Wittgenstein's Scapegoat

In his essay "Wittgenstein on Language and Ritual"1, Rush Rhees quotes with approval a remark written by Wittgenstein several months before he began writing his comments on Sir James Frazer's The Golden Bough. The remark refers to the ritual of the scapegoat, one element in the elaborate procedures described in the sixteenth chapter of Leviticus. These procedures served two functions; they were necessary for Aaron's safe entry into the Shrine (Lev.16:3) and, more generally, constituted the rite of the Day of Atonement (Lev. 16:29-34). Here is a translation of the verses dealing specifically with the scapegoat:

Aaron shall take the two he-goats and let them stand before the Lord at the entrance of the Tent of Meeting; and he shall place lots upon the two goats, one marked for the Lord and the other marked for Azazel. Aaron shall bring forward the goat designated by lot for the Lord, which he is to offer as a sin offering; while the goat designated by lot for Azazel shall be left standing alive before the Lord, to make expiation with it and to send it off to the wilderness for Azazel...When he has finished purging the Shrine, the Tent of Meeting, and the altar, the live goat shall be brought forward. Aaron shall lay both his hands upon the head of the live goat and confess over it all the iniquities and transgressions of the Israelites, whatever their sins, putting them on the head of the goat; and it shall be sent off to the wilderness through a designated man. Thus the goat shall carry on him all their iniquities to an inaccessible region; and the goat shall be set free in the wilderness. (Lev.16:7-9;20-22)2

And now Wittgenstein's comment:

The scapegoat on which sins are laid and which goes out into the wilderness with them, is a false picture and like all the false pictures of philosophy. Philosophy might be said to purify thought from a misleading mythology.3

Wittgenstein's comment is of singular importance for the exposition and defense of his ideas on religion. His followers have been frequently accused of "Wittgensteinian Fideism", an epithet coined by Kai Neilsen4 to refer to their alleged unwillingness to criticize religion. Here, at last, in Wittgenstein's comment, is an instance in which the criticism of a particular religious practice is undertaken by the master himself. Wittgenstein's disciples are well aware of the comment's apologetic value. In his book Belief, Change and Forms of Life 5 D.Z. Phillips quotes the comment in order to demonstrate that "one cannot ascribe to Wittenstein the view that anything that is called religious or ritualistic is free from confusion." Rhees himself originally introduced the comment in order to dispell the impression that "Wittgenstein was coming forward in defense of the ancient rituals".6

Wittgensteinian criticism of the scapegoat ritual is philosophically problematic. It requires the underlying assumption that we possess an infallible appreciation of ancient Israelite sensibilities. To quote Rhees, "Wittgenstein thought the symbol, in the role that was given it, was badly mistaken."7 In order to make such a judgment, Wittgenstein would have to know a) what account of the expiation of sin would make sense for the ancient Israelite religion and b) how the ancient Israelites were likely to misinterpret the symbolism of the scapegoat. Rhees suggests that the ancient Israelites thought along the following lines:

If the people assembled here do bear the sins of their fathers, and of their brothers now living, then why should the priest not bring in some animal to be made one of them in this sense only - that it bears their sins - and then, after laying his hands on it, send it with their sins away from them into the wilderness?'

Rhees himself admits that "it is hard to see the substitution in the scapegoat that delivers them of their sins. I do not know how they thought of this. Nordoes anyone now." Yet he concludes with the non-sequitor "This does not affect the point Wittgenstein is making."

After freely admitting that he lacks a basic understanding of what he sees as the central point of the scapegoat ritual, Rhees still remains confident enough to lable it as misleading. The conventional dictates of interpretational charity point to a different conclusion. If a particular interpertation of an ancient ritual implies that the ritual was essentially flawed, this should motivate us to reexamine the validity of the interpretation itself. More specifically, we should consider the possibility that the investigator had inadvertently sought in the ritual the expression of ideas native to his own culture yet absent in the culture in which the ritual was actually practiced. In particular, while the scapegoat ritual might seem misleading to someone living in a Christian culture, it may have been completely unproblematic for Jews living before the rise of Christianity. After all, it is meaningless to talk about a "picture" being false outside of any cultural context. It is equally meaningless to talk about a ritual presenting a particular "picture" in an extra-cultural sense. Rhees does demonstrate some concern for the importance of the cultural context of ritual. H e believes that his interpretation of the scapegoat is grounded in an appreciation of tribal notions of "bearing the sins of others", ie. that one may bear the sins committed by the members of one's family, living or dead. Against this I contend that Rhees's attempt to uncover the source of the scapegoat's misleading nature merely underscores his own cultural prejudices. Let us first examine his analysis of the linguistic description of the ritual:

