Westermarck's Ethical Relativism

Tekijät: Stroup Timothy
Julkaistu: 1980
Julkaistu filosofia.fi sivustolla: 17.10.2013
Timothy Stroup
Westermarck's Ethical Relativism
Ajatus 38, 1980, 31-71.

Liitteet

PDF: Skanneroitu/Skannad/Scanned: Stroup1980.PDF
Timothy Stroup
Westermarck's Ethical Relativism
Ajatus 38, 1980, 31-71.

Westermarck's Ethical Relativism
Timothy Stroup

Let us at this stage of the argument about subjectivism take a brief rest and look round a special view or assemblage of views which has been built on the site of moral disagreements between societies. This is relativism, the anthropologist's heresy, possibly the most absurd view to have been advanced even in moral philosophy.1

Ethical relativism is a view with a noble, if spotty, history. Older than emotivism and attitudinal subjectivism, and one of the chief reasons for formulating those doctrines, it can be traced back to ancient times: Herodotus recorded the varying funeral practices of Indians and Callatians and concluded that "custom is king over all";2 and for Plato relativism provided a convenient foil which gave impetus to his own theory.3 But it was the great advance in anthropological method in the nineteenth and twentieth centuries which

-31-


appeared to provide a new and convincing support for an ethics which formed itself around actual moral practices.4

What shape emerged from this process? According to H. B. Acton, "All philosophers have some rough idea of what they mean by 'ethical relativism.'"5 Yet Acton himself speaks of "two main forms of ethical relativism," one which emphasizes the diversity of moral codes and another which focuses on the nature of the moral judgment itself;6 Richard Brandt distinguishes among descriptive, normative, and meta-ethical relativism;7 Paul Taylor identifies social or cultural relativism, psychological or contextual relativism, theoretical or logical relativism, and methodological relativism;8 John Barnsley classifies the phenomena as evaluative relativism, relativism of moral Tightness, epistemological relativism, and axio-

-32-


logical or value relativism;9 John Ladd notes differences even within cultural relativism between diversity and dependency theses;10 and so forth. If one thing is clear, it is that "ethical relativism" can stand for a bewildering number of views differing not only in content, but also in the level at which the inquiry is pursued.

Given this variousness, it is scarcely surprising that the critics of ethical relativism have found almost irresistible the temptation to single out the most superficial and untenable strands. As is the case with subjectivism in general, ethical relativism has suffered unnecessarily by being the target of outworn and jejune attacks against postitions which few have ever held; and the philosophical legacy of these attacks has been to prevent a relativist picture from receiving any fair consideration.

Edward Westermarck is generally regarded as one of the founders of anthropological relativism; indeed, W. T. Stace has credited Westermarck with coining the phrase "ethical relativity."11 Westermarck's book of that title,12 published in 1932, was substantially a reformulation and explication of theses he had advanced most comprehensively in his earlier work on The Origin and Development of the Moral Ideas (1906—08),13 which in turn grew out of discussions at the Philosophical Society of Helsinki and scattered preliminary articles.14 Westermarck declared that the object of his inquiry was

-33-


"to study the moral consciousness as it displays itself among mankind at large";15 ultimately he concluded that "moral judgments could not possibly possess that universality which is characteristic of truth..."16 Since Westermarck allowed a place in his ethics for an examination of the historical and comparative facts of moral diversity, it may erroneously be assumed that he falls prey to the stock objections to naive relativism and that his views can be dismissed as being of little philosophical interest. It is worthwhile, therefore, to consider the extent to which relativism itself is tenable and whether Westermarck was committed to an indefensible form of it.

The discussion of Bernard Williams in his book Morality: An Introduction to Ethics can serve as an example of an attack based upon caricature of the relativist's position. Williams writes that relativism:

consists of three propositions: that 'right' means (can only be coherently understood as meaning) 'right for a given society'; that 'right for a given society' is to be understood in a functionalist sense; and that (therefore) it is wrong for people in one society to condemn, interfere with, etc., the values of another society…Whatever its results, the view is clearly inconsistent, since it makes a claim in its third proposition, about what is right and wrong in one's dealings with other societies, which uses a nonrelative sense of 'right' not allowed for in the first proposition.17

Williams calls 'the view he attacks the "vulgar and unregenerate form" of relativism,18 but he 'does not explain why discussion of the subject is more profitable with this stalking-horse instead of a sophisticated, regenerate relativism;19 nor does he identify any actual

-34-


relativist who adhered to such a view. Westermarck certainly is not guilty of holding any of Williams's three propositions: he did not assert the first proposition because he did not offer a logical analysis of moral utterance which reduces concepts to simple equivalents in terms of individual or group attitudes, but rather sought a psychological, genetic description of the way moral concepts are formed; as Kai Nielsen correctly remarks, Westermarck "had no very clear theory about the logical status of moral utterances. His value as a moral philosopher lay elsewhere."20 He did not maintain the second proposition as a functionalist, which he was not, although he could have agreed with the general goal of contextual understanding of social practices. And he did not hold the third proposition, at least in Williams's sense, for reasons to be discussed later.21

But are the three propositions even inconsistent, as Williams claims? Surely not. In the first place, if there is any inconsistency, it would result from the joint assertion of the first proposition and a particular interpretation of the third proposition which anyone holding the first proposition would be unlikely to make; but, at any rate, the second proposition is unnecessary in generating the supposed inconsistency. Second, read literally, the first proposition gives no definition of 'right,' for the term appears in both definiendum and definiens. This fault can easily be avoided by substituting "accepted as the preferred practice within a given society," or some such phrase, for "right for a given society." Third, anyone who adopts such a definition of "right" clearly must admit the naturalistic consequence that moral judgments about what is right are true or false, depending upon whether the society in question does or does not prefer the corresponding behavior. Since the third proposition is a moral judgment, it will be true for any society in which tolerance is the preferred practice; and because it is possible for there to be such a society, it is possible for the third proposition to be true.

-35-


Hence, acceptance of the first proposition entails that there are some conditions under which the third could be true.

The viewpoint expressed by the first proposition has the further consequence that there can be no genuine inter-societal moral contradiction although there can still be inter-societal moral disagreement in any but the most dogmatic sense of "disagreement": people from different societies can argue about moral matters, but there may be an element of ultimate futility to their dispute. However, a relativism which accepts the first proposition in a thoroughgoing way thereby allows both for intra-societal contradiction and for a method of determining which disputant is correct, and this is a powerful concession to the objectivist. Relativists of this sort scarcely are relativists at all, because they admit a sphere, and a large one at that since moral judgments typically are intra-societal, in which moral judgments are objective: moral judgments are relative to society but absolute within a society.22 An exponent of such a view can arrive at it from two different perspectives: as a quasi-objectivist who places, for example, high moral value on obedience to authority while noting that authorities vary; or as a quasi-relativist who wants to reconcile a world without universal moral facts with moral judgments which are products of cultural forces. A relativism at the inter-societal level, coupled with an objectivism at the intra-societal level, is logically consistent, if perhaps implausibly half-hearted, and no inconsistency results when a methodological principle for understanding social practices is appended.

What happens, however, if Williams's third proposition is interpreted as advancing a general, inter-societal claim about tolerance? Clearly there is a way in which inconsistency can be generated: if a person says both "There are no absolute moral truths" and "It is an

-36-


absolute moral truth that everyone should be tolerant," then contradiction results. But would relativists embrace anything like the latter statement? To issue the first of Williams's propositions is to recognize that "right" has a meaning different from that in common parlance, because ordinary speakers are not initially relativists. Yet how is "right" — or actually "wrong," which is the word that appears and which will be understood as its contradictory — used when the third proposition is so interpreted? A relativist who uses the term in Williams's sense, which is a "nonrelative" sense, would be unusually obtuse, having only two propositions earlier advocated a conceptual clarification of the term "right." This, of course, does not mean that an overzealous and "vulgar" relativist may not do just this, but there is no reason to let the case for relativism be deflected by these unlikely pitfalls. Rather than being guilty of inconsistency, Williams's relativist appears to commit the fallacy of equivocation, which is, to be sure, not much better; but more likely it is Williams himself who encounters difficulties in shaking off the ordinary meaning of "right."

Furthermore, to deny that the relativist can consistently and unequivocally utter all three propositions is to conclude that relativism implies moral nihilism, for if the first two propositions cannot be held along with even so gentle a moral prescription as that of tolerance, there is little scope for the relativist to advance any moral judgments at all. But this is far from being the case, and Westermarck, who adopted a stance on the issue of tolerance which was wholly consistent with his ethical relativism, provides the best example. If by "nihilism" is meant "value-nihilism," the second-order thesis that moral judgments are neither true nor false, then it is clear from his chapter on "The Moral Concepts" that Westermarck never seriously considered a radical emotivism, let alone prescriptivism. He writes: "I thought it was generally recognized that every proposition is either true or false, and that this must consequently be the case also with the proposition 'this is good,' whatever be the meaning of its predicate."23 As for a more general first-order nihil-

-37-


ism, a reading of Westermarck's autobiography gives ample evidence 'that he was devoted to moral causes.24

Hence Westermarck was neither nihilist nor objectivist, but was he consistent in so being? He certainly was aware of the charge that relativism or subjectivism entailed moral nihilism, for he said: "... my denial of objective moral standards does not prevent my pronouncing moral judgments which are expressions of my own moral feelings... "25 The remark comes in response to a criticism which G. E. Moore levelled in his paper "The Nature of Moral Philosophy." Moore argues that it is "commonly believed that some moral rules exhibit a higher morality than others."26 To say, using Moore's example, that it is better to be kind to friends and enemies alike than merely to friends is to recognize a higher standard of conduct. The illustration is poorly chosen, because outside of Christian theory (which is quite different in this respect from the practices of most Christians), the injunction to treat enemies kindly is often seen as little more than, foolishness and hardly the epitome of moral wisdom. But Moore's general point behind 'the illustration holds: when we argue about whether one moral rule is higher than another, we do not think that we are arguing merely about attitudes. Moore continues:

