what is already relevant and important
that we can pick out what might serve as stimulus, response, or sen-
sation, in the frst place. Taylor’s close analysis (in Chapter Seven) of
futile attempts to constrain and defne what might count as a stimulus,
or a cue, brings this point home.
Or take the phenomenon discussed in the key chapter on Spatial
Orientation. For the behaviorist, the animal’s oriented behavior—
knowing how to make its way to a goal, for example—is explained
by supposing it to issue from learned associations of stimulus and
response. The animal learns to go right at one cue, and left at another.
But this doesn’t capture the phenomenon to be explained. Genuine
orientation, Taylor insists—and Merleau-Ponty might have made the
same point—is not particular, determinate, and explicit in these ways.
It is open and fexible. As Taylor writes: “To be oriented … is to be
able to dispense with directions, or rather, to know more than can be
set down in any list of directions” (177). Knowing your way around
a maze, or a box, or a feld, or whatever, “is a more general capacity
to get around, to go from any point to any point in the environment.”
It “cannot be expressed in a list of facts known about the environ-
ment.” And so crucially, it is not a form of “knowing that,” as Taylor
notices, interweaving these Merleau-Pontyan insights with those of
Ryle, but rather a kind of “know-how” (176). Strikingly, as Taylor’s
analysis demonstrates, the behaviorist account falls short not only
when it comes to human beings, but also in its efforts to comprehend
the minds of rats.
And when Taylor argues, in Chapter Two, that desire is not only in-
tentional, in the philosopher’s sense, but that it is also, as it were,
motivated, we can detect more of this productive interweaving of ana-
lytic philosophical themes with insights from Phenomenology. Ryle’s
dispositionalism—itself a variation on the theme of behaviorism, al-
though non-reductive—informs Taylor’s idea that desire and its object
do not stand in a relation of, as it were, contingent correlation. The
 FOREWORD TO THE ROUTLEDGE CLASSICS EDITION xv
desire is for the water; if that were not its target, it would not be the
desire that it is. But this observation has a crucial upshot for how we
think about the explanation of behavior. Taylor argues that it is your
desire for water itself that explains why you drink; you don’t need a
further explanation of how or why the desire manages to actuate you
to drinking behavior; there is literally no room for further explanation
here, causal, or otherwise. What would stand in need of explanation
would be the fact of a person who desires water but does not drink
the water before him or her. Now this idea that the desire is, as it were,
intrinsically motivating, is closely related to Merleau-Ponty’s thought
that the body, and our attitudes, are governed by pre-cognitive forms
of intentionality such as what he called “motor intentionality” (mo-
tricité). Crucially, for Merleau-Ponty, we typically do not need, nor can
we make room for, the appeal to deliberation, thought, or planning to
bridge a gap between desire and action.
Taylor also appeals, at different junctures, to what I think of as the
principle of the primacy of the situation. With this, Taylor goes be-
yond the familiar observation that what matters for the explanation of
behavior, typically anyway, is not how things are, but how they are for
the animal, how the animal takes them to be, i.e. what Taylor calls the
“intentional environment.” For there is a still a question about how
this intentional environment gets into the act and guides behavior.
Taylor’s suggestion—and again, this is an insight with a counterpart
in the work of Merleau-Ponty—is that although animals and people
act out of desire and belief, to be sure, much of what we do is less
governed by our own agency than it is rather called for, or solicited, by
the situation in which we fnd ourselves. Or rather, so as not to oppose
agency and skillful responsiveness to what the situation calls for, we
should say, with Taylor, very often our agency consists just in that kind
of readiness to act in light of the situation in which we fnd ourselves.
And again, this goes for rats as well as people. Crucially, a situation is
never just a locus of stimuli; it is, for the animal, a never fully explicit
but manifest feld of signifcance and opportunity.
On the dust-jacket to the frst edition of The Explanation of Behaviour,
Anscombe notes that the frst part of the book, the philosophical
xvi FOREWORD TO THE ROUTLEDGE CLASSICS EDITION

part, “displays the most remarkable grasp of the contemporary phil-

osophical situation and its historical roots … There is also a satisfac-
tory absence of the tones and attitudes of any particular philosophical
school.” To judge from this, it was perhaps the fashion at the time
for philosophical exchange in print to proceed without much in the
way of explicit mention of the views of colleagues to whom one is
responding or from whom one has taken inspiration. And this may
explain why Taylor’s engagement with other philosophers, like Ryle
and Merleau-Ponty, is left so often unannounced.
Another example of such an unannounced philosophical presence
is Wittgenstein, especially in connection with the critical theme of the
ordinary. The ordinary shows up in The Explanation of Behaviour as a topic
in two distinct ways. First, Taylor wonders whether the possibility of
successful non-teleological explanation of human behavior would
threaten, undermine or delegitimize our ordinary ways of explaining
behavior in terms of purposes and our sensitivity to what situations
demand of us. Second, Taylor frequently compares ordinary styles of
thought and talk with those of the behaviorist, or with what we are
permitted to say in the sanctioned language of “behaviorese.” Here
the issue is complicated and borders even on the political. What is the
basis of our putative knowledge of what is ordinary? Is this something
we learn by introspection? With what authority can we assert, as Taylor
does, that ordinary thought and talk about behavior is teleological?
And who are we anyway? The readers of Taylor’s book who are, as it
were, within the sound of his voice? The larger culture? All people
everywhere? These questions are not posed in The Explanation of Behaviour,
although they come up urgently in the reception of Wittgenstein and
so-called “ordinary language philosophy,” and in Taylor’s later work
in which the notion of the hermeneutic comes to play an increasingly
important role. So readers would be well advised to be on the look-out
for uses of the world “ordinary.”
Graduate students and junior faculty are often required to read with
their eye on a prize such as the extraction of a nugget they can use
in their next article. It is not very often that we allow ourselves the
chance to just read and permit the author to bring us along on their
 FOREWORD TO THE ROUTLEDGE CLASSICS EDITION xvii
journey. But it can be valuable to do so. I recommend that readers fnd
a way to let them do so here with Taylor’s book. It is a youthful book,
to be sure; a frst attempt at laying down a path forward; but, to return
to the theme with which I began, it is also an opportunity for a whole
new discovery of where we think we are now.
Alva Noë, 2020

REFERENCES
Chomsky, Noam. 1959. Review of Verbal Behavior. Language, 35 No. 1, 26–58.
Dennett, Daniel C. 1991. Consciousness Explained. Boston MA: Little Brown.
Dreyfus, Hubert, and Charles Taylor. 2015. Retrieving Realism. Cambridge MA:
Harvard University Press.
Merleau-Ponty, Maurice. 1945/1964. The Phenomenology of Perception. London:
Routledge.
Ryle, Gilbert. 1949. The Concept of Mind. London: Penguin.
Taylor, Charles. 2016. The Language Animal: The Full Shape of the Human Linguistic
Capacity. Cambridge MA: Harvard University Press.
 Preface to the Routledge
Classics Edition

