Home | Bibliography | Site Map | Page 1 of 2

Wittgenstein's "Remarks on Frazer's Golden Bough"

Preface: the topic of this paper is the Philosophy of Religion, but here the subject of magic is looked at in a very different way. Whether in the context of religion or not, authority is not the issue here, nor is belief (or at least not very much).

Religion does not rest on opinions, and where there are no opinions there can be no error. There is progress in science, but not in magic.

Outline of this page ...

• Outline of page two's topics: Are words such as 'ghost' and 'spirit' superstitious words? Many rituals are not done to effect change, although others are.

Science and Magic

Simple though it may sound, we can express the difference between science and magic if we say that in science there is progress but not in magic. There is nothing in magic to show the direction of any development. (RFGB p. 13)

Nor could there be. If there were development in magic, if it were possible to get a better result by changing the ritual, then it would no longer be magic. Because then there would be an hypothesis that could be put to the test: If I change the ritual in such-and-such way, then I may get a better result. And further, if when I test my hypothesis, I do get a better result, then I can form new hypotheses hoping for still better results .... That is not magic. That is science (of the correlation, although not of the tracing a mechanism, kind). The whole point of magic -- of our concept 'magic' -- is that it is pure ritual: it has no power to change anything. The ritual simply satisfies our longing to have that power.

[In religion there is as well the commandment: you shall not put your God to the test. (And you shall be punished if you do.) This outlaws hypothesis formation. (But 'religion' and 'magic' and 'superstition' are different concepts: we cannot simply substitute one word for the others without loss or change of meaning; we have to look at context.)]

By 'progress', Wittgenstein may mean 'a change in the way a thing (e.g. a ritual) is done that makes it more effective'. For example: If you want it to rain, do this rather than that. [Why shouldn't reciting a formula of words consistently result -- i.e. be followed by -- a rain shower? And even if in our world, ritual does not have such results, nonetheless we can imagine [describe] a world in which it did. In such a world there might be progress in magic: it would be a world where ritual held the place physics holds in our world. But that is not our world, and this page is about our world.]

Text: Ludwig Wittgenstein. "Remarks on Frazer's Golden Bough", ed. Rush Rhees, tr. A.C. Miles, rev. Rhees (Brynmill Pr./Humanities Pr., 1979). Part II, pages 13-18, dates from "not earlier than 1936 and probably after 1948" (p. vi). Part I, pages 1-12, dates from about 1931 (p. v).

The point is that if there were evidence, this would in fact destroy the whole business. (LC p. 56)

Description versus Explanation

The difficulty in philosophy is to say no more than we know. (BB p. 45)

The description [RFGB p. 2] we give is "an account of what we know". An explanation is not among the things that we know. (Otherwise it would just be one more fact among others.) The account of what we know belongs to logic [i.e. what I am calling 'logic' in the context of the "logic of language", which is the point of view from which this page is written]; an explanation might come from any discipline (academic subject), from any point of view [from any discipline's point of view], or just be a matter of common sense [i.e. of what is regarded as sound judgment within a given "community of ideas"]. But with respect to magic [ritual], it was Wittgenstein's view that:

We can only describe and say, human life is like that. (ibid. p. 3)

We can only describe what we know, setting aside any conjectures [theories], and say.... This remark is like: "It is there -- like our life" (OC § 559). [The first remark is from 1931, the second from 1951.]

The question is: would "an explanation of magic" satisfy us better than a description alone does? Is an "explanation" what we want in the case of magic? The double quotes mean: do you think there must be an explanation -- that in all cases it is possible to invent an explanation that will satisfy us? ['possible' not in an idle sense of "Who knows whether someday someone may ...", but in sense of: "Here are the criteria, the guidelines, for an explanation; now invent one."] What do we want an explanation for in this area of our life -- is it not that we feel a disquiet and that we wish to set our minds at rest? Why must that always be possible? [Here it is easy to follow a false path, that is, to picture an explanation on the model of physics where a mechanism is traced. But there is nothing to correspond to that model in the case of magic.]

But will not, Wittgenstein asks, a thorough account of what we know satisfy the desire we have for an explanation? (ibid. p. 2) For example, the confession of sins [which is a ritual or rite -- i.e. magic by any other name].

This can also be "explained" (made clear) and cannot be explained. (ibid. p. 4).