What then, could Westermarck mean by saying that A's morality is higher than B's? So far as I can see, what, on his own views, he would have to mean is merely that he himself, Westermarck, shares A's morality and does not share B's: that it is true of him, as of A, that neglecting to do good to enemies excites his feelings of moral indignation and not true of him as it is of B, that it does not excite such feelings in him. In short he would have to say that what he means by calling A's morality the higher is merely 'A's morality is my morality, and B's is not.' But it seems to me quite clear that
-38-


when we say one morality is higher than another, we do not merely mean that it is our own.27

Much of what Moore says in this passage is an accurate reading of Westermarck's views, but it is not (therefore a valid criticism of those views. This can be seen by distinguishing between Westermarck and the plain man, between a theorist whose account of moral judgments is colored by an empirical understanding of social behavior, and a pre-philosophical, perhaps a little naive, ordinary moral disputant. Some of Moore's claims relate to the former and some to the latter. Moore charges that Westermarck would have to accept the consequence that when he, Westermarck, brands a moral practice as higher he is merely revealing his own acceptance of that practice; and this is a consequence that Westermarck is prepared to accept, for he is not, to borrow Williams's word, an unregenerate moral analyst.28 On the other hand, moral judgments are, for the plain man, a combination of feeling and belief, expression and objectification. The latter components of these pairs, insofar as the corresponding moral judgments make categorical and ultimate, rather than hypothetical and derived, claims, have illusory aspects, for there are, in Westermarck's view, no moral facts. It therefore behooves the theorist who recognizes the error of objectivizing to avoid it in his own pronouncements about morality, to forgo professing the objective and universal validity of the moral emotions which he happens to feel.

This is precisely where the issue of tolerance enters the picture; Westermarck does not advocate tolerance as a universally binding moral fact, which would then be inconsistent with the rest of his view; indeed, he is far from commending extreme tolerance even as personal preference:

-39-


I do not even subscribe to that beautiful modern sophism which admits every man's conscience to be an infallible guide. If we had to recognise, or rather if we did recognise, as right everything which is held to be right by anybody, savage or Christian, morality would really suffer a serious loss. 29

Rather, Westermarck offers the empirical claim that "could it be brought home to people that there is no absolute standard in morality, they would perhaps be on the one hand more tolerant and on the other hand more critical in their judgments."30 One of the chief causes of intolerance is the fact that, as Leslie Stephen has put it, "Each man thinks that his own morality is (the right morality, and that the ordinary standard is mistaken and immoral so far as it deflects from it. He does Mot say that your morality is erroneous, but denies it to be morality at all."31 Hence, the first point is that those who recognize the error of moral discourse would, within limits, be more tolerant. Even this is a matter of degree, because feelings can be held intolerantly; one of the props for intolerance has been removed, yet others remain. But also the regenerate moral relativist would have a more critical understanding of moral judgments: they should not continue to be used to make absolute moral claims. Is this "should," however, a word of moral injunction? The answer is, No more than it is a moral command not to contradict oneself.

-40-


There is no absolute moral obligation to avoid self-contradiction; it is rather an axiom of logic 'than of morals that a consistent error theorist cannot continue to commit the same old errors, and from this it is but a small step to propose that the error theorist be consistent. If, of course, the step is one which is viewed as entailing even a bit of genuine objectivism, then it need not be 'taken, for the error theorist can fall back on the weaker claim of mere personal preference for consistency, 'although such preference is widely held and close to being indispensible, yet not absolutely so — one can live with glaring inconsistencies. As Westermarck assents, "... what unprejudiced person can help changing his views if he be persuaded that they have no foundation in existing facts?",32 and this holds true as well of the imprudence of being Stubbornly inconsistent.
   
To return to Moore, the reconstructed moral judgments of a consistent error theorist will shun objectivizing elements, and hence Westermarck accepts the charge that he is announcing his approval when he calls one moral practice higher than another. But this is not the only thing that the ordinary moral disputant does, as Moore recognizes; and Westermarck can agree with this, because he realizes that he and the plain man are not doing the same thing when they engage in moral disputes, It is therefore a non sequitur on for Moore to move from the consideration of what Westermarck's own moral judgments involve to conclusions about ordinary moral judgments. Moore says, in the last sentence quoted above, (that we do not simply mean that a given view is our own when we call it higher than another. Westermarck replies: "I have no doubt that this is the case with most people's judgments, but this does not disprove my view that their assumed objectivity is an illusion."33
   
The regenerate relativist of the Westermarckian sort continues to have moral emotions and these will influence his moral judgments. He will even continue to use the ordinary moral vocabulary, for there is not much alternative:

-41-


. . . every one of us makes use of the words sunrise and sunset, which are expressions from a time when people thought that the sun rose and set, though nobody now holds this view. Why, then, should not the ethical subjectivist be allowed to use the old terms for moral qualities, although he maintains that the objective validity generally implied in them is a mere illusion?34

All that is required of him, on grounds of consistency, is that an illusory belief .that moral judgments state moral facts should no longer play a role in his moral psychology. He is therefore entitled to call one thing right and another wrong, but "right" and "wrong" will now have different meanings for him from what they once had. They now relate, insofar as they are used in ultimate judgments, not to facts at all, but to the relativist's endorsement of modes of conduct. Only by holding fast to the ordinary meaning of moral concepts in his personal moral judgments, while at the same time offering an entirely different theory about what people mean when they make moral judgments, can the "vulgar" relativist conjure up the inconsistency against which Williams directs all his guns. That Westermarck was not similarly "vulgar" can be seen quite clearly from the conclusion of a sentence which was previously quoted only in part:

There is no inconsistency in this [that A's morality is my morality, and B's is not] : my denial of objective moral standards does not prevent my pronouncing moral judgments which are expressions of my own moral feelings, and whatever terms I use they have to be interpreted accordingly. 35

The subject of interpretation of words should not be left until some further difficulties are removed. Westermarck sometimes employs words which typically are used to express moral approval, and these have been cited by critics as examples of inconsistency. It has been argued that the complaint in general is unfair, but in some cases it is particularly unfair because Westermarck 'intends no moral connotation at all. Three such words are "higher" (in

-42-


some contexts), "development," and "progress."36 "Higher" is, perhaps, the clearest example, because Westermarck uses it in two ways: he is insistent that he be allowed to use it in moral contexts, but also emphatic that it has non-moral uses as well. In Is Conscience an Emotion?, Hastings Rashdall denounces Westermarck for referring to some emotions as higher than others, when Rashdall claims that no such judgment is warranted on an emotionalist account.37 Westermarck replies that the context in which the words "higher emotions" occurred "ought to have made it quite plain that I attached no moral significance at all to this expression."38 An examination of Westermarck's personal copy of Rashdall's book makes this point even clearer. At the passage where Rashdall says "In one place he actually talks about emotions depending upon cognition as 'higher emotions,' " Westermarck observes in the margin: "not morally higher/Higher animals, etc."39 It is no more inconsistent for Westermarck to talk about higher emotions than it is for any moral skeptic to talk about higher education, higher mathematics, or higher criticism.

On the other hand, Westermarck was quite aware of the misguided attempts of some writers to blur the distinction between description and evaluation by using loaded terms. His discussion of the phrase "sufficiently developed moral consciousness" shows that he did not intend to admit objectivity through the back door by appealing to criteria which are only apparently empirical:

... when speaking of a 'sufficiently developed' moral consciousness (beyond insistence upon a full insight into the governing facts of

-43-


each case), we practically mean nothing else than agreement with our own moral convictions. The expression is faulty and deceptive, because, if intended to mean anything more, it presupposes an objectivity of moral judgments which they do not possess, and at the same time seems to be proving what it presupposes.40

Westermarck uses the word "higher" because he believes that it has legitimate non-monad as well as moral uses. The same is true of "development" and "progress." In an unpublished lecture on "Den moraliska utvecklingen" which he delivered in Stockholm shortly after 'the publication of .the first volume of The Origin and Development of the Moral Ideas, Westermarck was quite emphatic in pointing out that he intended no moral overtones in his use of the word "development":

I use the expression 'moral development' in a very simple sense, without any hint of some general evolutionary formula such as that which, e.g., provides the basis for Herbert Spencer's description of the development of morality. By 'moral development' I understand purely and simply the variations in the moral ideas which the human race has undergone on the way from primitive existence to civilization.41

"Higher," "development," and "progress" are words which tend to appear in each other's definitions. For example, one of the meanings of "progress" in The Oxford English Dictionary is:

Going on to a further or higher stage, or to further or higher stages successively; advance, advancement; growth, development, continuous increase; usually in good sense, advance to better and better conditions, continuous improvement.

The words have both moral and non-moral uses, although, in the case of "progress" in particular, the former are more common. In-

-44-


deed, Westermarck himself sometimes uses these words in commendatory ways, as is only to be expected of someone who refused to devise a new vocabulary simply because moral words had been coined by others. But he also uses (them to describe changes to more accurate, precise, and complicated ways of perceiving things, as when he contrasts -the progress of science with that of morality. The former has witnessed "a perpetual succession of new discoveries,"42 an increasingly refined (again in a non-moral sense: exact, distinct) differentiation of the phenomena of human experience; the latter exhibits little progress at .all, because it is not a branch of knowledge, but rather only the sphere in which human moral emotions work themselves out.