When Wittgenstein calls this rite a misleading picture, he may mean something like this: consider

1)"Children carry the sins of their fathers."

2)"A goat, when consecrated, carries the sins of the people."

In the first sentence "carry" is used in the sense of the whole sentence. In the second sentence "carry" seems to mean what it does in "The goat carries on his back the basket in which we put our fire wood"; and yet it can't mean that.

It is always dangerous to subject a translation to linguistic analysis. In the present instance, the Hebrew verb translated as "carry" in Leviticus simply does not possess exactly the same range of meaning as the English word "carry". Perhaps this point is not sufficiently philosophical; but can Rhees be seriously suggesting that a proper understanding (an understanding so precise that it may serve as the basis for the rejection of the symbolism it means to explain) of the Old Testament may be gained through the analysis of contemporary English usage? Exodus 34:7 has been translated as Extending kindness to the thousandth generation, forgiving iniquity, transgression, and sin. The same verb which was translated as "carrying" in Leviticus is here translated as "forgiving". The philological situation does not allow for easy analogies to English.

Problems of interpretation aside, it is possible to develop a much more charitable interpretation of the scapegoat based on Rhees's own schema. Consider the remark made by Hegel after first seeing Napoleon: “This morning I saw the Emperor - this world-soul - ride through the town."8 Presumably, Napoleon was riding on a horse. The horse bore Napoleon on his back. Hegel's remark might be restated as "This morning I saw a horse bearing a world-soul". Compare the following two sentences:

A) The Emperor bears a world-soul.

B) The horse bears a world-soul.

We may now rewrite Rhees's comment:

In the first sentence "bears" is used in the sense of the whole sentence. In the second sentence "bears" seems to mean what it does in "The horse bears on his back the basket in which we put our fire wood"; and yet it can't mean that.

The point is, given the context of Hegel's statement, that it can mean just that. Napoleon's horse was just a horse; he let his master worry about the consciousness of freedom. Although B) may sound like a joke, it is not hard to imagine a cultural situation in which it could become a perfectly normal and comprehensible utterance, i.e. if Hegel's remark had entered into to everyday speech as a way of referring to Napoleon. I would suggest that Rhees's statement 2) about the scapegoat be understood in the same fashion. The ritual of the scapegoat was in fact symbolic; it symbolized God's forgiveness for the sins of the people, the sins had been symbolically banished to the uninhabited wasteland. However, it was unnecessary for the goat to substitute for anyone or be granted honorary membership in the nation of Israel. No one had to take these sins upon himself; God was going to forgive the sins, make them "go away". The high priest symbolically placed the people's sins on the goat - but this does not mean that the goat was thought to have taken on some kind of spiritual responsibility for them. The goat's role was purely practical; he served as a means of transportation. Perhaps this would be clearer to us if instead of merely uttering a verbal confession the priest had tied a written list of sins to the goat's head. The goat's job would be to physically carry the list far away to the wilderness. Of course there was no written list, but we can hardly fault the ancient Israelites for not introducing yet another material element into a ritual which was meant to express an entirely spiritual concern.

A comparison with a Jewish ritual, known as tashlih, which is still observed today, may make this clearer. On the afternoon of the first day of the Jewish New Year, it is customary to visit a natural body of water, throw crumbs of bread into it, and read certain verses from Scripture, including Micah 7:19:

He will take us back in love; He will cover up our iniquities, You will hurl all their sins into the depths of the sea.