We can describe what we know: we can describe what human beings do and have done, e.g. how they confess their sins. But if someone asks for an explanation of why people confess their sins (meaning, if someone asks for the deeper reason rather than only asking for testimony from those who confess their sins, for the reasons they are inclined to give, which may be things like "I want to feel at peace with life, with God and the world" or "Because I believe that if I don't, I will be cast into the fires of Hell at the Last Judgment, as my church teaches" or "This is the what my elders taught me to do"), what would this explanation look like? Wouldn't it simply be speculative rubbish about "psychological motives", an unverifiable-unfalsifiable way of looking at things? ['Rubbish' is my word, my attitude, not Wittgenstein's.]

Compared with the impression that what is described here makes on us the explanation is too uncertain. (ibid. p. 3).

This applies to both sincere acts of worship [religion] -- and also to fanciful magical rituals (e.g. fire festivals) of primitive human beings. Of course, sometimes we might say: "No explanation is necessary: we too light bonfires at night -- Bonfires at night are exciting."

But, on the other hand, how a bonfire strikes us depends on the context we place it in. Guy Fawkes Day e.g. seen either as innocent fun ("Bonfire Night") or as an act of anti-Catholicism for the sake of promoting British nationalism. Throwing "the Guy" (i.e. an effigy of Guy Fawkes) into the fire in some contexts may horrify us -- that is, horrify us that we human beings do things like this (because of course human beings have been burned at the stake more than once in our history). Compared to that impression, any "explanation" of bonfire lighting is bound to appear more or less arbitrary and unsatisfactory.

"Every explanation is an hypothesis." (ibid. p. 3) Every explanation is consistent with the evidence (or we would not even call it an explanation), but is not proved by the evidence (otherwise we would call it a verifiable statement, or, proposition). "Something not proved by the evidence, although it may be disconfirmed by it" -- This is a definition of the word 'explanation' [a grammatical account of its use, at least of its use on this page; there is not necessarily an essence of explanation].

A religious symbol does not rest on any opinion.
And error belongs only with opinion. (ibid. p. 3)

But without a definition of 'opinion' we can do nothing with this. Does 'opinion' here mean a conjecture about cause and effect? For example, is Frazer's view that magic is a form of [pre-scientific] error, which people are cured of when they discover the error, although an error is difficult for them to discover (ibid. p. 2), an example of what we call an 'opinion'?

Here Frazer could say: "In my opinion ..." or "My theory is that ..." or "The meaning [in the sense of magic's "significance"] of magic is ..." For example, if someone says, the "meaning" of the story of Abraham's sacrifice is that God does not demand human sacrifice, he is stating an opinion. But then someone else can reply: "Is that the only possible explanation?" (LC p. 60-61), meaning that there are other ways to account for the story than the way you suggest (e.g. as a story about "Abraham, our father in faith", i.e. absolute obedience to God). Here 'opinion' does not mean 'testable statement' ['proposition']; it means a 'theory' in the sense of: a statement consistent with all the evidence, but not provable by means of it [although it can be disproved by the evidence. The only standard -- criterion of correctness -- is consistency with the evidence]. A 'theory' is a selection of facts organized by human imagination into "models, maps, pictures".

Maybe the example of Abraham's sacrifice is an extreme case, however, one where there may be some historical justification for offering an opinion. At the other extreme, and in most cases where we classify something as magic, this is no such background: we know no more than that a ritual exists. And I might say that, if that is all an explanation is here [i.e. a conjecture] -- what we mean by the word 'explanation' in the case of magic --, then there is not much point in explanations. "I would prefer the description alone. Then if I want to conjecture about it, I can do that for myself. Just give me the facts; I can think [about it] for myself."