The trouble with many of these descriptive words is that there is fairly general agreement that it is better to have the properties which the words attribute than not to have them, and eventually this tacit approval gets built into the connotation of the word. It is then difficult to recall that a word like "progress" can be used to describe a process as well as to evaluate it.43 Thus Ronald Fletcher has cited a passage from The Origin and Development of the Moral Ideas which he views as inconsistent over the notion of progress:

Could it be brought home to people that there is no absolute standard in morality, then they would perhaps be somewhat more tolerant in their judgments, and more apt to listen to the voice of reason. If the right has an objective existence, the moral consciousness has certainly been playing at blindman's buff ever since it was born, and will continue to do so until the extinction of the human race. But who does admit this? The popular mind is always inclined to believe that it possesses the knowledge of what is right and wrong, and to regard public opinion as the reliable guide of conduct. We have, indeed, no reason to regret that there are men who rebel against the established rules of morality; it is more deplorable that the rebels are so few, and that, consequently, the old rules change so slowly. Far above the vulgar idea that the right is a settled something to which everybody has to adjust his opinions, rises the conviction that "in it has its existence in each individual mind, capable of any expansion,

-45-


proclaiming its own right to exist, and, if need be, venturing to make a stand against the whole world. Such a conviction makes for progress.44

Fletcher comments:

This passage is utterly riddled with inconsistencies; but the logical absurdity of Westermarck's position is made starkly clear in the word 'progress' with which he ends it. How could he possibly believe that his own ethical proposals were 'better' or more 'progressive' than the other ethical opinions to which he referred, if all moral persuasions are subjectively adequate for every person and every society in accordance with the feelings and conceptions they happen to possess?45

For reasons already advanced, no part of the quotation from Westermarck should be viewed as inconsistent. Some parts make empirical assertions (the comment about tolerance), some express moral preferences (it is "deplorable that the rebels are so few"), some exhibit a combination of the two (the last sentence). Westermarck's remark at the end about progress has two components: rebels force change, and intellectual rebels change knowledge in the direction of greater precision and accuracy (an empirical observation); and change of this sort is worthwhile (a moral valuation).

The regenerate relativist is entitled to hold regenerate moral beliefs, but is he similarly entitled to compare his own beliefs favorably with those of others? This is one of the questions that Fletcher raises in his comment about the passage from Westermarck, but, despite Fletcher's use of "better" in quotation marks, Westermarck does not use this word in the passage. One of the reasons he avoids the word is perhaps that his emphasis on tolerance as a virtue has the consequence that moral comparisons should, on his account, be made with a little more reticence and a greater understanding of other people and the contexts in which they find themselves. On the other hand, to a regenerate relativist, the word "better," when used to compare two moral practices, does not function in the same

-46-


way that it does in ordinary moral discourse; the relativist who expresses his moral preferences by calling them better than the preferences of others is not thereby laying stake to absolute moral prerogatives.

This conclusion may appear to reduce morality to a wholly arbitrary phenomenon, and this possibility must be examined carefully. For Westermarck's account is intended to be an empirical explanation of human behavior, and if it entails that moral beliefs are whimsical when, quite to the contrary, they are everywhere accorded supreme importance, then the theory will fail to explain the facts of moral experience. The charge of arbitrariness is, in fact, the second prong of the attack against naive relativism, arguing as it does that the tolerance principle of the relativist, which is claimed to be inconsistent with his non-objectivism, is purchased at too high a price: we do not want to tolerate some things and the whole point of morality is (to provide a criterion whereby we can distinguish the immoral (indeed, in some cases, the atrocious) from the moral. So long as the great bulk of mankind continues to be pre-philosophical, innocent objectivists, there is little to worry about, even though they rely upon the opium of moral truth; but if everyone were to become persuaded of the correctness of a relativistic account, would not the result be a moral disorder in which all views would be on an equal footing? The less sanguine critics offer a fearful picture: "... Westermarck's subjectivist doctrine is both disastrous and disadvantageous for ethics once we understand the implications of this view. On the practical side, the implication is moral chaos — interpersonal and international."46 Rashdall says, "The emotionalist theory of ethics however little intended to have that result by its supporters, is fatal to the deepest spiritual convictions and to the highest spiritual aspirations of the human race."47 Williams, on the other hand, sees the chief threat as resulting from the collapse of relativistic tolerance into indifference: "It is possible for someone persuaded of subjectivist views to cease to care about moral issues."48

-47-


The concern about indifference is a needless anxiety. It is, of course, possible that an indoctrinated subjectivist will cease to care about morality, but it is only as likely as his ceasing to have moral emotions. Indeed, to think otherwise is to commit what Ayer has called "the fallacy of thinking that the absence of grounds for morality entails the absence of motives."49 Individual tastes, for example, are widely thought to be subjective, but they still thrive. If a person recognizes that his love of whiskey and his detestation of milk are not universal dictates of reason, it does not follow that he will therefore drink the two indiscriminately. On the contrary, as D. H. Monro has observed,50 there may even be reasons why we will argue for our admittedly subjective tastes: that is one of the best ways to secure them. Moral paralysis or apathy resulting from relativistic tolerance is surely a less real and troublesome possibility than the insufficient tolerance which is now characteristic of moral beliefs.

Moral chaos, however, is by far the least attractive of the supposed consequences of relativism. The positing of this nightmare, of course, greatly overstates the likelihood that a Westermarckian account would ever be 'adopted on a large scale, and it also assumes erroneously that the removal of the rational component of moral beliefs would result in the dominance of the basest emotions. Mill tried to assuage the parallel fear that adherence to utility as the standard of conduct would encourage people 'to act like pigs, as (though pig-like desires were only held in check by suppressive reason.51 But moral emotions can survive intellectual changes with fundamental

-48-


attitudes remaining much the same: there are Protestant, Catholic, and Jewish atheists.52 Furthermore, if every single vestige of the external validity of moral truths were removed, if ultimate moral judgments were seen as merely expressive and prescriptive, would our own moral principles then deserve no respect from us: "... why should the moral law command less obedience because it forms a pant of ourselves?"53 And even if the dire consequences predicted by Westermarck's antagonists did ensue, that would not invalidate the meta-ethical (theory of human behavior. As Westermarck noted:

It is needless to say that a scientific theory is not invalidated by the mere fact that it is likely to cause mischief. The unfortunate circumstance that there do exist dangerous things in the world, proves that something may be dangerous and yet true.54

But the major fault in such a criticism of Westermarck in particular is that it entirely mistakes or ignores his analyses of the natural constraints on moral conduct and the role of reason in morality. It claims that if Westermarck's picture were to become generally accepted, there would be nothing left to morality but the expression of emotions in arbitrary judgments.

There are at least four senses of the word "arbitrary" in which this conclusion does not follow: moral judgments, on Westermarck's account, are neither capricious, artificial, unexplainable, or irrational. Westermarck has two supports for a defense from :the charge of arbitrariness, one derived from his view of how we arrive at our

-49-


emotions, the other from his account of the relationship between reason and emotion in moral judgments.

Westermarck offered an evolutionary account of the genesis of moral emotions and moral concepts, according to which moral development is a natural process. Morality is not some artificial device elaborately constructed to achieve an arbitrary goal, but rather a growth of patterns of reacting to the natural settings in which people find themselves: "We approve and disapprove because we cannot do otherwise; our moral consciousness belongs to our mental constitution, which we cannot change as we please."55 Westermarck explicitly says of moral feelings that they "are not arbitrary."56 They are not capricious, but rather fairly well-settled ways of responding to important concerns; they are not artificial, but spring from our moral consciousness, which is itself a naturally acquired phenomenon; and they are not unexplainable, because there exist psychological, biological, and sociological methods for describing how moral beliefs have come to be what they are.

More surprising is the indictment that Westermarck deprecates the role of reason in moral judgments, which then are seen by critics as becoming irrational exercises of the emotions. Rashdall is the most notorious offender on this score. He quotes Westermarck's statement in The Origin and Development of the Moral Ideas that "could it be brought home to people that there is no absolute standard in morality, they would perhaps be somewhat more tolerant in their judgments, and more apt to listen to the voice of reason."57 Rashdall replies, " 'The voice of reason,' forsooth, when the whole chapter is a diatribe against the notion that Reason has anything to say about conduct!"58 In the margin of his copy of Rashdall's book, Westermarck has written "! ! ! ! Shameless!" He expounds on his outrage in a footnote in Ethical Relativity: "This inaccuracy is astounding. In that very same chapter I have said ... 'The influence of intellectual considerations upon moral judgments is certainly

-50-


immense.'"59 In the second volume of The Origin and Development of the Moral Ideas Westermarck again says that "the influence of intellectual considerations upon moral judgments is naturally very great."60 This awareness of the role of reason in the formation of moral judgments is found in Ethical Relativity also: "Though all moral judgments are ultimately based on emotions, the influence that intellectual factors exercise on such judgments is very great indeed."61 And one of the chief causes of the variability of moral judgments is assigned by Westermarck to the intellectual determinants of the judgments, to different ideas about the objective nature and consequences of certain modes of conduct.

No adequate treatment of Westermarck's account of moral beliefs can fail to allow for the importance he attaches to reason as a component of them.62 With some simplification, it can be said that if the origin of moral beliefs is due chiefly to emotional factors, their development is due chiefly to rational factors. But it should be recognized that simplification is involved in this assertion, because Westermarck, although he does distinguish sharply between emotion and reason, does not thereby exclude either from the analysis of those processes which constitute moral behavior.