I am very pleased that this book, the basis of which was my doctoral
thesis, is being published again in the Routledge Classics series.
In fact, this book was an intervention in a crucial debate which has
been going on for a very long time, and which may never reach a
fnal agreed conclusion. This is the debate between those who believe
that an adequate account of human life, its evolution and historical
development can be given in the (atomistic and mechanistic) terms of
post-Galilean natural science, and those who hold that this attempt is
fatally fawed. It is a contest between those who hold that really valid
knowledge must take “scientifc” form, whose paradigms are found in
the sciences of nature, and those who believe that no adequate account
of the human can ignore or sideline the goals and achievements we
seek, and the self-understandings we grapple with, neither of which
can fnd a place in a science built on these paradigms.
My thesis was (and remains) that purposes and self-understanding –
in the terms of this work, teleology, and intentionality – cannot be
xx PREFACE TO THE ROUTLEDGE CLASSICS EDITION

ignored in the sciences of human life, on pain of sterility, and ul-

timately failure. The aim is to “reduce” complex culturally defned
thinking, feeling, and actions to accounts in terms of acceptable build-
ing blocks of a materialist science: stimuli and motor responses, when
it comes to “behaviour”; the operations of computer programs as the
supposed basis of thinking; the replication of DNA tracing the conti-
nuity through history of “selfsh genes”, and the like.
But it is in the nature of this debate that it is fought out around
different such reductive projects. At the beginning, exaggerated and
far-reaching claims are made for the new project: all animate action
will be explained by stimulus–response bonds built up by “habit”, all
thinking will be accounted for by self-correcting computer programs
run on the brain as computer, all heredity through selfsh genes. Great
enthusiasm is generated by these exciting perspectives. But then as
time goes on, problems develop, diffculties recur, too simple hypoth-
eses don’t pan out, and doubts creep in.
Finally, a crisis point is reached where the project is, if not aban-
doned, at least indefnitely shelved. But for this move to happen, there
must be an alternative on the horizon. For in fact, these two philo-
sophical outlooks (one might also say: temperaments – the reductive
and the humanistic) are widely shared. The reducers can’t be easily
weaned away from a given program unless and until a replacement
for it appears on the horizon. And in fact what happened sometime in
the 1960s was that the vogue for computers as models of mind was
on the rise, and this created the background for a mass abandonment
of Hull–Skinner behaviourism, and younger scholars leapt on the new
bandwagon.
And The Explanation of Behaviour (TEoB) was published in 1964 just at
this critical transition! Which might seem to put me in the role of the
legendary cock, who couldn’t but believe that his crowing brought
about the sunrise. But of course, I never thought that – well, not for
long. (The actual coup de grace of the old theory was administered by
Noam Chomsky, in his famous review of B. F. Skinner’s book on lan-
guage, Verbal Behaviour, after which behaviourism lay in tatters.)
 PREFACE TO THE ROUTLEDGE CLASSICS EDITION xxi
TEoB was my frst book, but it is hardly a one-off, because the is-
sues around human science, and the conditions of an adequate non-
reductive explanation of human action have continued to pre-occupy
me throughout my life. These issues take different forms in different
disciplinary contexts, but there is an obvious analogy between the
debates in these different sites.
My position on all these was inspired from the very beginning by
the phenomenological tradition, and in particular by the work of
Maurice Merleau-Ponty. His infuence is obvious in TEoB. The model
was laid down by La Structure du Comportement and La Phénoménologie de la
Perception which run together philosophical and experimental fndings
to make their main points.
Later on, I found myself in the Department of Political Science,
where issues about the nature and method of social sciences became
inescapable. And the analogies with the questions dealt with in TEoB
were starkly obvious. It is not that anyone was proposing to borrow
theories from natural science and explain human action in terms of
stimulus–response. But the idea was widespread that the explana-
tion of political action had to invoke tangible, concrete interests. This
was an approach that was often characterized as “materialist”, a term
which played a central role in the Marxist variant, but the notion of
“interest” often played a central role in the work of political scientists
who were virulently anti-Marxist.
But even where the reductivism was far from crude, there was a
general tendency to shy away from attempts to explore in depth the
different self-understandings of political actors, or to examine in
depth the different political cultures which are operative in different
countries today.
What I’m calling here “political cultures” are shared understand-
ings of the shared institutions and practices of a society. I later coined
the term “social imaginaries” to designate these shared understand-
ings. But the problem is that they are usually under-theorized, or to
put it differently, they are not fully articulated. The excellent work of
Irving Goffman, for instance, shows how much of what goes on in
xxii PREFACE TO THE ROUTLEDGE CLASSICS EDITION

everyday exchange is based on tacit understanding, which is nowhere

articulated.
The result is that probing the nature of these implicit understand-
ings requires that one try to articulate them, very often in terms which
the agents concerned wouldn’t immediately recognize. The skills and
practices here are best seen in the work of skilled ethnographers. But
there are no guarantees that one will get things right. Indeed, we
might even say that there is no such thing as getting things defnitively
right, in the sense that no improvement could possibly be made.
We are dealing here with what in the phenomenological tradition
is called “interpretation”, or “hermeneutics”, and this is a practice
which can yield genuine insight, but can never claim to have arrived
at a fnal version, which cannot be improved on. One of the key stages
in my thinking came with the drafting of a paper fve years after the
publication of TEoB, called “Interpretation and the Sciences of Man”
(to which I would today give a gender-neutral title, but 1970 was
still in the benighted era where the necessity for this was not widely
recognized).
The insights here came to me from the phenomenological tradition,
because it was thinkers in this lineage who developed the key ideas:
Heidegger, Gadamer, Ricoeur.
Why do I think that the “hermeneutical turn” is very important?
Because in its absence, social and political theorists are tempted to deal
in dummy universals, processes in different societies grouped together
under a single name, which are in fact very different, because even
if the agents concerned use the same vocabulary, or terms which are
held to be translations of each other, the actual self-understandings
can be widely different from culture to culture, and the generaliza-
tions we make can only take fright at the price of a lot of fudging.
Two areas where I have participated in attempts to show this are:
“modernization” theory and (as a component of this) “secularization”.
In the late twentieth century, the vogue of comparative politics for
a universal “modernization theory” was at its height. The movement
in world history was towards economically developed, geographi-
cally and socially mobile, highly urbanized, “secular” societies (just to
 PREFACE TO THE ROUTLEDGE CLASSICS EDITION xxiii
name some of the main features), and since these societies belonged
historically to different civilizations, this represented a convergence
towards much greater similarity. “Modernization” was the name for
the process which brought (or is bringing) this about, and it happens
in different civilizations at different times, but it brings about the same
result, and as a process is very similar everywhere.
But when one looked more closely at the actual societies, “develop-
ing” and “developed” that make up our world, it became evident that
important differences remained, alongside the convergences men-
tioned above. Indeed, the very notion “development” which underlay
the crucial classifcation just invoked, could also blind us to important
differences.
What was needed was a recognition of “alternative modernities”,
an idea developed by the Center for Transcultural Studies with whom I
have been working for the last 35 years. (And a similar conception of
“Multiple Modernities” was developed by a group around the fertile
work of Shmuel Eisenstadt.)
As for the second concept mentioned above, “secularization”, one
could argue that a similar unfounded homogenization has been im-
posed here by Western social science. This was often thought to be a
single process, happening at a different pace, and occurring in differ-
ent countries at different times, but fundamentally the same.The result
was a very Western-centric picture: we in the West were the pioneers,
and other followed after, sometimes reluctantly, in our train.
Whereas it seemed to me evident that, although there were analo-
gies between the religious (or anti-religious) developments in differ-
ent civilizations, the terms in which they were carried out, and the
dynamics involved were rather different. The way to proceed would
be to attempt a study of secularization in one civilization (and maybe
even that would be too broad, because there were also important dif-
ferences within any great civilization), and then contrast this picture
with what went on elsewhere, and build a more general picture out of
a combination of these more limited studies.
This was the idea which launched me on the drafting of A Secular Age,
which deliberately concentrated on what happened in Western society,
xxiv PREFACE TO THE ROUTLEDGE CLASSICS EDITION