I keep repeating myself on this topic, saying the same thing over and over, trying each time to get a little clearer about it. On the other hand, these remarks do not make Wittgenstein's statement clearer (i.e. we still have not explained what 'opinion' is to mean here):

A religious symbol does not rest on any opinion.
And error belongs only with opinion. (RFGB p. 3)

Family Resemblances among Rituals

The most noticeable thing seems to me not merely the similarities but also the differences throughout all these rites [or, rituals, of the fire festivals of Europe]. It is a wide variety of faces with common features that keeps showing in one place and in another. And one would like to draw lines joining the parts that various faces have in common. (RFGB p. 13)

You could do this, for example, with games. You could draw a chart on a large sheet of paper listing the names of all the games you could think of, and underneath each game's name you would list all the features of the game [and here there are many types of features to list]; and then you would draw lines connecting the held-in-common features, in this way showing what various games have in common, such as equipment (like nets, balls, rackets, cards, boards). You could use different colored inks to connect different common features, and so on. In this way you could show what Wittgenstein meant by 'family resemblances'. Not all games will have the same features in common. Even within families of games, such as board-games, card-games, and ball-games, there will be variations (e.g. not all ball-games use nets). And in some cases, two games will have no feature in common with one another.

You might also do this with a picture book of human faces, drawing lines to connect similarly shaped noses, while at the same time going around [i.e. avoiding] dissimilarly shaped noses. You would also make connections between similar ears, eyes and so on. But here you must not imagine a biological family; rather, you must imagine that you know no more about the ancestry of this collection of faces than you know about the origins of our concept 'game'. Those origins are lost in our natural history; logic knows nothing about them and offers no theory, which in this case could only be an unverifiable picture ("It may have been that way", meaning that your account is consistent with the evidence but not proved by it), which would be of no grammatical interest, because it would "stand in the way of our seeing the use of the word as it is" (Philosophical Investigations § 305), something which is not theoretical, but describable in public rules for using the word.

For "family resemblances" to be a useful metaphor, we must carefully delimit its meaning [application]. This picture (metaphor) is not overly subtle, but it must be applied in the way Wittgenstein intended.

[There was an exhibition (Cambridge University, 2011) of photographs taken by Wittgenstein, "Wittgenstein and Photography", which claimed this as the source of his notion family resemblances. But another claim is that the notion family resemblances comes from Spengler. But I make no hypotheses: a resemblance (whether to photography or to another writer's ideas) is proof of nothing more than itself.]

Our Investigation has Two Aspects

But our investigation of religious rituals does not consist only of an objective description of rituals, such as we might give by drawing attention to family resemblances (the scattered distribution of common features). To give such a description would be the task of logic, which is what is objective in philosophy. [A 'logic of language' in my jargon means: "a method for making distinction between sense and nonsense"; in "Wittgenstein's logic of language" the method is to give a grammatical account [a statement of the public semantic rules for using a sign], or, a description of a language-game: "language and the actions into which it is woven" (PI § 7), the word 'language-game' being used to make a comparison between using language and making moves in a game, although that comparison is not the best method in every region of philosophy].

But then a part of our contemplation would still be lacking, namely what connects this picture with our feelings and thoughts. This part gives the contemplation its depth. (RFGB p. 13)

The description alone, that is to say, would be an inadequate treatment of religion (magic). That is to say that the reader, the one who reads the description, must also bring something to it. This is related to Wittgenstein's view that religion is full of color; it is warm, not cold [without feeling] like logic, like philosophy (discursive reason), like wisdom. Without our reaction to it, the ritual seems as it were dead [colorless, lifeless, just one more fact among countless others]. The description alone would also be inadequate without our reaction to [our judgment about] the type of human beings who take part in the ritual. [ibid. p. 14]

We have the expression "empty ritual". But what makes a ritual deep -- when do we say that a ritual is deep (on the basis of what, with what grammatical justification)? [ibid. p. 15-16] Is it not because of the context that we ourselves place it in; because one might see it in many contexts. [That is to say: In some contexts the ritual will seem to us deep, in others not. So should you not ask: Is it not because of the context that we ourselves bring to our attempt to understand the ritual that the ritual appears deep? (Is there any use for the expression 'is deep' as opposed to 'appears deep'? This is a grammatical question.)] For example, if we judge [feel] that such-and-such feature of life is deep, and then we see people performing a ritual that seems related to this, then we will say that what the people do is deep [is a deep ritual], and not just something empty, [not something we can't see anything in].

How do we verify what is deep and what is not deep in life? Obviously this is not a question of verifying anything: it is, instead, our own response to life -- both to our common life, but also to our individual life. All we have is a conceptualization (i.e. way of looking at the thing): if we see it in such-and-such contexts.... (Is it important here that we tend to have similar reactions to the reactions of the people we live among? -- or that if we do not, then if we are horrified by a ritual we are also horrified by the people we live among. [We could speak of differences in form of life here.] I mean: is it important to logic? Here we can make a comparison to the grammar of 'beauty' -- the description of an entire way of life is needed to understand the role that certain concepts have in our thinking [life]. The concepts 'deep' and 'shallow' are like that.)