First, the origin of moral beliefs. Westermarck is adamant in asserting that moral concepts and moral judgments are ultimately based on emotions, and the emotional origin of moral ideas is the chief principle of his construction of an alternative to objectivism.

-51-


Yet reason still enters into beliefs at the formative stage, because the perception of facts determines what attitude we shall take towards them:

... I recognized at the same time the enormous influence that intellectual considerations exercise upon those judgments, in the first place through the cognitions by which the moral emotions are determined.63
Emotions depend on cognitions and are apt to vary according as the cognitions vary...64
The belief that gives rise to an emotion, the cognitive basis of it, is either true or false ...65
Emotions are determined by cognitions and differ in nature or strength according as the cognitions differ.66
Emotions are determined by specific cognitions, and last only as long as the influence of those cognitions last [sic].67

By way of practical illustration of the manner in which reason operates on emotion in the formation of moral beliefs, consider the analysis that Westermarck gives of the ascription of moral responsibility. One of the many consistent threads in his moral philosophy is the contention that moral judgments are passed on the intentions of moral agents. But if the will or character of the moral agent is the focal point of a valuation, then the question of the origin of 'that character "has something to do with the thoroughness of the judgment."68 Westermarck's example is of a libertine whose ideas are shaped by an environment that is itself corrupt; such a person is generally held less blamable than one who, exposed only to good influences during his maturation, nevertheless sinks to licentiousness.69 It may be granted that even the trained libertine could have

-52-


resisted the training and consequently cannot wholly be absolved from blame; "but in any case 'the influences of environment and the circumstances of upbringing are not irrelevant to the degree of his guilt."70 On the contrary, we distinguish between original character and acquired Character, between, as Westermarck puts it in a manuscript fragment, "a person's original I and the I a person becomes, which is a product of external circumstances."71

Our perception of the factual circumstances in which a given action occurs will affect our perception of the agent's acquired character and hence our moral judgments about him. But what about blame directed at the "original I," the innate character of the individual? People appear to have, if anything, less control over their innate than their acquired characters; does the notion of responsibility therefore collapse? Westermarck cites the views of Martineau that to ascribe blame is "in itself as absurd as to applaud the sunrise or be angry at the rain..."72 and Balfour that "the virtuous man will, indeed, continue to deserve and receive admiration of a certain kind — the admiration, namely, which we justly accord to a well-made machine ..."73 For the present purposes, however, it is not

-53-


necessary to reach a particular solution to the problem of free will, nor even to consider the general relationship between free will and responsibility, but rather to see the effect that reason has on the origin of moral beliefs about responsibility; for if, on the determinist's account, human conduct is wholly determinate, then it seems absurd to use the ordinary, free-will-based concepts of responsibility to describe the way people act.

But do determinists find it easy in actual life to forgo such ascriptions? No, because changes in rational (albeit empirical) perceptions of the general structure of the universe do not necessarily effect corresponding changes in the feelings of indignation and admiration which provoke the use of blaming and praising words. More specific perceptions of relevant facts do enter into our ascriptions of responsibility: we retract or diminish blame upon hearing that an action was not intentional, but accidental, and we cease feeling sympathetic sorrow if we find that we mistook a friend's joy for sadness. But large-scale changes in world-views are less likely to have an immediate impact on the continuing existence of moral emotions:

. . . however convinced I am that a person's conduct and character are in every detail a product of causes, that does not prevent me from feeling towards him retributive emotions — either anger or gratitude, moral resentment or approval. This suggests that a retributive emotion is not essentially determined by the cognition of free-will.74

The determinist is not unlike a moral skeptic who, despite a theory which emphasizes the role of objectivizing in the formation of moral beliefs, does not thereby cease to have moral emotions. This supports Westermarck's view that "ethical writers are often inclined to overrate the influence of moral theory upon moral practice..."75 The example of determinism and responsibility shows that emotions are basic to the origin of moral judgments, but that reason does enter into the process also, not in a general way to eliminate all

-54-


moral emotions which do not conform to an empirical picture of the ultimate nature of reality, but in a specific way to assure that particular emotions are directed at appropriate causes of pleasure and pain.

The scope of reason in the development of moral beliefs is even more obvious than its scope in their origin. Fletcher has overstated its role, but there is much truth in his observation: "Westermarck's further theory is to the development of moral ideas was entirely a matter of his views concerning the place of reason in morality ... ."76 This exaggerates the importance of reason, because altruism also has a significant function in the development of moral ideas, but this is a less egregious error than Rashdall's of ignoring Westermarck's treatment of reason entirely.

The clearest way to see the effects of reason on the development of moral beliefs is to determine whether the evidence of human moral behavior reveals a diversity of moral judgments. At the same time such an inquiry can thwart another naive objection to naive relativism, which pictures the relativist as concluding from his anthropological investigation of moral practices that "chaotic diversity of opinion is the rule in the moral realm"77 and that eo ipso objectivism cannot be true; the objectivist's reply, on the other hand, is that the apparent diversity can be explained and at any rate would not in itself be a disproof of objectivism. Curiously, on this issue Westermarck sides in large measure with the objectivist.

By the development of moral ideas Westermarck refers to the processes by which they evolve, grow, spread, and mature not only within one society, but in all societies, and in each society through a period of time. Even in his earliest scholarly investigation, the undergraduate honors thesis he submitted on "Gör kulturen människosläktet lyckligare?,"78 Westermarck pictured a world in which there is a spectrum, for the most part chronologically ordered, along which moral practices change from the less to the more sophisticated. The contrast between "savage" and civilized man is still

-55-


drawn in Ethical Relativity, but in the latter work he is more cautious in noting discrepancies between the moral beliefs of the two. There is, indeed, a great amount of agreement about moral practices concerning the prohibition of such serious crimes as homicide and theft and the praise of acts of charity: "when we study the moral rules laid down by the customs of savage peoples we find that they in a very large measure resemble the rules of civilized nations."79 Nor is this surprising, even on a relativistic account, for if the emotions provide the ultimate source of moral judgments, then those judgments will show many similarities, because people resemble each cither in their basic emotional urges as well as in their intellectual orientation and capacities:

Such judgments are passed on conduct and character, and if they were guided by sufficient knowledge of facts and reflection there would be no essential difference between them as regards the general subjective conditions of the modes of conduct to which they refer. This uniformity as regards the nature of their subjects is due to the facts that they are all based on moral emotions, and that the moral emotions are retributive emotions felt towards persons conceived as causes of pleasure or pain.80

Various peoples exhibit basic similarities in their natures, but there are also rational and emotional differences which influence the ways in which moral ideas develop. Consider first the cognitive causes of moral diversity, which can be attributed to different ideas about the objective nature of similar modes of conduct and the consequences of such conduct. On Westermarck's account, these differences arise in three specific ways. First, the moral behavior of mankind encompasses varying degrees of knowledge and reflection. Societies which manifest a high level of scientific sophistication will have different moral codes from those which rely on animistic explanations of natural events. In Christianity and Morals Westermarck cites an example: "... at the lower stages of civilisation there is a considerable lack of discrimination between intentional injuries and accidental ones,"81 and this imprecision is reflected in moral

-56-


beliefs. Unlike abrupt individual Shifts in rational outlook which may not affect the individual's moral emotions, the evolution of social ideas and perceptions Ito more complex and differentiated forms is usually a gradual process which exerts a pervasive force on 'the members of the society. Second, much moral diversity results from different ideas which arise from the different external circumstances in which people find themselves. Westermarck offers the practice of infanticide as an example.82 Societies vary greatly in the extent to which they are favored by conditions propitious to life. Subsistence societes cannot tolerate large populations, and unlimited childbirth may be viewed as disadvantageous on economic grounds, although these considerations are not always drawn so explicitly. The ultimate consequence is that two societies may agree to the general moral principle of placing a premium on human life, yet disagree about whether infanticide is necessary to avoid degrading the lives of those who cannot be cared for, or is abhorrent because it involves the taking of life. Third, Westermarck cites differences in beliefs which result from superstition or religion, and this category is really a subcategory of the first: "In almost every branch of conduct we notice the influence which the belief in supernatural forces or beings or in a future state has exercised upon the moral ideas of mankind, and 1ihe great diversity of this influence."83 In fact, Westermarck's last work, Christianity and Morals, is entirely devoted to tracing the extent to which a particular set of religious beliefs has resulted in an elaborate code of moral behavior.