more precisely in (some of) the societies which emerged from the
Latin Christendom of the Middle Ages.This was an exercise in herme-
neutics, and it has all the vulnerability of these. In keeping with what
I said above about hermeneutics, it makes no claim to be defnitive.
On the contrary, I know that a lot more needs to be said, and that the
picture will be very altered through these new contributions.
My most recent book, The Language Animal explores another facet of
the same brace of issues. Reductive theories of human life and devel-
opment have always seen language as primarily important through
its function of encoding, recording, and communicating information.
The early modern pioneers of language theory which preceded and
overlapped and contributed to the Enlightenment – Hobbes, Locke,
and Condillac – were very focused on the way in which language
contributed to – indeed made possible – the development of modern
science. And this focus remains in the heirs of this tradition, which
can be found in the post-Fregean strains of contemporary analytical
philosophy.
I in no way want to contest the importance this function of lan-
guage, and the way in which it has been developed and enhanced
through logic and philosophy. But language plays other crucial roles
in human life, which have been explored in an alternative tradition of
philosophy, taking its origin in German Romanticism, and in which
important fgures are Hamann, Herder, and Humboldt (and in the
twentieth century, Heidegger). Language is crucial to intimacy for hu-
mans; we foster intimacy, and also mark distance in the ways we talk
to each other. But also on a broader social level, the language we use
creates or underlines hierarchies. Think of what for many languages
rides on the distinction between intimate and formal address (the tu/
vous, or du/Sie distinction).
Not to speak of the roles of literature and art; nor of the importance
of narrative for our understanding of ourselves.
There are vast felds to explore here, which need to be integrated
into our theory of language, and therefore of our understanding of
ourselves as “animals possessing language”, the original defnition of
the human offered by Aristotle.
 PREFACE TO THE ROUTLEDGE CLASSICS EDITION xxv
The above description of my writings makes me sound like a mon-
omaniac. And maybe there’s some truth to this. But in any case, it
should be clear that TEoB started me on a path that has led me to many
destinations of which I had little idea at the time, but in retrospect
form a single itinerary through many different sites. I am very pleased
that the book will once more be available, and I thank Routledge for
this.
My work has fallen mainly in what we call the “human sciences”,
which we could roughly characterize as the disciplines in which cul-
ture, and particularly cultural differences, play an obvious role: history
politics, social theory, religion, linguistics, and the like. But a central
area in which the dispute between the two contrasting outlooks has
been argued out lies in the biological or life sciences. This is certainly
the area where the reductive stance can appear most plausible.
While I don’t feel competent to intervene here, I have taken much
inspiration from the work of Evan Thompson, Denis Noble, and Lenny
Moss.
But once the reductive presumption is lifted, the road to a herme-
neutical account of human cultures is open, and this is where new and
important discoveries remain to be made.
Charles Taylor, 2020
 Acknowledgements

This book was written while I was a Fellow of All Souls College. I am
very grateful for the opportunity that this provided me to complete
the work.
I would also like to thank Isaiah Berlin, Stuart Hampshire, Roy Hol-
land and Bernard Williams who helped by advice, and even more by
criticism, to make this book somewhat less inadequate to the subject
than it would have been.
PART 1
EXPLANATION BY PURPOSE
 1
PURPOSE AND TELEOLOGY

It is often said that human behaviour, or for that matter the behav-
iour of animals or even living organisms in general, is in some way
fundamentally different from the processes in nature which are stud-
ied by the natural sciences. This opposition is variously expressed. It
is sometimes said that the behaviour of human beings and animals
shows a purposiveness which is not found elsewhere in nature, or that
it has an intrinsic ‘meaning’ which natural processes do not. Or it is
said that the behaviour of animate organisms exhibits an order which
cannot be accounted for by the ‘blind accident’ of processes in nature.
Or again, to draw the circle somewhat narrower, it is said of human
beings and some animals that they are conscious of and direct their
behaviour in a way which fnds no analogue in inanimate nature, or
that, specifcally in an account of human affairs, concepts like ‘sig-
nifcance’ and ‘value’ play a uniquely important part which is denied
them in natural science.
4 EXPLANATION BY PURPOSE

Against this view stands the opinion of many others, in particular

of many students of the sciences of human behaviour, that there is no
difference in principle between the behaviour of animate organisms
and any other processes in nature, that the former can be accounted
for in the same way as the latter, by laws relating physical events, and
that the introduction of such notions as ‘purpose’ and ‘mind’ can
only serve to obscure and confuse. This is in particular the point of
view of the widely spread school of thought in psychology known as
‘behaviourist’.
Now the issue between these two views is one of fundamental and
perennial importance for what is often called philosophical anthropol-
ogy, the study of the basic categories in which man and his behaviour
is to be described and explained. That this question is central to any
science of human behaviour—if such a science is possible—needs no
showing. But this does not by any means exhaust its importance. For it
is also central to ethics.Thus there is a type of ethical refexion, exem-
plifed for instance in the work of Aristotle, which attempts to discover
what men should do and how they should behave by a study of human
nature and its fundamental goals.This is the attempt to elaborate what
is often called a ‘humanism’. The premiss underlying this refexion,
which is by no means confned to philosophers, is that there is a form
of life which is higher or more properly human than others, and that
the dim intuition of the ordinary man to this effect can be vindicated
in its substance or else corrected in its content by a deeper under-
standing of human nature. But this premiss collapses once it is shown,
if it ever is, that human behaviour cannot be accounted for in terms
of goals or purposes but must be explained on mechanistic principles;
for then the concept of fundamental human goals or of a way of life
more consonant with the purposes of human nature—or even the
existentialist notion that our basic goals are chosen by ourselves—will
be shown to have no application.
A similar premiss, that a purpose or set of purposes which are in-
trinsically human can be identifed, underlies all philosophical and
other refexion concerning the ‘meaning’ of human existence, and
this, too, would collapse if the mechanistic thesis were shown to hold.
 PURPOSE AND TELEOLOGY 5
These brief remarks, which still do not exhaust the ramifcations
of this question, are enough to show why it has been of perennial
interest to philosophers and laymen. And yet in spite of, or perhaps
because of this, it still awaits resolution. We might try to explain this
simply by pleading lack of evidence. In fact the sciences of man are
in their infancy. But this cannot be the whole explanation. In fact, it
seems simply to put us before the same question in another form. For
we might just as truly say that the sciences of man, and particularly
psychology, are in their infancy because this question remains unre-
solved. It is not enough, therefore, simply to invoke research. In fact
the trouble is deeper: we have frst to know where to look. And when
we ask ourselves this question, we fnd ourselves well and truly at sea.
In fact there has never been agreement among philosophers or
other students as to what is at stake here, that is, on the meaning of the
claim that human behaviour is purposive, or, what is the same thing,
on what the relevant evidence is which would decide it. As a matter
of fact, it is not even generally agreed that it is a matter of fnding evi-
dence in the frst place, for some thinkers hold that the issue is not in
any sense an empirical one, but rather that it can be decided simply by
logical argument.
This confusion might tempt us to say that the question is insoluble,
or even that it is a pseudo-question. But this radical ‘solution’ would
itself have to be established by some argument about the nature of the
putative issue involved. Before fnally turning our backs on the matter
in this way, therefore, it is worth trying once more to defne what is
at stake.This is what will be attempted in the frst part of this book.

1. TELEOLOGICAL EXPLANATION
What, then, does it mean to say that human, or animal, behaviour
is purposive? Central to this claim would seem to be the view that
the order or pattern which is visible in animate behaviour is radi-
cally different from that visible elsewhere in nature in that it is in
some sense self-imposed; the order is itself in some way a factor in its
own production. This seems to be the force of the rejection of ‘blind
6 EXPLANATION BY PURPOSE

accident’: the prevalence of order cannot be accounted for on princi-

ples which are only contingently or ‘accidentally’ connected with it,
by laws whose operation only contingently results in it, but must be
accounted for in terms of the order itself.
The point, then, could perhaps be put in this way: the events pro-
ductive of order in animate beings are to be explained not in terms
of other unconnected antecedent conditions, but in terms of the very
order which they produce. These events are held to occur because of
what results from them, or, to put it in a more traditional way, they
occur ‘for the sake of’ the state of affairs which follows. And this of
course is part of what is meant by the term ‘purpose’ when it is in-
voked in explanation. For to explain by purpose is to explain by the
goal or result aimed at,‘for the sake of’ which the event is said to occur.
Explanation which invokes the goal for the sake of which the expli-
candum occurs is generally called teleological explanation, and thus at
least part of what we mean by saying that human or animal behaviour
is purposive is that it is to be accounted for by a teleological form of
explanation.1
But does this take us any further ahead? What is meant by teleo-
logical explanation and how can we establish whether it holds or not
of a given range of phenomena? The remainder of this chapter will
be devoted to an attempt to answer this question, and to cut through
the skein of confusions which usually makes it diffcult to bring this
question into focus.