Could you say that the quality of depth is something we add or bring to the facts; it is not a fact itself?

What makes human sacrifice [throwing a man into the fire e.g.] something deep and sinister anyway? Is it only the suffering of the victim that impresses us in this way? (ibid. p. 16)

Why should a purely objective (and therefore external to our own feelings) description of any human activity make such an impression on us? What is there in the description itself that is "deep and sinister"? It is rather that "we impute [what is deep and sinister] from an experience in ourselves" (ibid. p. 16) -- i.e. from the insight into the human heart that we have gained through the things we have ourselves seen and felt and read.

For example. A cat may toy with a mouse before killing it, letting it partly escape and then recapturing it, knocking it about. But only a human being can torture, just as only a human being can be sincere (PI II, xi, p. 229a). It is the feelings (or lack of them; the inner life) of the torturer that strike us as "deep and sinister". As to a purely objective description of the human suffering involved: "All manner of diseases bring just as much suffering and do not make this impression [on us]." (RFGB p. 16)

When Abraham undertakes to sacrifice his child, what impression does this make on us? Need the simple story make any impression on us? God commands a father to cut his son's throat. "Well, that is what gods do after all: demand complete obedience. And, anyway, in the end God doesn't make Abraham go through with it." If we read no cruelty into the story [into Abraham's part in it], if we see none of the seven deadly sins, does it impress us as "deep and sinister"? We might even, if that were the way we were inclined to look at religion, offer a "naturalistic explanation": sometimes people report hearing voices in their heads, but we now regard this as a treatable form of mental illness.

But, on the other hand, if we really try to imagine cutting a child's throat, we may be horrified by the story -- be horrified that human beings will do such things in answer to the gods man's imagination has conjured up ... Whether or not a story (ritual, magic) is "deep and sinister" will depend on the context we place it in [the setting we provide].

But not as if the Philosophy of Religion were in a sense not philosophy ("Feeling, intuition, the irrational" where they are out of place). Recognizing that there are many contexts -- and not just one -- in which we might place a ritual (religion, magic) is, we might say, the point of our investigations. [This is like when we are looking for a rule for using a word, a rule that can lead us out of our confusion about the word's meaning (PI § 309), only finally to recognize that there is no such rule, that in our everyday language we do not as it were play "games" (as in language-games) according to strict rules with this word.] And that we cannot understand ritual (religion, magic) if we do not first recognize that there are many points of view that we may not share and do not understand (although we can describe [give an account of the facts]). Human sacrifice e.g. is not something that is easily explained away as "error" (Frazer); -- there may be something deeper here than that.

The fact that for the lots [they cast to see which among them will be thrown into the fire for their Beltane fire festival] they use a cake, has something especially terrible (almost like betrayal through a kiss), and that this does seem especially terrible to us is of central importance in our investigation of practices like these.

If I see such a practice, or hear of it, it is like seeing a man speaking sternly to another because of something quite trivial, and noticing in the tone of his voice and in his face that on occasion this man can be frightening. The impression I get from this may be a very deep and extremely serious one. (RFGB p. 16)

Think of the impression that adult anger may make on a child, and that this impression is something that can be reawakened [re-invoked] in us as well when we are adults if we witness such a scene.

Again, I think these remarks are connected to Wittgenstein's view of religion as colorful (vital and alive). That is, that we cannot understand religion if we regard it cold-bloodedly, with detachment. We must, instead, look for what is human in it, and that means looking for what is human in ourselves as well. ("Only a man [human being] can understand another man, the human form of life.")

So Wittgenstein seems to say again and again: it not the description [alone, not just the description of the ritual] -- but also the response that is evoked by our placing the description in a particular context [which is something that we ourselves bring to the investigation (cf. PI § 107)] -- that makes the investigation of magic or religion [or religious ritual] complete. (Although: is Wittgenstein saying that it should be this way, or simply that it is this way?)