The conclusion of Westermarck's inquiry into the rational bases of moral development is that a great amount of moral diversity can be explained as resulting from intellectual factors and consequently provides no ammunition for the relativist:

In so far as differences of moral emotion depend on knowledge or ignorance of facts, on specific religious or superstitious beliefs, on different degrees of reflection, or on different conditions of life or other external circumstances, they do not clash with that universality

-57-


which is implied in the notion of the objective validity of moral judgments.84

But what about emotional differences? It has been conceded that the emotional characteristics of different peoples exhibit general similarities, and this is indeed what makes possible Westermarck's classification of the retributive emotions. Yet this is not the whole Story. Morris Ginsberg has distinguished six types of variation in moral beliefs, and most of those that would be accepted by Westermarck can be explained as intellectual in origin. However, this is not true of Ginsberg's first category, "variations in the range of persons to whom moral rules are held to be applicable."85 Variations of this sort are, for Westermarck, mainly due to variation in the sentiment of altruism, which has an emotional origin. The degree of altruism in a society is a function of many factors, the chief among them being the size of the social unit and the amount of intercourse with other societies: social isolation and differences of race, language, and customs work to inhibit the feeling, and peaceful inter-societal encounters foster it because prudence requires friendliness in such relationships.86 Altruism is manifested in many ways. In "primitive" morality exclusive reference is made to members of the same community, and an act which is condemned when perpetrated against a member of the same tribe may be praised when an outsider bears its brunt; "civilized" society has not managed to obliterate the unequal treatment of citizens and foreigners, but "both law and public opinion certainly show a very great advance in humanity with regard to the treatment of foreigners,"87 and the improvement is even more noticeable in the injunctions of the normative systems of Western society than in the informal behavior of their adherents. And altruism does not have merely a human reference: it is evident in the attitude we take towards animals, although in this respect the so-called "civilized"

-58-


societies have not always been very advanced. Yet Westermarck notes an improvement in nineteenth century Europe in the kindness accorded to animals, and he believes that this can be accounted for partially by intellectual factors — for example, by more vivid reflection about what happens when animals suffer — but that emotional factors also enter in, with the result that diversity in moral judgments about animals prevails because there is a diversity of altruistic sentiment towards them:

... though greater intellectual discrimination may lessen the divergencies of moral opinion on the subject, nothing like unanimity may be expected, for the simple reason that humanity to animals is ultimately based on the altruistic sentiment, and sympathy with the animal world is a feeling which varies greatly in different individuals.88

Thus the specific manner in which moral rules have been extended so that they prohibit actions against strangers and animals reflects the nature and degree of the altruism that prevails in a society. Westermarck concludes, "Whatever part reflection may have played 'in the expansion of the moral rules — prudence has also, no doubt, had something to do with the matter — it seems to me obvious that the dominant cause has been the widening of the altruistic sentiment."89

To return to the naive objection to naive relativism, that attack was two-sided: it claimed that the relativist misstates the evidence of anthropology as being a record of thoroughgoing divergence in moral belief and that he then moves illicitly to conclude that moral diversity proves that objectivism is false. It can now be seen that this criticism is inappropriate when applied to Westermarck. First, at the very heart of Westermarck's account of moral judgments is an emphasis on objectivizing. According to Westermarck, a moral judgment purports to describe 'the actual moral properties of things, but succeeds only in revealing the utterer's moral emotions, which are projected onto the phenomenon. Yet this very process requires

-59-


that there be substantial uniformity in the intellectual and emotional structure of mankind, for otherwise there would be mo such tendency to objectivize our feelings:

This objectivity ascribed to judgments which have a merely subjective origin springs in the first place from the similarity of the mental constitution of men, and, generally speaking, the tendency to regard them as objective is greater in proportion as the impressions vary less in each particular case.90

A considerable amount of regularity must be accepted in order to be consistent in explaining moral objectivizing. Second, the empirical evidence of diversity does not counteract objectivism in any simple way, for to a large extent differing moral beliefs are the products of differences in intellectual outlook. The scope of irreconcilable moral diversity must therefore be considerably reduced. Nor does the plain fact of diversity offer an iron-clad argument against objectivism, because mere dispute does not in itself show that neither disputant is right. Actually, this is somewhat too Strong, because the very appearance of persistent differences which cut across the moral beliefs of both ordinary people and moral theorists is strong evidence against any meta-ethical claims that moral truth is readily grasped through simple acts of intuition.91 Third, it is important, in determining what divergent moral beliefs show about moral truth, to inquire into the causes of those divergences. As Westermarck observed, "The validity or fallacy of this argument [from diversity to subjectivity] depends in the first place upon the causes to which the variability of moral judgments is due."92 That is why the emphasis placed on altruism in the development of moral ideas raises the level of discussion considerably beyond Westermarck's earlier account in "Normative und psychologische Ethik,"93 a lecture which he presented to an interna-

-60-


tional congress during his formative years. There he contrasted disagreement in ethics and disagreement in science and conceded that much of the former resulted from varying intellectual perceptions. But the young Westermarck, still in the initial stages of his inquiry into moral practices, did not examine in any detail the causes of ethical diversity. Once this is done, it becomes clear that some changes in the moral ideas within a society over a period of time, as well as some variations in moral ideas from society to society at a given time, have "fundamentally different"94 causes from disagreements in theoretical matters. Whatever differences exist in the latter "can be removed by sufficient observation and reflection"95 but differences in the former have a tenacity which discloses a residue of varying emotional reactions behind them. Even among the most sophisticated individuals — moral philosophers perhaps — there are enduring divisions over moral questions, and this is only to be expected if moral judgments are ultimately based upon emotions. Rational considerations do enter into the analysis, for moral emotions are dependent upon cognitions; but cognitions which are qualitatively and quantitatively similar can give rise to disparate moral beliefs in individuals whose emotional make-ups differ. While basic regularities set broad limits to emotional dissimilarities — "certain cognitions inspire fear into nearly every breast"96 — there is still room for individual variation: heroes and cowards will have quite different reactions to a situation in which they have identical rational perceptions of the level of danger. Westermarck concludes:

The same holds true of the moral emotions. To a large extent, as we have seen, their differences depend upon the presence of different cognitions, but very frequently the emotions also differ though the cognitions are the same. The variations of the former do not interfere with the belief in the universality of moral judgments, but when the variations of the moral emotions may be traced to different persons' tendencies to feel differently in similar circumstances on account of the particular nature of their altruistic sentiments, the supposed universality of moral judgments is a delusion.97

-61-


Westermarck's theory is, after all, an explanatory one, and it can be judged by how well it accounts for the phenomena of human moral behavior. As such it is compatible with a world in which some moral arguments between individuals and societies get resolved, while others remain stubbornly fundamental. It makes no hasty conclusion that diversity reveals the error of objectivism, but rather examines the reasons why apparently diverse moral practices reality are diverse in order to show that not all moral disputes can be assimilated to the model of theoretical knowledge.

Throughout the preceding discussion the assumption has been made that moral practices can be compared from society to society, and that the only question is whether such comparison reveals them to be fundamentally similar or divergent. This assumption appears reasonable enough, but the methodology which it may reflect has been subjected to such vitriolic attack that it may be worth while to append a brief discussion of Westermarck's views about the value of a comparative sociological approach in reaching conclusions about human behavior. Obviously, no attempt can be made within the present confines to rescue 'the comparative method from the weight of contrary opinion, but it is of some interest, in appraising the role of Westermarck's sociological presuppositions in the development of his moral philosophy, to establish the nature and extent of his debt to cross-cultural comparison. E. R. Leach, for example, has remarked that "the evolutionist comparative method has achieved a kind of massive futility in the vast tomes of Frazer and Westermarck,"98 and Westermarck has been severely criticized on methodological grounds by Lowie, Mills, and others.99

-62-


Westermarck offered the most detailed defense of his scientific credo in the Introduction to the expanded edition of The History of Human Marriage100 and in his Huxley memorial lecture on "Methods in Social Anthropology,"101 but even so he cannot be charged with the general criticism levelled by T. D. Campbell that "many modern works in social science spend too much time expounding their methodological approach and too little time doing anything significant with it."102 Westermarck was sensitive to, but not preoccupied with, matters of method, and his sensitivity was increased by the realization that his comparative approach to the study of social phenomena, "which for half a century has been dominant among British students of social anthropology,"103 was no longer fashionable. This change was hardly surprising, as he himself comments, because the very fact that the method had been applied on a large scale for so many years guaranteed that mistakes would be made and detected.104 But Westermarck tried to defend a comparative method against three specific and serious charges 'that had been made against it.

In the first place, it was argued, by critics like W. H. R. Rivers, that comparative sociology too often confused the separate disciplines of sociology and psychology, and that only when the former "has made far greater advances than the present, will it be profitable to endeavour to explain the course of social life by psychological processes."105 This is a criticism which, if valid, would be devastat-

-63-


ing to Westermarck's entire work, which has a consistent emphasis on psychology throughout it. Levi-Strauss notes that "for Westermarck, sociological explanation is never satisfactory in itself and ... to grasp the phenomenon in an intelligible fashion one must transcend the social and attain... the psychological level always, and the biological level whenever one can."106 Westermarck is willing to concede that psychological explanations encounter special difficulties and in many cases are no more than provisional, but why should psychological considerations not be joined to sociological investigation; why must sociology be "pure" in Rivers's sense? Westermarck observes, "It is strange that this extraordinary faith in sociological explanations should be coupled with an equally extreme distrust in our capacity of learning the motives by which social conduct is determined."107 Human beings are not automata, and only distortion can result from focusing sociological attention exclusively on mechanical processes detached from the intentions and motives behind them. This is not intended to license uncritical psychological dabbling in the study of social behavior, nor to deny that the disciplines of social anthropology and psychology have differences in level, approach, and emphasis. But Westermarck's goal was to achieve, in the form of testable hypotheses, a synoptic vision of human moral beliefs, for which purpose it was necessary to draw upon both sociology and psychology. No claims were made that further refinements were unattainable or irrelevant in either field, that caution was unnecessary in relating psychological to sociological facts, or that the two fields could be directly fused into an undifferentiated synthesis; and the contrast was not with what the future might hold, but what the past actually did hold — grandiose systems which never went beyond mere speculation because they were based upon little or no factual data at all. Westermarck did not attempt to do more — or less — than to make the best use he could of whatever empirical materials were at hand, and since psychological and sociological materials were fundamentally complementary rather than inconsistent, he believed that he was entitled

-64-


to appeal to both to give a comprehensive, though tentative and improvable, picture of morality.