2. AN EMPIRICAL QUESTION?
Now a frst diffculty arises straight off with the objection that this
question, whether or not a teleological explanation holds, is not an
empirical one at all. In fact many theorists, and particularly students of
what can roughly be called the behavioural sciences, would hold that
the claim that animate behaviour must be explained teleologically or
in terms of purpose is a meaningless one, empirically empty or ‘meta-
physical’, that the whole question is a ‘pseudo-question’.This is espe-
cially true of many theorists in the feld of experimental psychology,
 PURPOSE AND TELEOLOGY 7
those of the behaviourist school, on which the discussion later in
this book will mainly centre. These thinkers, extremely hostile to the
claims of teleological explanation, make short work of it by purport-
ing to expose, in summary fashion, its non-empirical character.
If this objection is a valid one, then our whole enquiry is stopped
before it starts. But, as a matter of fact, it is not. In fact, it reposes on an
interpretation of the notions of purpose and teleological explanation
which is arbitrary and by no means imposed on us.
Thus the claim that we must explain the behaviour of a given sys-
tem in terms of purpose is often taken to mean that we must explain
it by laws of the form x = f(P), where ‘x’ is the behaviour and ‘P’ is the
Purpose considered as a separate entity which is the cause or anteced-
ent of x. Of course the view that an explanation in terms of purpose
involves the postulating of a special entity is by no means confned
to those who are hostile to the idea. Many who were on the ‘vitalist’
side of the controversy in biology made use of a hypothetical entity
of this kind. (Cf. Driesch’s ‘entelechy’.) But there is no doubt that the
end result of this is to create a handy Aunt Sally for the mechanists. For
a theory of this kind can neither be confrmed nor add in any way to
our power of predicting and controlling the phenomena.
This can be readily seen. In fact, the only empirical evidence for the
operation of the purpose is the behaviour which its operation is used
to explain. There is thus no conceivable evidence which could falsify
a hypothesis of this kind because whenever the behaviour is emitted,
the purpose responsible is ex hypothesi assumed to have been operating.
And at the same time we would never be able to predict behaviour
with the aid of such a hypothesis. For if x having a value of x1 is due to
P having the value P1, and if the only evidence for P1 is the occurrence
of xl, then we have no way of knowing beforehand what the value of
x will be.
Now, of course, we might fnd some antecedent conditions for P,
such that we could determine the value of P ex ante by means of a
function such as P = f(a). But then we would be turning P into what is
often called an ‘intervening variable’,2 that is, a term useful in calcu-
lation which is nevertheless without empirical content, which is not
8 EXPLANATION BY PURPOSE

itself an empirical descriptive term. For in this case the entire empirical
content of the two functions P = f(a) and x = f(P) could be expressed
in one more complex function linking a and x directly, x = F(a). What
is meant by saying that ‘P’ is not an empirical descriptive term is that
no single proposition about P is open to empirical confrmation or
infrmation. Thus, in the case above, neither of these functions can be
verifed singly. We have seen above that this is true of x = f(P), but it
is equally so of P = f(a). The proposition formed by the conjunction
of both is open to empirical confrmation, but then the evidence for
this is the same as the evidence for x = F(a) which makes no mention
of P. That is, no empirical sense can be given to the supposition that
x = F(a) be true and the conjunction of the two functions false. Thus
the question whether or not the functions containing ‘P’ are to be
accepted is not an empirical but purely a stipulative question, to be
determined by the convenience in the calculation. ‘P’ is therefore not
an empirical descriptive term.
Thus those who hold that ‘purpose’ is essential to the explanation
of the behaviour of animate organisms are left with the unattractive
choice either of making an unverifable claim of no explanatory utility
in science or of winning their point at the expense of making the laws
true by stipulation. This view of the matter is very common among
behaviour psychologists who are unsympathetic to this claim. Their
view seems to be that their opponents adopt the frst position, that of
positing an unobservable entity, propositions about which cannot be
verifed. Thus Hebb in the frst chapter of his The Organization of Behaviour
speaks interchangeably of ‘animism’ (the view that animate behaviour
must be explained in terms of ‘purpose’) and ‘interactionism’ (the
view that behaviour is the result of the interaction of observable phys-
ical and unobservable ‘inner’ or mental processes) and of course ‘mys-
ticism’ (which doesn’t seem to have a very clear sense in Hebb’s usage
but which means something counter-empirical, unscientifc and gen-
erally nasty). Similarly, Spence3 speaks of animistic theories as those in
which the relation of the (unobservable) constructs to the empirical
(observable) variables is left entirely unspecifed (and hence they are
unverifable as in the frst alternative above).
 PURPOSE AND TELEOLOGY 9
The upshot of this view, then, is that the claim that animate organ-
isms have a special status is undecidable, or rather that even to make it
is to say something which cannot be verifed. If the question is not to
be closed here we shall have to examine explanation by purpose more
closely in order to determine whether it must involve the postulating
of an unobservable entity which is the cause or the antecedent condi-
tion of behaviour.
Now, as we have said, explanation by purpose involves the use of a
teleological form of explanation, of explanation in terms of the result
for the sake of which the events concerned occur. Now when we say
that an event occurs for the sake of an end, we are saying that it occurs
because it is the type of event which brings about this end.This means
that the condition of the event’s occurring is that a state of affairs ob-
tain such that it will bring about the end in question, or such that this
event is required to bring about that end.4 To offer a teleological ex-
planation of some event or class of events, e.g., the behaviour of some
being, is, then, to account for it by laws in terms of which an event’s
occurring is held to be dependent on that event’s being required for
some end.
To say that the behaviour of a given system should be explained
in terms of purpose, then, is, in part, to make an assertion about the
form of the laws, or the type of laws which hold of the system. But
qua teleological these laws will not be of the kind which makes behav-
iour a function of the state of some unobservable entity; rather the
behaviour is a function of the state of the system and (in the case of
animate organisms) its environment; but the relevant feature of system
and environment on which behaviour depends will be what the con-
dition of both makes necessary if the end concerned is to be realized.
Thus for instance, we can say that the conditions for a given action,
say a predator stalking his prey, are (1) that the animal be hungry, and
(2) that this be the ‘required’ action, i.e., the action in his repertoire
which will achieve the result—catching his next meal. The condition
of an event B occurring, is, then, not a certain state of P, but that the
state of the system S and the environment E be such that B is required
for the end G, by which the system’s purpose is defned.
10 EXPLANATION BY PURPOSE

Now the fact that the state of a system and its environment is such as
to require a given event if a certain result is to accrue can be perfectly
observable, and the fact that this antecedent condition holds can be
established independently of the evidence provided by the occurrence
of the event itself.This type of law, therefore, does not suffer from the
disabilities of functions of the type x = f(P). On the contrary, whether
laws of this kind hold can be verifed or falsifed, and, if true, they can
be used to predict and control the phenomena like any others. To say
that a system can only be explained in terms of purpose, then—at least
insofar as this is an assertion about the form of the laws—does not in-
volve making an unverifable claim any more than it involves postulat-
ing an unobservable entity. The element of ‘purposiveness’ in a given
system, the inherent tendency towards a certain end, which is con-
veyed by saying that the events happen ‘for the sake of’ the end, can-
not be identifed as a special entity which directs the behaviour from
within, but consists rather in the fact that in beings with a purpose an
event’s being required for a given end is a suffcient condition of its
occurrence. It is not a separable feature, but a property of the whole
system, that by which it tends ‘naturally’ towards a certain result or
end. It is this notion of a ‘natural tendency’ towards a certain result or
end—which we shall discuss at greater length in the next section—
which lies behind the notion that, in systems whose behaviour must
be accounted for by laws of this kind, the order which results cannot
be attributed to ‘blind accident’, that is, to principles which are only
contingently connected with the bringing about and maintenance of
this order; for the principle underlying the laws by which the behav-
iour is explained is itself a tendency to produce this order.