So was I correct in suggesting that Philosophy of Religion can at times only be written in the first person? Is it like this:

I wonder if you can ever speak about religion except in the first person. -- In the sense here that: can you avoid seeing it through your own techniques of understanding? It's not as if I could suspend my critical reason, philosophical skepticism: that would be like saying "Think, but not really", and I don't know how to do that. (At your own level only.)

But in this case, rather, it is that: One can only make a complete investigation of religion at one's own depth, at one's own level of maturity as a human being (cf. CV p. 32). Although anyone might give a purely external description: "They cast lots [and the one who loses is treated as disgraced] and then [they make as if to] throw that man into the fire [and afterwards they talk about the man as if he were dead]." A child might write the story up using exactly those words. [RFGB p. 14]

[Who told human beings that they must placate gods, and who told them that e.g. sacrificing a man would placate the gods? This is lost in our natural history; its origin is unknowable (like the origin of natural language) -- that is, if we can even imagine a sense for the word 'origin' to have here. Why shouldn't it just be "as old as mankind itself"? Sometimes "explanations" just strike us as silly (as do some questions, e.g. "At what point in man's evolution did God inject a soul?") Do you think there must be an explanation for everything? It is of course the scientific spirit [method] to seek one -- but that is not what I am asking ["belief in the causal nexus"].]

What is an "error" in religion (magic)?

This is not how it is in connection with the religious practices of a people; and what we have here is not an error [a mistaken notion, a false science]. (RFGB p. 2)

Was Augustine mistaken, then, when he called on God on every page of the Confessions?

Well -- one might say -- if he was not mistaken, then the Buddhist holy-man, or some other, whose religion expresses quite different notions, surely was.

But none of them was making a mistake except where he was putting forward a theory. (ibid. p. 1)

[Although it seems obvious that in many ways Hindu, Greek, and Jew mean different things by the word 'god' ('God'), there are also reasons why it is natural to use the same word, namely 'god', when talking about them all -- and indeed calling them all 'religion'. (Russell's "Theory of Descriptions" as Wittgenstein applies this, may make the meaning of -- i.e. how we use the word -- 'God' clearer.)]

For there to be an error, there must be some type of verification [of error] possible. If there is no type of verification possible [the only possibility in logic is grammatical possibility: 'not possible' = 'not defined'], then we are dealing with a ritual and with a ritual there is no error. This is related to "there is progress in science but not in magic" (See above). Magic is ritual [a ceremony].

This is why Augustine or the Buddhist holy-man were only making a mistake "if they put forward a theory" (ibid. p. 1). -- They weren't necessarily making a mistake [i.e. error] if they invented a picture (the Last Judgment e.g.), unless the picture expressed an hypothesis (i.e. a statement that might be inconsistent with this-or-that fact). If there was no hypothesis, then there was no error possible. But this is only a definition. And it may not be what Wittgenstein meant -- or all that he meant -- by 'error' in this case. He might [also] have meant: they do not understand the logic of religion [magic] if they try to offer a justification, or if they confuse religion with theory-making -- i.e. try to model religion on science or metaphysics.

But again, our understanding is being blocked by a vague concept-word, namely 'theory' (cf. 'opinion' above). What did Wittgenstein mean by the word 'theory' here?

Religion is not hypotheses [things testable: e.g. conditionals or verifiable propositions], or the making of hypotheses, but the expression of something which is not hypothetical. [Does ritual exist to express something that we feel the need to express? That is, if we look at it from this point of view, does its nature seem clearer to us?] Religion can only err if it attempts to form hypotheses -- or to regard its doctrines as theoretical. (Wittgenstein about Moore: "A thing isn't made clearer by simply saying it over and over again" (Letter to Russell (R.1), 11 June 1912))

For many centuries Catholics have been offering the sacrifice of the Mass (And it is a sacrifice -- that is why the table is called an 'altar' and that is why the celebrant is called a 'priest', and why bread and wine become flesh and blood, etc.). This ritual hasn't changed in centuries: the sacrifice of Jesus Christ is offered daily to God. What is its efficacy? What is it supposed to [visibly, demonstrably] effect? The answer is: nothing testable ['untestable ritual' is a pleonasm]. "If we do not offer the sacrifice, God will not forgive our sins." But there is no way to measure whether or not sins have been forgiven [If there were, this would not be a ritual]. There is nothing testable in magic. That is the difference between magic and science. And because there is no test, no hypothesis [If ..., then ...] and subsequent verification or disconfirmation, there cannot be any progress, progress in the sense of improvement, nor deterioration either, in the Mass. There can be no coming to a better understanding in the sense of inventing something more efficacious, more effective.