Second, the comparative method has been held to be inherently prone to error and simplification because its scope is too broad to permit the careful scrutiny of individual societies that is necessary to generate accurate data for comparison. Westermarck has often been accused, usually without any concrete examples being provided, of careless methods which necessarily incorporate inaccuracies into his account.108 His autobiographical comments about his research procedures belie the general charge of carelessness,109 but he was ready to acknowledge that there was truth in the accusation of fallibility; even in the first edition of The History of Human Marriage he noted, "As the sociologist is in many cases unable to distinguish falsehood from truth, he must be prepared to admit the inaccuracy of some of the statements he quotes."110 But the remedy for error, he held, was greater accuracy rather than lesser scope. On the other hand, Westermarck's research methods were not characterized only by large-scale comparative studies. As a result of his fieldwork in Morocco, his approach to anthropological

-65-


investigation became more direct and modern.111 Describing in his autobiography why he restricted his travels to Morocco and gave up his more ambitious itinerary, Westermarck declared "better much about little than a little about much."112 His own fieldwork in Morocco is a clear disproof of any criticism that he did not appreciate the possibilities opened up by first-hand, in-depth methods in social 'anthropology.

The third Objection argues along functionalist lines that in principle the attempt to isolate the responses of different societies to a particular behavioral issue wrenches the individual practice out of an organic cultural whole in which it alone makes sense.113 Moral beliefs, for example, are not simply individual products, but must be considered within their social context. But this criticism can scarcely be levelled against Westermarck. He surely understood and emphasized 'the broad cultural influences out of which moral beliefs are manufactured, and his Moroccan research indicates that he prized monographic studies devoted to examining every aspect of a particular culture (indeed, more so than Durkheim, who never carried on comparable research).114 Yet Westermarck did not view the comparative and functionalist approaches as incompatible:

There is no real opposition between the study of a cultural phenomenon as it is distributed among different races and the study of it which is restricted to a particular ethnic group. Here again the

-66-


methods differ simply because the subjects differ. But the two kinds of investigation complement each other.115   

Furthermore, like the determinist who cannot shed his ordinary notions of responsibility, the functionalist can scarcely forgo drawing comparative conclusions if he is to know the significance of the institutions he seeks to study in a given society:

It is easy to criticize the comparative method in the point we are now considering [the debt of the generalist to the specialist], but it is impossible for any modern student of human civilisation to ignore its results. The writings of Professor Durkheim and his disciples are thoroughly pervaded by the teachings of the very school whose method they have so severely criticised. Does not this show that there must be exaggeration in their criticism?116

The reason why a comparative approach works, given safeguards to insure that the data are accurate and contextually understood, is that people all belong to the same 'animal species and therefore have certain basic resemblances which are more pervasive than the individual differences which result from the varying circumstances in which they find themselves:

How could we disclose these resemblances in any other way than by comparison? How could we otherwise distinguish that which is local from that which is general? Nay, how could we fully explain the social environment itself without taking into account the mental characteristics of the human species? I think there is sufficient evidence to show that innumerable customs and beliefs are not so closely interwoven with the social tissue that they cannot, with due precaution, be abstracted from it for the sake of comparison.117

-67-


The efficacy of such a general application of a refined comparative method can be seen from Westermarck's works on moral philosophy themselves. As Fletcher has described them:

The work of Westermarck is an example of scholarly study in which, at one and the same time, theories were being formulated to explain a wide, systematic survey of concrete facts, and the investigation of facts was being undertaken in the light of testable assumptions — both at the level of documentary evidence and at the level of face-to-face observation in the field… Though it is true that Westermarck did draw comparative details from their wider context, he was, nonetheless, well aware of their functional setting within societies, and, given the mass of comparative details which he succeeded in bringing together.. . the merits of his work far outweigh this particular inadequacy of his methodological approach.118

Westermarck set two tasks for himself, those of amassing a compendium of moral practices and beliefs and of drawing general conclusions about moral behavior based upon the most systematic empirical study that was possible at the time.119 The former he approached as an ethnographer, the latter as a comparative sociologist; but the tasks were not unrelated. He recognized (that his survey of anthropological and historical data seemed to have "disabled some of my earlier readers from seeing the wood for the trees,"120 and he noted with astonishment McTaggart's assertion in a review that The Origin and Development of the Moral Ideas "consists not so much in any theory which could be quoted and examined as in the vast mass of well-arranged facts which he has collected

-68-


with reference to every branch of morality."121 Indeed, McTaggart's appreciation wholly ignores the goal behind the compilation of moral data.122 Westermarck was a careful historian of moral ideas, but he considered his purpose to extend beyond mere accumulation of historical details, as can be seen from his disparaging comment about Spencer: "Spencer's theory might at the most be a contribution to the history of the growth of moral ideas, but could have no bearing whatever on the question of their validity."123 Westermarck's appeal to facts on a large scale has more philosophical than sociological significance: an empirical ethics must be set upon a solid evidentiary base, and this requires a description and comparison of the particular workings out in society of mankind's biological endowments. Thus he said that The Origin and Development of the Moral Ideas "was so constructed that the comparative and historical treatment of the moral ideas in the last three-quarters of the book was necessary to confirm the general theory of the nature of the moral consciousness set forth in the earlier chapters."124 Errors may creep into the analysis, as Westermarck admitted, and the ideal procedure would be to compare only on the basis of properly conceived monographic studies of particular societies, yet the practical requirements of reaching tentative ethical hypotheses cannot await ideal conditions, but must rely upon the best evidence available; and piecemeal studies on their own, in the absence of generalization, are insufficient documentation for an empirical analysis of human behavior in general. Sophisticated comparison serves a purpose for which a restricted functionalism offers no substitute, and to deny the validity of such an investigation is to rob ethical inquiry

-69-


of much of its empirical foundation and to postpone the drawing of conclusions until some nebulous, future ideals of precision can be attained. As A. R. Radcliffe-Brown has observed:

For social anthropology the task is to formulate and validate statements about the conditions of existence of social systems (laws of social statics) and the regularities that are observable in social change (laws of social dynamics). This can only be done by the systematic use of the comparative method, and the only justification of that method is the expectation that it will provide us with knowledge of the laws of social development.125

The ultimate defense of Westermarck's methods, which combined intensive field study with a wider observation of human society in general, lies in the explanatory power of, and evidentiary support for, his hypotheses; and as he claimed, no other theory, "whatever arguments have been adduced in support of it, has been subjected to an equally comprehensive test."126

There has been in the presentation of Westermarck's relativism an element of talking around the issue, of disposing of actual and possible objections to his view without pausing to consider what that view actually consists in. To a large extent this approach mirrors the treatment of the subject in Ethical Relativity, where, despite the title, Westermarck is strangely unforthcoming in offering any straightforward or systematic account of what he means by relativism. Indeed, the words "relativity" and "relative" are scarcely used in the work. But the style of analysis in this article has been prompted by .the belief that Westermarck's relativism can best be brought out indirectly along two fronts: first, by indicating the empirical concerns which motivated him to dissect moral beliefs; and second, by stressing what his ethical relativism was not. This negative caution results from the very nature of Westermarck's inquiry, which did not address ethical issues along modern episte-

-70-


mological lines, but rather sought to erect a unified theory which could account for the facts of moral experience. By adequately allowing for the interplay of reason and emotion, by separating personal preferences for tolerance from the descriptive functions of an empiricist meta-ethics, and by shunning the more recent temptation to analyze moral discourse in any simple (and oversimplified) way, Westermarck was able to present a sociological, psychological, and biological relativism which survives the misplaced complaints of his critics.


John Jay College of Criminal Justice, The City University of New York

This research was supported (in part) by a grant from the PSC-BHE Research Award Program of the City University of New York.