3. ASSUMPTIONS OF ATOMISM
What considerations, then, led to the belief held by mechanists and
vitalists alike that the claim to some special status involved the pos-
tulating of some special entity? The background to this belief is very
complex and exploring it would involve unravelling a skein of con-
nected questions which surround this subject, some of which we
 PURPOSE AND TELEOLOGY 11
hope to return to in the fourth chapter. But at this stage one of the
causes can perhaps be laid bare.
We can readily see that, in any explanatory functional law, the
antecedent and the consequent must be separately identifable.Thus it
cannot be a logical condition for the occurrence of the antecedent that
the consequent occur. Something like this was what was the matter
with our x = f(P). True, it was not a condition (logically) of P having
a certain value that x have the corresponding value demanded by the
function, but since the latter was the only evidence for the former, it
came to much the same thing. Similarly, the antecedent’s occurring
cannot be a logical condition of the consequent’s being held to occur.
Now teleological laws meet this requirement. For the antecedent may
occur independently of the consequent, and vice versa.
But there is a stronger requirement which teleological laws cannot
meet.This not only demands that the two terms which are linked in a
law be identifable separately from each other, that is, that it not be a
condition for identifcation of either term that it be linked to the other
term in the law,5 but also that each term be identifed separately from
any law in which it may fgure, i.e., that it not be a condition for the
identifcation of any term that it be linked to any other. Now this more
stringent requirement arises from the atomism which is part of the
tradition of empiricism, and is ultimately founded on epistemologi-
cal grounds. The notion is that the ultimate evidence for any laws we
frame about the world is in the form of discrete units of information,6
each of which could be as it is even if all others were different, i.e.,
each of which is separably identifable from its connexions with any
of the others. Our knowledge of the world is built up from the empir-
ical connexions which are found to hold (contingently) in experience
between these units. Thus the evidence for any law can ultimately be
given, although perhaps with great tedium, in terms of connexions
between such discrete units.Thus if a given chemical C, with defning
properties x, y and z, is held to produce a certain result R in some con-
ditions, although we may usually for convenience sake speak of this
law as C — R, the ultimate evidence for it is the concomitance of x, y,
z and R, each of which is identifable separably from the others.
Exploring the Variety of Random
Documents with Different Content
mancipia, proximo vero paterni and slaves, but the land to the
generis terram relinquat. next of the paternal kin.
(VIII) Usque ad quintam (VIII) As far as the fifth
generationem paterna generation the paternal kin
generatio succedat. Post succeed. But after the fifth, a
quintam autem filia ex toto, daughter, whether on the father’s
sive de patris sive de matris or on the mother’s side, may
parte, in hereditatem succedat; succeed to the whole inheritance;
et tunc demum hereditas ad and then finally let the inheritance
fusum a lancea transeat. pass over from the spear to the
spindle.

As in the other laws so under these rules the

alod clearly embraced both the land and the The alod included
both land and
‘pecunia’ and ‘mancipia’ upon it. Its object, like that cattle.
of the similar clauses in the other laws and also like
that of the Edict of Chilperic, seems to have been to protect the land
in ordinary cases from passing over ‘from the spear to the spindle,’
while at the same time sanctioning inheritance by females even in
the land of the alod when otherwise there would be danger of its
passing away from the kindred altogether.
In certain cases the land of the alod was made to go to male heirs
while the ‘pecunia’ and ‘mancipia’ upon it went to females.
Whether the word ‘pecunia’ in such cases should be translated by
‘cattle’[170] or the wider word ‘chattels,’ it must have included the
cattle, and at first sight it is not easy to see how the rule would work
which gave the cattle of the alod to a female and the land to a
distant male heir. The cattle must in the nature of things have
remained or be put upon land, and the awkward question arises
upon whose land they remained or were put. And so we are brought
once more to the practical question of the position of women in
relation to the land. That in certain cases in default of male heirs
they could inherit land is one thing; but this question of the cattle
and slaves involves quite another.
 When a sister received her portion or gwaddol
under Cymric custom, and when she received so Male next of kin
takes the land
many cows for her maintenance from the chief of and chieftainship,
kindred, she must have had rights of grazing for but females may
her cattle in the family herd of her gwely. Till she have cattle upon
the land.
married, her cattle would graze with the cattle of
her paternal gwely; and when she married, with the cattle of her
husband’s gwely. And so under the rules of this clause ‘De alodibus’
it does not follow that the distant male heir succeeding to the land
of the alod was to evict her and her cattle from it. With the land he
had to take also the responsibilities involved in the family holding.
Clause V. states that to whomsoever the inheritance of the land shall
come, to him ought to pertain the coat of mail, i.e. the birnie, and
with it the duty of the chief of the kindred to avenge his kin and to
see to the payment of wergeld if any one of the kin should be slain.
Read from this point of view this clause ‘De alodibus’ becomes good
evidence that, whatever changes may have been made as to female
inheritance, the land of the alod had not yet lost all its tribal traits. It
had not yet become the ‘res propria’ of an individual possessor
under Roman law.

V. THE SO-CALLED LEX CHAMAVORUM.

This document, according to most recent authorities, relates to a
district between the Frisians and Saxons to the North and East, with
the river Meuse to the South.[171]
Its real title seems to be Notitia vel
commemoratio de illa euva quæ se ad Amorem The Chamavi
under Frankish
habet, and it seems to be not so much a code as a law.
memorandum of the wergelds and fines of a
Frankish people settled in the district alluded to. Probably in date it
may belong to the time of Charlemagne, but before his changes in
the currency.
 It is of some interest to this inquiry because of its peculiar
position, as relating to a tribe or people under Frankish rule, and yet
with customs of its own which have survived Frankish conquest.
The Notitia starts with the declaration that in ecclesiastical
matters, as regards the bannus dominicus, the same laws prevail ‘as
other Franks have.’
And then it at once describes the wergeld, as
follows:— Wergeld of the
Homo Francus
three times that
The wergelds of this law are as under. Whoever of the ingenuus.
kills—
Homo Francus 600 solidi et pro fredo 200 sol.
Ingenuus 200 ” ” 66⅔ ”
Lidus 100 ” ” 33½ ”
Servus 50 ” ” 16⅔ ”

Then follows a clause (VII.) which states that if any ‘Comes’ be

slain in his own ‘comitatus’ the wergeld is to be three times that
according to his birth.
The Homo Francus thus has a triple wergeld, like the Comes. But
the Comes may possibly be not ingenuus. He may be a lidus with
official position, and so presumably, according to Clause VII., with a
threefold wergeld of only 300 solidi.
In the next clause the Royal ‘Missus’ is put in the same position
while on the King’s business. His wergeld is also to be trebled.
What, then, is the Homo Francus with a wergeld three times that
of the ordinary ingenuus of the district of Amor?
The wergeld of the latter is the full normal wergeld of 200 solidi.
The Homo Francus in this district was therefore very much above the
ordinary freemen of other laws. He was evidently a Frankish
landowner on a large scale, towering in social position above the
ordinary freemen of the district.
 The casa and curtis of the Homo Francus alone were protected by
special clauses (XIX. and XX.), and of him alone are any hints given
as to kindred or inheritance. Clause XLII., in the following few
words, enlightens us as to his social position:—

If any Francus homo shall have sons, his inheritance in woods and
in land shall pass to them, and what there is in slaves and cattle.
Concerning the maternal inheritance, let it go in like manner to
the daughter.

We must probably consider the privileged position of the Homo

Francus as presumably the result of Frankish conquest. The great
landowner may have been the holder of a benefice, or a tenant in
capite placed upon the royal domain with ministerial and judicial
duties, and the triple wergeld may fairly be assigned to his official
position.
But to return to the wergelds.
The payment pro fredo seems to have been equal to an additional
one third of the wergeld.
From clauses XX. and XXXII. it appears that the
value of an eye or hand or foot was one quarter of Payment for the
eye etc. one
the wergeld, instead of half as in the Salic and quarter the
Ripuarian Laws. wergeld.

Theft was to be paid for ninefold with four solidi pro fredo.
The further clauses regarding theft in this border district of forests
and cattle and mixed population are not quite easily understood, nor
need we dwell upon them.
In c. XXX. the penalty for letting a thief go without bringing him
before the Comes or centenarius was 60 solidi, as in the Ripuarian
Laws.