What is the sacrifice of the Mass supposed to bring about? Nothing that can be tested. And in this sense, the sacrifice of the Mass is not a mistake (error). If people do this act of "magic", they do not do it as an attempt to bring about any verifiable change. And, according to Wittgenstein and contra Frazer, you cannot understand religion, religious practices, without understanding this: religion is not an error [a false science; it does not have the relationship to science of alchemy to chemistry].

How Drury understood Wittgenstein's Remarks

At one time for a short period Wittgenstein got me to read aloud to him the opening chapters of Frazer's Golden Bough. Frazer thinks he can make clear the origin of the rites and ceremonies he describes by regarding them as primitive and erroneous scientific beliefs.... Now Wittgenstein made it clear to me that on the contrary the people who practiced these rites already possessed a considerable scientific achievement: agriculture, metalworking, building, etc. etc.; and the ceremonies existed alongside these sober techniques. They were not mistaken beliefs that produced the rites.... Thus today ... if we enter a church we take off our hats and speak in a low voice ... [We do this from the need to express] reverence [not because we believe] that to keep one's hat on in church is dangerous!

[The] ceremonies were [and are] a form of language, a form of life. (The Danger of Words (1973), p. x)

To "make clear" in this instance means: to offer a satisfying explanation of why human beings did or do these things. 'Clear' = 'explained' = 'nothing to wonder at'.

I think you could say the people Frazer considered had considerable practical achievement or ability -- not "scientific" achievement however (I am using the word 'science' the way Drury does when he writes about the nature of scientific theories): they had knowledge of how to do things; why they thought they got the results they did is another matter, and not likely to be what we would call 'science' at all: This is what I must do, and this is how I explain why I must do it, where the second half is not a practical question but a myth. (The peasant who puts her dough in a warm, draft-free place to rise, also performs a ceremony when she places it there: "If I do not make the Sign of the Cross over the dough, it will not rise." That is practical knowledge [Knowing the physical conditions dough needs in order to rise], but it is not science [Making the Sign of the Cross is an explanation of the dough's rising belonging to magic, i.e. mythology, not science].)

I doubt that any single-cause, or, reductionist theory [explanation] of primitive religious rites and ceremonies is going to be acceptable in all cases. Perhaps in some cases you will be inclined to agree with Frazer, in other cases clearly not. (Just as you might say that some dreams express wishes, others anxiety e.g., not always the same thing in all cases.) And it is a question of inclination, of being satisfied to look at the thing this or that way.

"Conceptions of God"

Why shouldn't some rituals be done for their own sake, not as a means to any end? Some rituals may point beyond themselves (in the sense of having the intention of bringing about change, e.g. the forgiveness of sins), but not all must. When Augustine called out to God in the Confessions, was this not often done for its own sake, in order to express wonder e.g.? Need it always have been a cry for help ("How long, Lord?")? And neither Augustine nor the Buddhist are mistaken, even though they have different conceptions [pictures] of the "really real". There is no mistake here because there is only a conception.

If when I visit the cemetery I take flowers to the grave, if I decorate an altar with flowers, if I throw rose petals at the statue of a saint, if I give someone I love a bouquet, I may believe that I will receive something in return for doing this (Perhaps I am petitioning Heaven e.g.), but it may also be nothing more than this: that, Well, we just do things like that when we love someone. Human beings are not only manipulative; we are also expressive. There is no explanation for this: it is just the way we are; it is a given, a form of life. (Of course, that is not the only way this might be looked at: there is no necessary end where theory-making must stop.)

Religion is better seen as a "conception", the as it were birth in the mind of a way of looking at things, rather than as an hypothesis (that is, a testable statement of fact). This is not to say that religion is without content: No, the practicer or believer holds fast to his conception, to his pictures -- but the pictures of religion (e.g. the Last Judgment) are not used the same way as picture-hypotheses: only in the latter case is there an attempt [a defined way] to compare the picture with what it is claimed to be a picture of [and in this way to confirm or disconfirm it]. If anything were hypothetical, it wouldn't be religion (what we call [how we define by means of examples] 'religion', what we call 'magic').