The point of this ceremony is to symbolize the longing for divine forgiveness. It might be said that the bread crumbs represent sins, and that they are borne away by the waters of the river or stream. No one would think that the water was somehow going to suffer for the sins of the participants in the ritual, or that Micah meant to say that the depths of the sea would be spiritually burdened with the sins of men. The point is that the sins are forgiven, cast away to the inaccessible depths from which they can never return.

Even when understood in this way, the scapegoat ritual (or rather the entire ritual procedure of the Day of Atonement of which the scapegoat was merely one element) might still have been misconstrued by some worshippers as mechanically forcing God to forgive their sins. The same might be said of the tashlih ceremony, or for that matter, of the Christian sacraments. In fact, this kind of misunderstanding is endemic to practically all religious ritual and hardly gives cause to isolate the scapegoat for special criticism.

The deeper cultural roots of the misunderstanding of the scapegoat ritual become clear a few lines later in Rhees's article:

Perhaps we'd not find it incongruous - we'd not find the picture jams in in symbolizing what is intended for it - if you said that a man might take on himself the sins the people have had to bear, and to offer himself in atonement for them. But a goat?

Rhees has let the cat out of the bag. The notion of a man taking on the sins of others and offering himself (or should I write Himself) in atonement for them is all too familiar. Rhees has been mislead by the application of a picture - the Christ idea - to a culture in which it simply had no application. I would suggest that the "false picture" which the scapegoat is thought to project was simply not available to those who actually performed and observed the ritual.

By modeling his interpretation of the scapegoat ritual on the crucifixion, Rhees reduces it to a kind of confused religious farce. Of course, a goat does not make for a terribly impressive Christ figure. Worse yet, the ritual does not make for a convincing analogy to the crucifixion. Scripture gives us no reason to believe that the goat suffered at all: and the goat shall be set free in the wilderness, not such a terrible fate for an animal living in a culture with an active sacrificial cult! (Later, Talmudic, sources do state that the goat was killed in the wilderness, but apparently the point of this was to make sure that it did not wander back into inhabited territory)9. But why force an analogy which merely obscures that which it is meant to make clear? Isn't it more reasonable to conclude that the scapegoat is simply not comparable to Christ?

Ever since Peter Winch published his celebrated book The Idea of a Social Science and his subsequent essay, "Understanding a Primitive Society"10, philosophers in the Wittgensteinian camp have attacked the idea that the magic and ritual of primitive societies constitute some failed form of proto-science. They view this idea as the product of the imposition of Western thinking on non-Western societies; perfectly good magic is misinterpreted as hopelessly bad science. By forcing the scapegoat ritual into a Christological mold, Rhees becomes subject to an exactly parallel criticism. Certainly if we are prohibited from reducing magic to bad science, we should avoid reducing biblical Judaism to bad Christianity.

Phillips's treatment of Wittgenstein's comment offers no improvement on Rhees. Before quoting Rhees at length, he offers his own version of the scapegoat ritual:

In Leviticus we are told that on the Day of Atonement, a goat said to be laden with the sins of the people is driven into the wilderness, the abode of Azazel, leader of the evil angels. As the scapegoat is driven into the wilderness, so the sins of the people depart with it to the spirits of darkness to whom they belong.11

Even a cursory comparison of this description with the text of Leviticus (unless by "Leviticus" Phillips means "Leviticus as translated in the Syriac version" rather than the Masoretic Hebrew text, the ancient Septuagint and Vulgate translations, or any of the modern translations) makes it clear that Phillips places the scapegoat in the worst possible light. Scripture makes no mention of mythic elements such as "evil angels" and "spirits of darkness". The meaning of the word "Azazel"(which is not capitalized in the Hebrew, since that language lacks capitalization) is not explained by the text; it may designate a geographical location. Most translations I have seen treat it as such. Phillips's description does reflect, without any of their scholarly caution, the views of some modern Bible researchers. Yet Phillips goes much farther than they do in depicting the ritual in crassly barbaric terms. Father Roland de Vaux, in his classic work, Ancient Israel arrives at an interpretation of the scapegoat similar to that of Phillips; and then he adds:

It is important to remember that the transferring of sins and the expiation which results from it are said to be effective only because the goat is presented before Yaweh (v. 10): Yaweh brought about the transfer, and the expiation.12

Someone who had only read Phillips's description of the ritual might not even be aware that God was thought to have any role in it, or that it was part of the practice of a monotheistic religion. Certainly he gives no indication that the ritual might be intended to express some more subtle sentiment through symbolic means.