VI. CONCLUDING REMARKS.

 Before passing from the laws, the compilation of which seems to
date from the conquests of Charlemagne, it may be well to note
that, regarded from the point of view of the wergelds, the tribes
whose customs have been examined in the last two sections seem to
have belonged to the Frankish group with wergelds of 200 gold
solidi, while on the other hand the Frisians and Saxons seem to have
belonged to the other group with wergelds of 160 gold solidi.
This grouping of the tribes may not be exactly what might have
been expected.
Geographically the Frankish group is sufficiently
compact. The other is widely extended and The two groups of
tribes with
scattered. Frisians and Saxons remain in their wergelds of 200
ancient homes. The Alamannic, Bavarian, and and 160 solidi.
Burgundian tribes have wandered far away from
theirs. But in their northern home they may have been once
sufficiently contiguous to have shared many common customs and
among them a common wergeld of 160 solidi.[172] Settled in their
new quarters, the Rhine and its tributaries seem to have been the
great highways of commercial intercourse and the connecting links
between them. Immigrants from them all met as strangers (advenæ)
in the Ripuarian district, and, as we have seen, we owe our
knowledge of some of their wergelds very much to the recognition of
them in the Ripuarian law.
 CHAPTER VIII.
THE TRIBAL CUSTOMS OF THE OLDEST
SCANDINAVIAN LAWS.

I. THE MONETARY SYSTEM OF SCANDINAVIA.

The facts needful for the understanding of the monetary system of
the Scandinavian tribes need not detain the reader very long.
The weight system applied to gold and silver was that evidently
derived from the Eastern Empire.
It consisted of the mark, the ore, and the ortug.
The mark was divided into eight ores or ounces, Marks, ores, and
ortugs. The ortug
and the ore or ounce into three ortugs, which were the Greek stater
in fact staters or double solidi. The ounce being the or ox-unit.
Roman ounce of 576 wheat-grains, the ortug
contained 192 wheat-grains, and was the exact counterpart in
wheat-grains of the Greek stater, i.e. Professor Ridgeway’s ox-unit.
Reckoned in wheat-grains, two Scandinavian marks of 8 ounces
were, as we have seen, exactly equal to what the early metrologists
called the (light) Mina Attica, which consisted of 16 Roman ounces
or 9216 Roman wheat-grains. Four gold marks thus made a heavy
gold mina, traditionally representing a normal wergeld of 100 head
of cattle.
But this heavy gold Mina of four marks had been seemingly
twisted from its original Greek character to bring it into consistency
with Roman methods of reckoning. It was divided no longer into 100
staters, but now into 96 ortugs, so as to make the ortug double of
the solidus and one third of the Roman ounce, thus throwing it out
of gear, so to speak, with the normal tribal wergelds of 100 head of
cattle. It was thus made to contain only 96 ox-units, although in
actual weight its 32 Roman ounces really did contain, so long as the
standard of the Roman ounce was adhered to, 100 Attic staters or
ox-units.
That the light mina of two marks or 9216 wheat-grains had found
its way by the Eastern trade routes into Scandinavia appears from its
survival in the monetary system of countries on both sides of the
Baltic to quite modern times.
In Northern Europe the pound of twelve ounces
was not, as elsewhere, the usual larger unit. The The pound of two
marks.
pound of two marks or sixteen ounces had taken
its place. And except in Norway and Denmark, which sooner or later
adopted the monetary and weight system of Charlemagne, the
ounce remained the Roman ounce of 576 wheat-grains. At the same
time, as in the case of the Merovingian system, in spite of the
Imperial influence of the gold solidus, there were evident marks of a
tendency towards the ancient Eastern standard of the stater rather
than the heavier standard of the double solidus. The ortug of 192
wheat-grains seems to have often sunk in actual weight below even
the Attic weight to that of the ancient Eastern stater of 8·18
grammes.
Thus when the Russian weight system was recorded in the time of
Peter the Great the unit both for precious metals and goods was
found to be the Zolotnic or gold piece. Thus—

Dolja = ·0444 grammes = wheat-grain.

Zolotnic = 4·265 ” = 96 w.g.
Funt = 409·511 ” = 96 zolotnic, or 9216 w.g.

Here, then, in wheat grains the Funt is the light Mina Attica over
again, Romanised in its divisions. The Zolotnic is the solidus or half-
stater. But in actual weight the pound is exactly half of the ancient
Eastern gold mina of 818 grammes.
 The Pfund of Silesia (Breslau), according to Martini, was 405
grammes, and that of Poland (Cracow) the same. Only Sweden and
Riga seem to have adopted or preserved higher standards, the
double mark of Sweden being 425 and that of Riga 419 grammes;
but even these fell far short of the standard weight of 16 Roman
ounces, viz. 436 grammes. But throughout, low as the standard of
the Baltic Funts or double marks may have been, they were divided
according to the Roman commercial weight system into ores or
ounces and loths or half-ounces, and gwentschen or drachmas of
one eighth of an ounce, just as if they were of full Imperial weight.
The marks and the ores remained, but the old division of ores into
ortugs or staters had long ago disappeared.
The division into marks, ores, and ortugs was, however, in full
force at the time of the Norse laws, both for gold and silver. And the
evidence of actual weights seems to show, not only that for the
purposes of the Eastern trade routes, reckoning in marks, ores, and
ortugs was in common use, but also that the standard, like that of
the Merovingian coinage, was the ancient Eastern standard.
Thus the following weights, believed to belong to the Viking
period, from the island of Gotland, are now in the Royal Museum at
Stockholm (Nos. 4752 and 5984).

The ortug in
weight = Eastern
stater or two
Merovingian
solidi.

819 grammes = 100 staters or ortugs of 8·19

57·25 ” = 7 ” ” 8·1
 32·65 ” = 4 ” ” 8·16

32·4 ” = 4 ” ” 8·1

24·35 ” = 3 ” ” 8·12

The unit of these weights is exactly the Eastern stater of 8·18 or

two Merovingian solidi.[173]
Whether this standard had been arrived at independently of the
Merovingian standard, or adopted from it, we must not stop to
inquire. It is enough that the ortug at the date of the laws through
Roman influence had come to be reckoned as one third of the
ounce.
Whatever may have been the early Byzantine influences and that
of Eastern trade routes, long before the date of the Norse laws,
Scandinavia had come under Frankish influences also.
Already during Merovingian times, chiefly
through the Frisian mint at Duurstede, Merovingian The mark of 8
Roman ounces
currency had become well known on the Baltic, and and
we have seen that the first Scandinavian coins Charlemagne’s
were copies from those of the Duurstede type. mark of the nova
moneta.
Hence it came to pass that in the most ancient of
the Norse laws the old Scandinavian reckoning in gold and silver
marks, ores, and ortugs had become connected with the Frankish
currency. During the period of Merovingian influence the Merovingian
ounce and the Norse ore were both, reckoned in wheat-grains, the
ounce of the Roman pound, whatever may have been their actual
weight. The mark of eight ounces contained 4608 wheat-grains of
gold or silver. But at last, as the result of Charlemagne’s conquests in
the North, his nova moneta with its higher standard was brought
into contact with Scandinavia. His mark of eight of his ounces or
5120 wheat-grains ultimately superseded in Norway and Denmark
the old mark of eight Roman ounces. Hence, as all the Scandinavian
laws as we have them, are of later date than Charlemagne’s
conquests, the question must arise, which of the two marks is the
one in which the wergelds and other payments are described.
In the oldest Norse laws the wergelds are stated mostly in silver
marks, ores, and ortugs. The ratio between gold and silver was 1:8,
so that an ore of gold equalled a mark of silver, and thus the
translation of silver values into gold is easy. The laws themselves, as
we shall find, make this perfectly clear. A wergeld stated as of so
many gold marks is divided in the details of payment into silver
marks, ores, and ortugs at the ratio of 1:8.