-71-


Notes
1. Bernard  Williams,  Morality:  An Introduction  to  Ethics  (New  York, 1972), p. 21.
2. Herodotus, History, III. 38.   See also the accounts of different customs in David   Hume, An Enquiry concerning the Principles of Morals, in Enquiries concerning the   Human Understanding and concerning the Principles of Morals,  ed. L. A.  Selby-Bigge,  2nd ed. (Oxford,  1962),  pp. 324—343  ("A Dialogue");  and Adam Smith,  The Theory of Moral Sentiments, new ed. (London, 1892),  pp.  290ff.  (V.ii.).
3. However, H. D. Oakeley tries to identify relativistic features even in Plato's view; see "Is Ethical Relativity Necessary?," in Action, Perception and Measurement, Aristotelian Society Supplementary Volume 17 (London, 1938), pp. 152—169.
4. For examples see William Graham Sumner, Folkways: A Study of the Sociological Importance of Usages, Manners, Customs, Mores, and Morals, rev. ed. (Boston, 1940); Ruth Benedict, Patterns of Culture (Boston, 1934); Melville J. Herskovits, "The Problem of Cultural Relativism," in Man and His Works: The Science of Cultural Anthropology (New York, 1952), pp. 61—78, and Cultural Relativism: Perspectives in Cultural Pluralism, ed. Frances Herkovits (New York, 1972); and Robert Redfield, "The Universally Human and the Culturally Variable," in Human Nature and the Study of Society: The Papers of Robert Redfield, ed. Margaret Park Red-field, vol. 1 (Chicago, 1962), pp. 439—453. General accounts of some of the main themes of cultural relativism are given by Clyde Kluckhohn, "Ethical Relativity: Sic et Non," The Journal of Philosophy 52:23 (1955), pp. 663—677; Kai Nielsen, "Ethical Relativism and the Facts of Cultural Relativity," Social Research 33:4 (1966), pp. 531—551; and Abraham Edel, Ethical Judgment: The Use of Science in Ethics (New York, 1955).
5. H. B. Acton, "Is Ethical Relativity Necessary?," in Action, Perception
and Measurement
, p. 170.
6. Ibid., pp. 170—171.
7. Richard B. Brandt, "Ethical Relativism," The Encyclopedia of Philosophy, ed. Paul Edwards, vol. 3 (New York, 1967), pp. 75—76.  A parallel terminology ("empirical,"   "normative," and "philosophical"   theses of ethical relativism) is offered by Steven Lukes in "Relativism: Cognitive and Moral," Aristotelian Society Supplementary Volume 48 (London, 1974), p. 173.
8. Paul W. Taylor, "Four Types of Ethical Relativism," The Philosophical Review 63:4 (1954), pp. 500—516.
9. John Barnsley, The Social Reality of Ethics: The Comparative Analysis of
Moral Codes
(London, 1972), pp. 323—336.
10. John Ladd, "Introduction," Ethical Relativism, ed. Ladd (Belmont,
Cal., 1973), pp. 1—11.
11. W. T. Stace, Man Against Darkness and Other Essays (Pittsburgh, 1967), p. 249.
12. Edward Westermarck, Ethical Relativity (London, 1932). Hereafter referred to as "ER."
13 Edward Westermarck, The Origin and Development of the Moral Ideas (London, 1906-08), 2 vols. Hereafter referred to as "ODMI I" or "ODMI II."
14. The manuscript of the Philosophical Society minutes, entitled "Filosofisen Yhdistyksen Pöytäkirjat, 1873—1915," is located in the Department of Philosophy, Helsinki University. The most important of the preliminary articles are "Remarks on the Predicates of Moral Judgments," Mind N.S. 9:34 (1900), pp. 184—204, and "Remarks on the Subjects of Moral Judgments," Proceedings of the Aristotelian Society N.S. 4 (London, 1904), pp. 111—126.15  ER, p. xvii.    .
16. ibid., p. 218.
17. Williams, op. cit., pp. 20—21.
18. Ibid., p. 20.
19. Williams does claim (p. 20) that his presentation captures the "most influential" form of relativism, but this seems scarcely consistent with his own assessment of  the blatant implausibility of  the position. Gilbert Harman, in "Moral Relativism Defended," The Philosophical Review 84:1 (1975), p. 3, has remarked of this passage, "It is easy enough to show that this version of moral relativism will not do, but that is no reason to think that a defender of moral relativism cannot find a better definition."
20. Nielsen, "Varieties of Ethical Subjectivism," Danish Yearbook of Philosophy 7 (Copenhagen, 1972), p. 284. For an account of Westermarck's treatment of moral judgments, see Stroup, "In Defense of Westermarck," forthcoming in The Journal of the History of Philosophy.
21. See below, pp. 39—41.
22. Actually, "absolute" is an unfortunate term, because it merely signifies that something is unconditional, has no exceptions. Thus, an emotivistic relativist can hold absolute principles and a naturalistic relativist can hold objective principles, but no relativist can maintain that there are absolute, ultimate, objective principles. For purposes of simplicity, "absolute" will be used, as it generally is in discussions of relativism, to imply "objective" and contrast with "relative."
23. ER, p. 142.
24. See, for example, his campaign for Finnish nationalism, related in Westermarck, Minnen ur mitt liv (Helsinki, 1927), pp. 185—194, or Memories of My Life, trans. Anna Barwell (London, 1929), pp. 149—156.
25. ER, p. 146.
26. G. E. Moore, "The Nature of Moral Philosophy," Philosophical Studies (London, 1922), p. 334.
27. ibid., p. 335.
28. See ER, p. 146.  Westermarck recognized this even in his early draft of ODMI: "If anybody says to me, 'It is right to kill,' and I answer him, 'What you say is not true,' my answer, strictly speaking, implies the assertion only that the judgment in question is not in harmony with a moral law which I recognize as binding." Westermarck, "Introduction," Edvard Westermarcks Brevsamling, Addenda, n.p. Åbo Akademis Bibliotek, Åbo, Finland. This collection is hereafter referred to as "EWB."
29. ODMI I, p. 19. A similar point is made in ER, p. 58. Thus Stace, The Concept of Morals (New York, 1937), p. 58, wholly misunderstands Westermarck when he quotes a passage (ER, p. 59) about tolerance and then observes, "Certainly, if we believe that any one moral standard is as good as another, we are likely to be more tolerant. We shall tolerate widow-burning, human sacrifice, cannibalism, slavery, the infliction of physical torture, or any of the thousand and one abominations which are, or have been, from time to time approved by one moral code or another." Objectivist widow-burners, however, could use a measure of relativist tolerance. And even Hume did not think that everything goes: "...nor is it conceivable, that any human creature could ever seriously believe, that all characters and actions were alike entitled to the affection and regard of everyone." Hume, op. cit, pp. 169—170 (I).
30. ER, p. 59; ODMI I, p. 20.
31. Leslie Stephen, The Science of Ethics, 2nd ed. (London, 1907), pp. 35— 36.
32. ER, p. 60. But the recalcitrance of the moral emotions may enable them to survive the onslaught of some kinds of facts; see below, pp. 53—54.
33. ER, p. 146.
34. ibid., p. 49.
35. Ibid., p. 146. (Emphasis added)
36. Others include "civilized," "advanced," "savage," "lower," "archaic," "enlightened," etc.
37. Hastings Rashdall, Is Conscience an Emotion? (London, 1914), p. 123. A similar charge is made in Rashdall, The Theory of Good and Evil: A Treatise on Moral Philosophy, 2nd ed., vol. 2 (London, 1924), p. 413, n. 1; this passage is cited with approval in William S. Kraemer, "Ethical Subjectivism and the Rational Good,"   Philosophy and Phenomenological Research 12:4 (1952), p.  531. The passage in Westermarck is ODMI II, P. 744.
38. ER, p. 146.
39. Rashdall,  Is  Conscience  an Emotion?,  p.  123.  Westermarck's  copy is in Åbo Akademis Bibliotek.
40. ODMI I, p. 12; ER, pp. 46, 217.
41. Westermarck, "Den moraliska utvecklingen" ["Moral Development"], delivered 5  July 1907, EWB,  Box 82,  p. 1. See also his Philosophical Society lecture, delivered 27 March 1908, on "Moralens utveckling" ["The Development of Morality"], in "Filosofisen Yhdistyksen Pöytäkirjat, 1873—1915"; and Westermarck's comments about Spencer in ER, pp. 20ff.
42. ER, p. 215.
43. The Swedish word "framsteg" ("progress") exhibits parallel features.44 ODMI I, p. 20.
45. Ronald Fletcher, The Making of Sociology: A Study of Sociological Theory, vol. 2 (London, 1971), p. 112.
46. Stuart L. Penn, "The Ethical Relativism of Edward Westermarck: A Critique"  (Ph. D. dissertation, Yale University,  1957), p. 576.
47. Rashdall, Is Conscience an Emotion?, p. 200; quoted in ER, p.  57.
48. Williams, op. cit, p. 27.
49.  A. J. Ayer, The Central Questions of Philosophy (New York, 1974), p. 225.
50. D. H. Monro, Empiricism and Ethics (Cambridge, 1967), p. 116.
51. John  Stuart Mill,  Utilitarianism,  12th  ed.  (London,   1895),  pp. 10ff. Darwin also encountered this objection. F. P. Cobbe, in "Darwinism in Morals,"  The Theological Review 8:33  (1871), p.  175,  argued that if the evolutionists had their way, "... I cannot but believe that in the hour of their  triumph would be sounded  the  knell  of the  virtue of mankind." Darwin, in The Descent of Man and Selection in Relation to Sex, new ed. (London, 1901), p. 152, replied, "It is to be hoped that the belief in the permanence of virtue on this earth is not held by many persons on so weak
a tenure." See also W. C. Swabey, "Westermarckian Relativity," Ethics 52:2 (1942), p. 224.
52. This point was made in conversation by J. L. Mackie, who ascribes it to Santayana.
53. ER, p. 59; ODMI I, p. 19. As Hume put it, "Nothing can be more real, or concern us more, than our own sentiments of pleasure and uneasiness; and if these be favourable to virtue, and unfavourable to vice, no more can be requisite to the regulation of our conduct and behaviour." Hume, A Treatise of Human Nature, ed. L. A. Selby-Bigge (Oxford, 1967), p. 469 (III.i.1).
54. ER, p. 58; ODMI I, p. 18. Hume, in An Enquiry, p.  279 (IX.ii.), commented: "And though the philosophical truth of any proposition by no means depends on its tendency to promote the interests of society; yet a man has but a bad grace, who delivers a theory, however true, which, he must confess, leads to a practice dangerous and pernicious."
55. ER, pp. 58—59; ODMI I, p. 19.
56. ER, p. 58; ODMI I, p.  19.
57. ODMI I, p. 20; quoted in Rashdall, Is Conscience an Emotion?, p. 124.