"Doctor Fischer of Geneva"

In Graham Greene's book Doctor Fischer of Geneva, Doctor Fischer, a very embittered man, plays god, according to his conception of God (i.e. his concept 'God' [rules for using that word], the picture he has created for himself, or which he has been drawn to), his conception [picture] of God being of an all-powerful being who gives us little prizes [presents, gifts] in exchange for our suffering deep humiliations. -- Here you cannot say there is an hypothesis about God; instead there is a conception. And it is a different conception from both Augustine's and the Buddhist holy-man's (or if not of divinity, then of the "meaning of reality"). But it is not a mistake. It is not, because it is not hypothetical [does not offer an hypothesis]. It is an explanation [but only in the sense that an unverifiable picture may serve as an explanation], but only in the sense that a description (or characterization) is [can serve as] an explanation. It is not a hypothetical explanation: it is not something you can disconfirm (and therefore not something that you can confirm either). It is akin to a classification scheme, a system of organization, something which can be done in many possible ways.

It is a description, a picture, which you may either accept or not accept. You may accept it if you share the perception, or you may say: "No, I cannot accept this; it is very wrong-headed; I don't think this should be looked at that way." But you cannot say: "Oh, but this is an hypothesis (in the sense of a proposition, statement of fact) which can be disconfirmed, or which does not stand up to the test when it is compared with the evidence." (Or even if it does appear not to stand up to the test or even make sense (is not consistent), that is not what matters here. The religious person ignores this, simply doesn't form that proposition at all -- or calls it a "mystery [of faith]".)

Ritual as a way of looking at things

So you could say that religion, ritual -- or at least some rituals -- is a way of looking at things. And that a ritual may or may not be a self-conscious act. If it is a self-conscious act, then you can say that it is a way of looking at things. Although not all rituals are self-conscious; people may do them without thinking about their "meaning", about how these rituals should be conceived or looked at or understood. [For example, Italian people and their descendants eat only fish and spaghetti on Christmas Eve: why do we not eat meat on that day? None of us has any idea, although we would not consider doing otherwise. Why? There is no "why"; -- do you think there must be a "why"?] What is the meaning of the first verse of the Hail Mary ["Hail Mary, full of grace, the Lord is with thee. Blessed art thou amongst women"]? It is simply the words of the Annunciation in the Gospel [Luke 1.28)] -- but is that the meaning of those words when they are repeated in the prayer? I doubt that many Catholics would have an answer to that question: having a "meaning" [e.g. being a move in the language-game of 'annunciation' (a greeting followed by a delivery of news)] just isn't the place this prayer has in their religion.

So religion should [at least sometimes] be compared to a way of looking at life -- of e.g. referring things to a non-hypothetical picture that makes sense of life for the believer; -- and this is Wittgenstein's view of religious belief: as a "passionate commitment to a frame of reference" (CV p. 64).

Confusion Words

An unsuitable type of expression is a sure means of remaining in a state of confusion. It as it were bars the way out. (PI § 339)

The word 'conception' may encourage confusion: because it appears to suggest a percept giving birth to a concept -- i.e. a single percept giving birth to a multitude of possible concepts. Now, I can look at something in a variety of ways -- but only if there is already something to look at. The word 'conception' or 'to conceive' suggests that there is already something to look at. For example, if we speak of "the Greek conception of perfection" [which was: what is permanent, not subject to change], this sounds as if there were a pre-existent something [invisible, of course] named 'perfection' which might be conceived in a variety of ways [Cf. "the Christian conception of God", "the Hindu conception of God", etc.] -- rather than a word, a sign (sound, ink-mark), to be assigned a use in our language [i.e. defined, given a grammar]. -- This is the philosophical fog, the misunderstood logic of our language, that Wittgenstein was trying to dissolve.

The return of the fog: the return of the confusion between a verbal definition (conceptual investigation) and a "real" definition (a factual investigation) (Zettel § 458). We must always make the distinction between a verbal and "real" definition with the greatest care in all our philosophical thinking.