II. THE WERGELDS OF THE GULATHING AND

FROSTATHING LAWS.
In approaching the consideration of the
Scandinavian custom as to wergelds and the The Gulathing
law.
structure of tribal society as disclosed in the
ancient laws, I do it with great diffidence, especially as, for the
translation of Old Norse, I am dependent on others.
On the whole it seemed best to concentrate attention upon the
Gulathingslög as the oldest of the Norse laws. The Danish and
Swedish laws and the Grágás of Iceland no doubt under competent
hands would yield valuable additional evidence, but the oldest of the
Norse laws may probably be fairly taken as the most representative
of early Northern custom, and at the same time most nearly
connected with the object of this inquiry.
Geographically the Gulathing law was in force in the southern
portion of Norway. It seems to have embraced, in about the year
930, three, and afterwards six, fylkis or districts each with its own
thing and local customs.[174] In this respect it resembled the Frisian
and Saxon laws, both of which recognised, as we have seen, the
separate customs of tribal divisions contained in the larger district
over which the laws had force.
 The Gulathing law must therefore be regarded as in some sense a
compilation or collection of customs rather than one uniform law. For
instance, there are three or four separate descriptions of the wergeld
and the modes of its payment and receipt. One of these is avowedly
of later date. The older ones may probably describe local variations
of general custom, belonging to one or another of the divisions, and
even these bear marks of later modification and additions.
As usual, the introduction of Christianity was the occasion and
perhaps the cause of the compilation, and therefore from the time of
the formation of Dioceses by King Olaf (a.d. 1066-93) ecclesiastical
influence must be expected. But on the whole this Gulathing law
presents in some points a far more interesting and instructive picture
of social conditions resulting from tribal custom than the laws of
other tribes already examined of much earlier date.
The next important of the ancient laws of
Norway is the Frostathingslög belonging to the The Frostathing
law.
more northerly district of Drontheim. Without
pretending to have made it the subject of special study, I have here
and there found it useful in elucidation of the Gulathing law, and as
showing that tribal custom, though with local variations, was in force
over a wider district than that under the Gulathing law.
The question of the structure of tribal society and the division of
classes in Norway may be most conveniently approached from the
point of view of the rett or ‘personal right,’ somewhat analogous to
the Irish ‘honour-price’ and the Welsh ‘saraad.’
Both in the Gulathing law and in the Frostathing
law this personal ‘rett’ lies at the root of the Grades of
personal ‘rett.’
graduated payments for insults, wounding, and
homicide. And the statements of it are practically identical in the two
laws. They are as follows:—

Gulathing (200) Frostathing (X. 35)

Leysing 6 ores Leysing before 4 ores
freedom’s ale
 Leysing after 6 ores
freedom’s ale
Leysing’s son 8 ” Leysing’s son 8 ores or 1 mark
Bónde 12 ” Reks-thane[175] 12 ores
Árborinn 16 ores or 2 marks
man[176]
Hauldman[177] 24 ” Hauldman 24 ores or 3 ”
Lendman and Stallare 48 ”

The chief difference is that the Frostathing law divides the leysings
into two classes, a significant point on which important
considerations turn.
The things for which full rett was paid may be described as insults.
If a man were knocked down, even if he fell on his knees, or if his
moustache were ‘seized with hostile hand’ (195), or if a man were
called ‘a mare or bitch,’ these were insults for which full rett was to
be paid (196).
The payments for inflicting serious wounds (sár) were regulated in
the same gradations according to rank as the rett, but were
threefold in amount. These payments were made in ‘baugs’ or rings,
each of twelve ores of silver.

Gulathing Law (185) Frostathing Law (IV. 53)

Leysing 1 ring
Leysing’s son 2 rings Leysing 2 rings
Bónde 3 ” Reks-thane 3 ”
Ár-borinn man 4 ”
Odal-born man[178] 6 ” Hauld 6 ”
Lend-man and Stallare 12 ” Lend-man 12 ”
Jarl 24 ” Jarl 24 ”
King 48 ” King 48 ”
 These were the penalties paid by the person inflicting the wound
—i.e. three times his own rett—and besides this he had to pay sár-
bót according to the extent and character of the wound, as in other
laws. He also had to pay the healing fee (185) of the injured person.
Passing from insults and wounds to homicide,
throughout the Gulathing law the hauld, or odal- The hauld or odal-
man the typical
born man, is taken as the typical tribesman. His tribesman.
wergeld is described, and then the wergelds of
other classes are said to vary according to the rett.
But before we consider the wergelds it must be remarked that
here, as elsewhere, there is no wergeld for a murder within the
family.
In clause 164 under the heading of ‘A madman’s manslaying’ is
the following:[179]

Nu hever maðr óðz mannz Now if a man has done the

víg vigit, vigr sunr faður, æða slaying of a madman, if a son
faðer sun, æða bróðer bróðor, slays his father, or a father his
æða systkin eitthvert, æða vigr son, or a brother his brother or
barn móðor sína, æða móðer any of his sisters and brothers, or
barn sitt, þá firi-vigr hann arve a child slays its mother or a
þeim er hann átti at taca. Scal mother her child, then he forfeits
sá þann arf taca er nestr er þá, the inheritance he ought to take.
oc helldr scal konongr hava en The one next to him in kin takes
hann. En hann være í lande, oc that inheritance, and the King
gange til skrifta, oc have sitt shall have it rather than he. But
allt. he shall stay in the land and be
shriven and keep all that is his.

In the Gulathing law the kindred within which

there is no wergeld is thus the actual family, and it No wergeld within
the family.
is in full accord with the instance in Beowulf in
which the old father is represented as having to put up with the
presence of a son by whose arrow another of his sons had been
slain, such a crime being one which under tribal custom could not be
avenged.
Turning now to the amount of the wergeld of the Gulathing law
and the Frostathing law, it must again be remarked that there are in
these laws varying accounts of it.
In the first place there are some avowedly of
later date than others. Thus, in Frostathing VI. 1 The wergeld of
the Frostathing
the description of the wergeld is commenced as law of later date
follows:— awarded in marks
of gold.