58. Rashdall, Is Conscience an Emotion?, p. 125.
59. ER, p. 147, n. 60. Westermarck is quoting ODMI I, p. 10.
60. ODMI II, p. 744.
61. ER, p. 147. In a review of ER, Huntington Cairns took a singularly aggressive stance in defending the role of reason in Westermarck's theory and the practical consequences of adopting Westermarck's view: "It seems to me that Professor Westermarck's general  position from the point of view of the good life is immeasurably superior to that of his opponents. In judging moral precepts, if his view is adopted, the force of tradition is minimized and the role of reason is increased. This is a merit which few objective moralists can claim for the systems proposed by them." Cairns, untitled review, The Baltimore Evening Sun 45 (6 August 1932), p. 4.
62. See Morris Ginsberg, "The Function of Reason in Morals," Proceedings of the Aristotelian Society N.S. 39 (London, 1939), p. 257, for an accurate appraisal.
63. ER, p. xvii; ODMI I, p. 10.
64. ER, p. 59; ODMI II, p. 744.   
65. ER, p. 60.   
66. Ibid., p. 147.
67. Ibid., p. 180. In EWB, Box 24, there is even a small note which Westermarck wrote to himself: "Influence of reason on emotions-state more fully." At one point he contemplated a chapter of ODMI on "The Influence of Intellectual Factors upon Moral Judgments." (EWB, Box 76.) See also ODMI I, pp. 2—3.
68. ER, p. 178.   
69. Smith, op. cit, p. 291 (V.ii.), has a similar example.
70. ER, p.  178; ODMI I, pp. 324—325.
71. Collection of the author; see also ODMI I, p. 325.
72. James Martineau, Types of Ethical Theory, 3rd  ed. (Oxford, 1891), pp. 41—42. This and the following quotation are cited in ER, p. 179; ODMI I, p. 320.
73. Arthur James Balfour, The Foundations of Belief, being Notes Introductory to the Study of Theology, 8th ed. (London, 1901), p. 27. Balfour's view is not too far removed from Westermarck's on this matter. Although Balfour notes (p. 22) that he has "nothing to do here with the truth or untruth of either of the contending theories," he stresses the difficulties the determinist faces in dispensing with notions of responsibility and also gives an evolutionary account of why conscious animals would have a sense of
free will. But Westermarck took strong exception to the thesis that man can be thought of as a mere machine. In an interview he declared that Balfour "made a great error when he said that, on the naturalistic hypothesis, man could only be regarded as a well-made machine. Whatever Determinist doctrines we may hold, we do not and cannot treat man as a machine. Even the philosopher who denies free-will to his dog will never treat his dog as a machine." The comment appears in F. J. Gould, "With
Dr. Westermarck," The Literary Guide and Rationalist Review N.S. 30 (1 December 1898), p. 186.
74. ER, p. 181; ODMI I, p. 322.
75. ER, p. 59. That is why he said only that subjectivism would "perhaps" lead to greater tolerance.
76. Fletcher, op. cit., p. 105.
77. Penn, op. cit., p. 2.
78. Westermarck, "Gör kulturen människoslägtet lyckligare?" ["Does Civilization Increase the Happiness of Mankind?"], EWB,  Box  77.
79. ER, p. 197.
80. ibid., pp. 183—184.
81. Westermarck, Christianity and Morals (London, 1939), p. 34.
82. ER, pp. 185—187.
83. Ibid., p. 187.
84. Ibid., p. 196.
85. Ginsberg, "On the Diversity of Morals," Essays in Sociology and Social
Philosophy
, vol. 1: On the Diversity of Morals (London, 1953), p. 101.
86. ER, p. 200.
87. Ibid., p. 199.
88. Ibid., p. 213.
89. Ibid., p. 207.
90. ODMI I, p. 8. On the next page Westermarck talks of the "comparatively  uniform  nature of the moral consciousness."
91. ER, pp. 42—43.
92. Ibid., p. 183.
93. Westermarck, "Normative und psychologische Ethik," EWB, Addenda.
94. ER, p. 215.
95. Ibid., p. 215.
96. Ibid., p. 216.
97. Ibid., p. 217.
98. E. R. Leach, "The Epistemological Background to Malinowski's Empiricism," in Man and Culture: An Evaluation of the Work of Bronislaw Malinowski, ed. Raymond Firth (London, 1957), p.  121.
99. Robert H. Lowie, The History of Ethnological Theory (New York, 1937), pp. 97—99; C. Wright Mills, "Edward Westermarck and the Application of Ethnographic Methods to Marriage and Morals," in An Introduction to the History of Sociology, ed. Harry Elmer Barnes (Chicago, 1948), pp. 654—667; see also Robert Briffault, The Mothers: A Study of the Origins of Sentiments and Institutions, 3 vols. (New York, 1927); V. F. Calverton, "The Compulsive Basis of Social Thought: as Illustrated by the Varying Doctrines as to the Origins of Marriage and the Family," The American Journal of Sociology 36: 5 (1931), pp. 689—720; and G. Duncan Mitchell, A Hundred Years of Sociology (London, 1968).
100. Westermarck, The History of Human Marriage, 5th ed., vol. 1 (London, 1921), pp. 1—25.
101. Westermarck, "Methods in Social Anthropology," Journal of the Royal Anthropological Institute 66    (1936), pp.  223—248. See also Westermarck, Minnen, pp. 402—404, or Memories, pp. 298—300. Westermarck's colleague L. T. Hobhouse also offered a defense of the method; see Hobhouse, "Comparative Ethics," in his Sociology and Philosophy: A Centenary Collection of Essays and Articles (Cambridge, Mass., 1967) pp. 237—267.
102. T. D. Campbell, Adam Smith's Science of Morals (London, 1971), p. 238.
103. Westermarck, History, p. 1.    "'
104. Ibid., p. 21.
105. W. H. R. Rivers, "Survival in Sociology," The Sociological Review 6: 4 , (1913), p. 304.
106. Claude Levi-Strauss,  "L'œuvre d'Edward Westermarck," Revue de l'Histoire des Religions 129:1&2—3 (1945), p. 87.
107. Westermarck, History, p. 10.
108. For example, see John Dewey, "Anthropology and Ethics," in The Social Sciences and Their Interrelations, ed. W. F. Ogburn and A. Goldenweiser (Boston, 1927), p. 27; Lowie, op. cit, pp. 97—99; Ragnar Numelin, "Edward Westermarck and the Finnish Sociological School," Le Nord 4 (1941), p. 273; Mitchell, op. cit., p. 68. The critics themselves are often none too masterly with details. Thus Mills, op. cit., p. 655, claims that Westermarck "conceived of the marriage study around 1887 .... Fourteen years later the book appeared." [The History of Human Marriage appeared in 1891.] And Mitchell, op. cit., p. 66, asserts that Westermarck "spent part of the year teaching philosophy at the Academy of Åbo in Helsinki…" [The Åbo Akademi is in Åbo.] Still, these errors are mild compared with this description of Westermarck, a childless bachelor: "His own marriage was happy, though his wife's name does not appear in the index of his book; a son, Jack, was killed under the wheels of an English lorry." Stanley J. Kunitz and Howard Haycraft, eds., Twentieth Century Authors: A Biographical Dictionary of Modern Literature (New York, 1942), p. 1502.
109 For example, see Westermarck, Minnen, pp. 83—84, 96—97, 113—114, 150, or Memories, pp. 79—80, 89, 101—102, 126.
110. Westermarck, The History of Human Marriage (London, 1891), p. 4; see also ODMI I, p. 2.
111. See Westermarck, Marriage Ceremonies in Morocco (London, 1914); Ritual and Belief in Morocco, 2 vols. (London, 1926); and Wit and Wisdom in Morocco: A Study of Native Proverbs, with the assistance of Shereef 'Abd-es-salam el-Baqqali (London, 1930).
112. Westermarck, Minnen, p. 181, or Memories, p. 146.
113. Curiously, it was Malinowski, a pupil of Westermarck's, who propounded one of the most explicit views on the subject;  see Malinowski, "Culture," in Encyclopedia of the Social Sciences, ed. E. R. A.  Seligman and A. Johnson, vol. 4 (New York, 1931), pp. 621—646. Before the article appeared,  Westermarck  wrote  to  Malinowski,  "I  was  very glad  to  hear that you are going to write the Article on Anthropology for the Encyclopaedia, since nobody could do it better."  Westermarck  to  Malinowski, 26 August 1926, Bronislaw Malinowski Papers, Yale University, New Haven, Conn.
114. Westermarck makes this point about Durkheim in "Methods," p. 240.
115. Westermarck, History, 5th ed., p. 17. In his spoken introduction to a  lecture of   Marcel Mauss, Westermarck commented,  "I think there should be — well I think there is — an entente cordiale between the two schools, the English and the French." EWB, Box 42. [Spelling errors have been corrected in this quotation.]
116. Westermarck, History, 5th ed., p. 17.
117. Ibid., pp. 17—18; see also Westermarck, "Prefatory Note," in Gerald C. Wheeler, The Tribe, and Intertribal Relations in Australia (London, 1910), PP. v-vi. After all, as Bryan Wilson has put it, in Magic and the Millenium: A Sociological Study of Religious Movements of Protest among Tribal and Third-World Peoples (New York, 1973), p. 3, "The end sought is not the subsumption and encapsulation of all reality in a set of (therefore necessarily tautological) formulae, but the interpretation of reality by principles of greater generality than are to be derived from a particular case, and by the conscious application of comparative method."
118. Fletcher, op. cit, pp. 97, 116—117.
119. "Behavior" should be used with some caution, however, because Westermarck later contrasted the term with "opinion," concluding that his inquiries dealt with the latter (ER, p. xvii); but "behavior" has been used throughout the present article in its broadest sense to include "opinion."
120. ER, p. xviii.
121. J. Ellis McTaggart, untitled review, International Journal of Ethics 20: 1 (1909), p. 94. Westermarck made a marginal notation at this passage in his copy of the journal.
122. Westermarck stated, "Facts in themselves leave me as a rule rather cold; but they become a different matter, component parts, indeed,  of a person's mentality, as soon as he thinks that he can abstract from them something which they do not directly express." Minnen, p. 26, or Memories, p. 28.
123 ER, p. 24.
124 Westermarck, Minnen, p. 296—297, or Memories, p. 232.     
125 A. R. Radcliffe-Brown, "The Comparative Method in Social Anthropology", The Journal of the Royal Anthropological Institute 81:1—2 (1951), p. 22. See also E. E. Evans-Pritchard, "The Comparative Method in Social Anthropology," L. T. Hobhouse Memorial Trust Lecture 33 (London, 1963).
126 ER, p. 177.