After reproducing Rhees's discussion Phillips quotes from The Interpreter's Bible:

Christ, as identified with man in his shame and sin, rejected by men and driven away bearing their sins and done to death for their forgiveness, is symbolically depicted, crudely and inadequately yet really, in the scapegoat.13

Phillips then comments:

Notice that here one has the possibility of criticism within a tradition. The ritual concerning the scapegoat is called crude and inadequate.

While the quotation from The Interpreter's Bible offers a fine example of the Christian reinterpretation of Judaism, it says nothing against the fact that the scapegoat ritual, as an anthropological datum, had absolutely nothing to do with Christ. One might as well criticize the rituals of Africa's Azande tribesmen for so poorly symbolizing the Immaculate Conception! As for Phillips's claim that this constitutes a "criticism within a tradition", one must wonder exactly which tradition he is talking about- that of ancient Judaism, which apparently found no fault with the scapegoat ritual, or that of Christianity, which radically reinterpreted the ritual in terms of a doctrine which had not yet been formulated during the period in which the ritual was actually practiced. It seems a bit messy to offer an example which spans the gap separating two theologically distinct and historically mutually antagonistic religions as an illustration of "criticism within a tradition". I would like to take this opportunity to point out a much "cleaner" example.

The Old Testament itself offers clear evidence of the rejection of a ritual element belonging to the ancient Israelite tradition due to the inherently confusing nature of its symbolism. I refer to the copper serpent from Numbers 21:9:

Moses made a copper serpent and mounted it on a standard; and when anyone was bitten by a serpent, he would look at the copper serpent and recover.

Considering the Israelites' much lamented predilection for the worship of graven images, it is not hard to imagine the "false picture" which Moses’ creation projected to the people. Much later, in the course of carrying out his religious reforms, King Hezekiah destroyed the serpent. The author of Kings apparently approved:

He did what was pleasing to the Lord...He also broke into pieces the copper serpent which Moses had made, for until that time the Israelites had been offering sacrifices to it; it was called Nehushtan. (II Kings 18:3;4)

Given such a well documented case of "criticism within a tradition" why pick on a scapegoat?

© Berel Dov Lerner. This is a slightly corrected version of the article that originally appeared in Philosophical Investigations 17:4:604-12 (1994).

NOTES
1) Pp. 450-484 of Essays on Wittgenstein in Honor of G.W. Wright, vol. 28 of Acta Philosophica Fennica (Amsterdam 1976).
2) All biblical quotations are from the Jewish Publication Society of America's translation (Philadelphia 1962 and 1978).
3) As translated by D.Z. Phillips in his Belief, Change and Forms of Life (Atlantic Highlands, New Jersey 1986) pg.30.
4) See his "Wittgensteinian Fideism", Philosophy Vol. xlii no.161 pp. 191 - 209.
5) Phillips pg. 27.
6) Rhees pg. 459.
7) Rhees pg.460. All further quotations of Rhees are from this and the following page.
8) As quoted in Shlomo Avineri's Hegel's Theory of the Modern State, Cambridge 1972,pg.63)
9) See the Mishna, Tractate Yoma 6:6. The unimportance of the scapegoat's death for the proper completion of the ritual may be inferred from Yoma 6:8, according to which the high priest was informed of the goat's arrival at the wildern
ess, but not of its death.
10) (London 1958) and American Philosophical Quarterly Vol. 1 no. 4 pp. 307-324.
11) Phillips pp. 29-30.
12) Volume II: Religious Institutions (New York 1965) pg. 509. This page also contains a discussion of the problems of identifying "Azazel" in the various versions and translations of the Bible.
13) Phillips pg. 31.