Her hefr upp oc segir í frá því Here begins and is told that
er flestum er myrkt oc þyrftu þó which to most is dark and yet
marger at vita, fyrir því at many had need to know, because
vandræði vaxa manna á millum difficult matters increase among
en þeir þverra er bæði höfðu til men and those grow fewer who
vit oc góðan vilja, hvesso scipta both had the wits and the goodwill
scylldi ákveðnum bótum ef þær for it,—how to divide the fixed
ero dœmdar, fyrer því at þat er bóts (bœtr) if they are adjudged,
nú meiri siðr at ánemna bœtr, for it is now more the custom to
hvesso margar mercr gulls uppi fix the bóts, how many marks of
sculu vera epter þann er af var gold shall be paid on account of
tecinn, oc velldr þat at marger him who was slain, and the cause
vito eigi hvat laga bót er, er þó of that is that many know not
at vissi, þá vilia nú fáer því una. what the lawful bót is, and though
En Frostoþings bóc scipter they knew it, few will now abide
lagabót hveriom epter sínum by it. But the Frostathing book
burð oc metorði, en ecki hinum divides the lawful bót to every one
bótum er þeir ofsa eða vansa er according to his birth and rank,
í dómum sitia oc sáttmál gera. and not those bóts (bœtr) which
they that sit in courts and make
terms of peace put too high or too
low.
 Here the writer clearly refers back to the ancient Frostathing book
as the authority for the ‘lawful bót,’ but on examination he seems to
add certain additional bóts which the courts now include in the
round amount of so many gold marks awarded by them in each case
as it comes before them.
The writer takes first the case of an award of six marks of gold
and describes how it is to be divided, and then the case of five
marks of gold, and so on.
The division is throughout made in silver marks,
ores, ortugs, and penningar. But when the items Division of it in
silver marks at
are added up, the total in silver divided at the ratio ratio of 1:8.
of 1:8 brings back the result as nearly as may be to
the number of gold marks from which the division started. Thus in
the clause describing the division of the wergeld of six marks of
gold, the silver items add up to 48 marks exactly, and the division of
this by 8 brings back the amount to six marks of gold. And so in the
clause dividing five marks of gold, the items seem to add up to one
ortug only less than 40 silver marks, and again a division by 8 brings
the amount sufficiently near to five marks of gold.
In each case, however, the writer adheres to the
same scheme of division. When he has 6 gold The group of
Bauga men. The
marks to divide he first assigns 18 silver marks to other group of
Bauga men (i.e. the near group of kinsmen of male Nefgildi-men.
descent on the paternal side only), and then he
adds half as much (i.e. 9 marks) to a group of Nefgildi-men[180]
among whom are included, with others, kinsmen of descent through
females on both paternal and maternal sides. So that these two
groups of Bauga men and Nefgildi get 27 marks. In all cases he
makes the group of Nefgildi receive only half the amount received by
the Bauga group, the whole amount being reduced according to the
number of gold marks to be divided. After the amount allowed to
these two groups, the remainder is made up of additional payments
some of which he expressly declares were not included under old
law. Thus (in clause 6) he adds an amount which he says was ‘not
found in the old Frostathing book’ and justifies it by saying that
there would be danger to the slayer if it was not paid. And so again
(in clause 9), there are additions for half-brothers, half-brothers’
sons, &c., of the same mother. And these additions are included in
the six marks of gold ‘according to new law.’
Evidently, therefore, we must not take these wergelds of six and
five marks of gold with their divisions as representing the ancient
customary wergelds of this class or that in the social scale, but
rather as showing the extent to which the system of wergelds had
become somewhat arbitrarily expanded and elastic in later times.
The total amount with additions was apparently increasing as time
went on.
As in the Frostathing law so also in the Gulathing
law (clause 316, p. 104) there is a statement of Later statement in
the Gulathing law.
wergeld, avowedly of a late date and added under
the name of Biarne Marðarson, who lived about a.d. 1223. And this,
too, seems to belong to a time when the amount of the wergeld was
awarded by some public authority in so many marks of gold. He
takes the case of a wergeld of six marks of gold and shows how it
ought to be divided; and then the case of a wergeld of five marks of
gold and shows how that should be divided—‘What each shall take
of five marks of gold’ and so on—just as was done by the writer in
the Frostathing law.
One might have supposed from this that, as the method of
awarding fixed amounts and the amounts to be divided in gold
marks were the same, so the groups and the persons included in
them would have corresponded also. But they differ considerably.
Biarne Marðarson up to a certain point follows the same scheme
as the writer in the Frostathing.
In his division of six marks of gold he, too, draws a line at the
amount of 27 marks, and he also divides this amount into thirds and
gives two thirds to one group and one third to the other. The son of
the slain and the brother of the slain form the first group and take
18 marks, and a second group take 9 marks, the two together taking
27 marks.
The group who together take 9 marks, like the Nefgildi-men of the
Frostathing, embraces however by no means the same relatives as
are included in the latter. The only persons included are the father’s
brother and his children, i.e. first cousins or brœðrungs of the slain,
but among them are included the sons of concubines and of female
first cousins. And after the mention of these is the statement, ‘All
that these men take amounts to 27 marks and 2 aurar.’ Out of the
remainder of the 6 gold marks or 48 silver marks other relations take
to the ‘fifth man’ on the male line and the sixth on the female line.
Biarne Marðarson seems, like the writer in the Frostathing law, to
have had to some extent a free hand in the division. It is clear that
there was much variety in the course adopted. Nor does he seem to
have been by any means so systematic and accurate as the other
writer. The silver amounts, when added up, do not so accurately
correspond with the six gold marks to be divided.
We turn, then, from these later statements to
what seems likely to be an older statement of the Earlier wergeld of
the Gulathing law.
Norse wergeld, viz. that which commences at In silver marks
clause 218 of the Gulathing law. and cows.

It describes the division of the wergeld of a ‘hauld’ or ‘odal-born’

man, and it begins with the explanation that the ‘mannsgiöld’ or
wergeld decreases and increases from this as other retts.
It describes the various amounts both in silver marks and in cows,
which the other statements do not, and this, so far as it goes, is a
sign of antiquity.
In clause 223 is inserted a statement of the various things in
which wergelds may be paid. The only item the value of which is
given is the cow, which is to be taken at 2½ ores if not older than
eight winters and if it be ‘whole as to horns and tail, eyes and teats,
and in all its legs.’ And this silver value of the cow—2½ ores—is the
one used in this older description of the wergeld.
 The wergeld according to this statement consists of bauga
payments and upnám payments. The first are received in three
baugs or rings thus:—

Höfuð (head) The 64 cows of

baug, taken by the Bauga group.
the son and the
father of the
slain 10 marks or 32 cows.
Bróður baug,
taken by
brother, or if
none, by the son
of the slain 5 marks or 16 cows.
Brœðrungs baug
taken by the
father’s brother’s
son, i.e. first
cousin of the
slain 4 marks or 13 cows - ½ ore.
19 marks or 60 cows + 2 ores.

To this is added
for women’s
gifts, i.e. the
mother,
daughter, sister,
and wife of the
slain, or in
default to the
son of the slain 1 mark or 3 cows + ½ ore.
Total 20 marks or 64 cows.

After this statement is the declaration, ‘Now all the baugs are
counted.’
 A clause is here interpolated changing the point of view so as to
show how, and by whom on the slayer’s side the same three baugs
were paid.

Nú scal vigande bœta syni (222) The slayer shall pay to the
hins dauða hafuðbaug. son of the dead the höfuð baug.
En bróðer viganda scal bœta The slayer’s brother (if he has
brœðr hins dauða bróðor baug, one) shall pay to the brother of
ef hann er til, ellar scal vigande the dead the bróður baug;
bœta. otherwise the slayer shall pay it.
Nú scal brœðrongr viganda The brœðrung of the slayer (if
bœta brœðrongi hins dauða he has one) shall pay to the
brœðrongs baug, ef hann er til, brœðrung of the slain the
ellar scal vigande bœta. brœðrungs baug; otherwise the
slayer shall pay it.
Sá er sunr hins dauða er við He is [reckoned] the son of the
giölldum tecr, hvárt sem hann dead who takes the giöld, whether
er faðer æða bróðer, æða hvigi he is father or brother or however
skylldr sem hann er. he is related.

Then follows the declaration, ‘Now the baugs are separated’ (‘Nú
ero baugar skildir’).
It seems clear, then, that the slayer was in the last resort
responsible for the whole of these baug payments, as it was the son
of the slain who would take any part of them lapsing through failure
of the designated recipients.
The small payments to the mother, daughter,
sister, and wife included in the baug payments are Women’s gifts.
evidently additional and exceptional payments in
regard to close sympathy. The slayer does not make these
payments. It is expressly stated that they are made ‘by the
kinswomen of the slayer,’ but they are included in the even amount
of 20 marks or 64 cows.
 The recipients of the three baugs, it will be seen, were limited to
the nearest relatives on the paternal side—fathers, sons, brothers,
and first cousins—with no descent through females, while the
recipients in the next set of groups or ‘upnáms’ include also relations
through females: but, again, only males receive.
There is, however, one exception. In clause 231 is the following:—

Nú ero konor þær allar er All those women who have sons
sunu eigu til sakar, oc systr are in the sök (suit), and sisters
barnbærar. þá scal þeim öllum capable of bearing sons. They
telia söc iamna, til þær ero shall all be held to have an equal
fertogar. part in it till they are forty.

Evidently they partake, as under Cymric custom, only in respect of

possible sons who if born would partake themselves. Indeed, the
sons only appear in the list of receivers and in no case the mother,
except among the women’s gifts included as above in the baug
payments.
Clause 224 describes the upnám set of recipients
as under. It will be seen that they include The upnám group
includes
descendants of great-grandparents, but no more descendants of
distant relations. great-
grandparents.

‘Sac-tal of upnáms or groups outside bauga men.’

1st upnám.

Father’s brother (i.e. uncle).

Brother’s son.
The slain person’s { Mother’s father.
Daughter’s son.

Each gets a mark from the slayer if a hauld be slain; and this
amounts to 4 marks.