Another unsuitable form of expression is 'what we call', because it encourages that mix-up [confusion]: because it may be used to indicate either a verbal definition (i.e. rule of grammar) or a "real" definition (i.e. proposition, statement of fact). For example, 'This is called magic'. We may mean by that combination of words: Such-and-such phenomenon is classified as magic -- i.e. is placed in the class named by the word 'magic' -- because it meets the criteria we have established [defined] for inclusion in this category; in which case, that combination of words is a statement of fact about the phenomenon: it has been verified that the phenomenon has such-and-such characteristics. But, on the other hand, by 'This is called magic' we may mean: this phenomenon is a model [a paradigm: defining model] of what we call 'magic': you are also to call other things that resemble this phenomenon 'magic'; -- this is a verbal definition (or if it is a statement of fact, it is a fact about the English language [such as may be reported in a dictionary], not as it were a fact about magic). And it is not always clear to us which of the two types of definition ("real" or verbal) we are trying to state -- unless we notice just that about our utterance and try to restate it in other words.

'Put forward a theory'

To 'put forward a theory' in the context of religion means to offer something that looks like [has an apparent resemblance to] an hypothesis, something that may have the form of a testable explanation [something verifiable or disconfirmable] -- but which does not have its content [because it is not testable]. Is this what Wittgenstein meant?

But in religion -- or, as I should say, divine theology -- do we find "metaphysical theories" or "pseudo-scientific theories" [i.e. statements that are defined in such a way as to make them unfalsifiable (Z § 259) -- i.e. tautologies]? What is the distinction here? Metaphysical theories are refuted by "arguments" -- that is, they are language, which when examined, shows itself to be a confused mix of "real" and verbal definitions, and refutation is accomplished by showing this to be the case. Whereas a scientific theory is refuted by a confrontation with evidence [the facts of the case] that is inconsistent with the theory, in the case of pseudo-scientific theories it is grammatically impossible to show inconsistency [They are "anomaly-proof" (cf. 'bomb-proof', 'weather-proof') -- i.e. they are consistent with any an all evidence (facts)]; that is what we mean by calling -- i.e. labeling -- them 'pseudo-'. [This is a definition of 'pseudo-scientific theory'.] Both types of "theories" are failures to understand the logic of our language [In the case of a pseudo-scientific theory, it is a failure to distinguish between a frame of reference and an hypothesis; to imagine that you are saying something about reality when you are only saying something about the way you look at reality (PI § 114)], and therefore either type is an error if put forward by a religious believer (or anyone else, of course).

But what would be an example of "putting forth a theory in religion"? Wittgenstein to Drury: "The symbolisms of Catholicism are wonderful beyond words. But any attempt to make it into a philosophical system is offensive." (Recollections p. 102). "To make it into a philosophical system" would mean -- what? (I don't understand this language, and so I am asking for examples as an explanation of meaning.)

In other words, were Augustine or the Buddhist holy-man to make an error in something, what exactly would they have to do to make that error? What would be an example?

But even if they did make an error, would it not be a philosophical rather than a religious error (which is what an incorrect doctrine [heresy] would be)? This is, their error would not be in their religion. We could say that Augustine's What, then, is time? was a misunderstanding of the logic of our language, but that is a philosophical question, not a religious one. When Augustine asks, "What do I love when I love my God?" and replies with many metaphors ("an embrace for the man within" e.g.), this is religion, not philosophy; and error (except in the sense of inconsistency with Church dogma) is not possible (i.e. is undefined. Likewise 'error' in magic, except in the sense of deviation from prescribed ritual, is undefined).

Philosophical Opinions

If we speak of "philosophical opinions" (Wittgenstein used the expression 'my considered opinions' when characterizing some Cambridge lectures to Drury (Recollections p. 141)), we have in mind cases of consistency with the evidence (as well as self-consistency): falsifiable but not verifiable assertions. However, we must remember that we offer more things [other types of things] in philosophy than opinions ("theories"): e.g. metaphors, definitions and methods are not opinions. A way of looking at things [a selected point of view, frame of reference] is not an opinion: here there is no assertion to falsify or verify. (These are of course all grammatical remarks [definitions] or reminders of what we actually do in philosophy.)


Notes about
Wittgenstein's "Remarks on Frazer's Golden Bough" - Page 2 of 2

Site copyright © September 1998. Send Internet mail to Robert Wesley Angelo. Last revised: 16 May 2006 : 2006-05-16

The URL of this Web page:

Back to top of page

Wittgenstein's Logic of Language | Bibliography | Search | Site Map