Gaita, Coeztee, and the Lives of Animals

This unfortunately will only be skin deep.
Writing about animals takes time and space, neither of which I have in a lowly blog post, so I'll have to make due with - and you'll have to forgive me for - skimming the surface. I want to talk about what it means to live with animals, and to do that I'll be doing a reading of a reading. Raimond Gaita, an Australian expert on Wittgenstein, uses a particular passage from J.M. Coetzee's novel The Lives of Animals to show how they way we live with animals is something deeper, and richer, than we generally understand. But first, I should tell you about Hugo.
Hugo is the peach-faced lovebird my family has had since I was twelve. A replacement, for Sam, the first bird who foolishly flew into a window and broke his neck. Hugo has a different temperament than Sam, though they are brothers, and since he is still alive it seems best to speak of him. What has it meant for me to live, on and off, with this creature, for a decade? Well, not much on the face of it. Nowadays when I come over he swoops down onto my shoulder to nestle into the nape of my neck. We're on better terms than we used to be. There were times when I was younger that I screamed at him for eating one of my books or screeching in my ear as I was trying to concentrate. But if I pause for a moment, and consider the way the house is when he is there, I realize that indeed there is much to plumb in this extra-species relationship.
My deportment around the house is entirely in line with the fact that there is another creature there with me. This is not something conscious. At no point did I decide to behave differently whether the bird is there or not. One might reply to this by saying, "well yes, obviously you'll act different, that sheds no light on anything,"but that would be, I think, to miss something. Yes, I might act a little differently if I walked into my house and the entire inside was completely different (this in fact happened once I had moved out and my brother-in-law moved in), but I suspect that this "acting differently" is of a different kind than the kind of deportment I assume when around Hugo. First of all, the former is something conscious, while the latter is embedded in my being in the world. There is something about the way we live with animals that comes before - and indeed makes up - our rational understanding of them. From here I shall turn to Gaita, and Coetzee.
First, a quote from Elizabeth Costello, the protagonist of three of Coetzee's books, the first being The Lives of Animals. Costello is a animal-rights thinker par excellence, and here she addresses the contention that animals cannot fear death and thus their deaths are lesser events than the death of a human.

Anyone who says that life matters less to animals than it does to us has not held in his hands an animal fighting for its life. The whole of the bring of the animal is thrown into that fight, without reserve. When you say that the fight lacks a dimension of intellectual or imaginative horror, I agree. It is not the mode of being of animals an intellectual horror: their whole being is in the living flesh. (The Lives of Animals, 65)

Gaita sees that within this particular line of argument, Coetzee, "challenges assumptions about the connection between our sense of an animal's body and our unhesitating preparedness to say that animals believe this or know that. He urges us to attend to the role that the living body, the body of flesh and blood, plays in the constitution of our concepts, including our concepts of belief and knowledge. Like Wittgenstein, he seems to believe that we misunderstand the importance of the infinitely subtle inflections and demeanours of the body, the many forms of its expressiveness, if we take them only on as the basis for hypothetical attributions of states of consciousness to animals. Rather (I take him to suggest), they partly determine the meaning that words like "knowledge" and "belief," "hope" and "fear," and so on have in our life with language - language, Coetzee emphasizes, as it is used in disciplined ways in literature, and elicited in imaginative living with animals" (The Philosopher's Dog, 71-72)

I doubt that Coetzee sees himself as being particularly Wittgensteinian. Considering that many, if not all, of his works exemplify aspects of deconstructionism, it is a safer bet that his position on the body comes from people like Lacan, Merleau-Ponty, and, ultimately, Heidegger and Freud. Furthermore, it is unclear whether Costello is speaking for Coetzee here. Considering that he's a strict vegetarian, I'm inclined to side with Gaita, but there's no real proof that Costello is speaking Coetzee's thoughts. But Gaita is right to pick up the body as a central theme in Coetzee's work. The living body, for both animals and humans, is the place of communication and knowledge that comes long before anything like states of consciousness. Gaita's larger point in The Philosopher's Dog is that many of our basic concepts - concepts that we assume have nothing to do with animal life - are at least partly constituted by our lives with animals.
Our lives with animals are made up of bodily interaction, and not merely of a brutish kind. When Hugo nestles onto my shoulder, there is more going on than him landing on something warm. Is this some kind of intellectual interaction? No, but neither is it an example of a rational being interacting with an automaton - and let's be serious, birds are pretty stupid creatures generally. And so when considering how we should treat animals, we should note that we are not wrong to consider their feelings and grant them rights as living things, but secondly that we can go to far into anthropomorphism, ultimately asking meaningless questions about whether dogs or horses possess the concept of death. What does it matter if they have the mental apparatus to comprehend death when they can feel it coming with every inch of their bodies? The former means nothing in the face of the latter.
As I said, I can only skim the surface, but what I hope the quotes above, and my time with Hugo, suggest is that our deportment towards animals is more deeply felt than we give it credit, and more important in constituting how we are in the world - i.e. who we are - than we may understand. Is that why I myself shy away from eating meat? Well, it wasn't at first, but it surely is now.

There are Some Things Words Can’t Describe

The Absence of Tacit Knowledge in Davidson and Kuhn

Thomas Kuhn’s response to Donald Davidson’s On the Very Idea of a Conceptual Scheme in Commensurability, Comparability, Communicability[1] successfully rebuffs Davidson’s criticisms of Kuhn’s work, but it does so on Davidson’s terms. There are further arguments to be made on Kuhn’s behalf that do not so readily accept Davidson’s assumptions. Most notable is the assumption that language should be the representative of a conceptual scheme. This is not to suggest that there may be a better representative, or that a representative should exist at all, yet neither Kuhn nor Davidson note that to speak only of language is to ignore other aspects of a conceptual scheme. Kuhn himself seems to forget some of the fundamental points he made in On the Structure of Scientific Revolutions[2]. Tacit knowledge, Kuhn’s term for understanding a concept in a non-linguistic way, is highlighted in many parts of Structure, especially the postscript. By playing on Davidson’s turf and never mentioning tacit knowledge in CCC, Kuhn does himself a disservice. I will firstly examine what is right about Kuhn’s argument against Davidson, before pointing out more fully what both OVI and CCC appear to have missed. Lastly, I will give a truly Kuhnian response to Davidson, one that combines his points in CCC with a discussion about the fact that Davidson does not seem to understand that talk about conceptual schema cannot be so easily degraded into talk about languages. Tacit knowledge must be recognized, something which Davidson never does, and Kuhn seems to have forgotten how to do.

Davidson’s Criticisms

In CCC, Kuhn notes that there are two major criticisms that Davidson, among others, launches at him. We should explore these criticisms, but first we must understand their context in Davidson’s overall project in OVI. Davidson wonders whether talk about differing conceptual schemes is intelligible. His conclusion is that there is “no intelligible basis on which it can be said that (conceptual) schemes are different…(though) neither can we intelligibly say that they are one” (OVI, 198)[3]. Integral to this conclusion is the argument that conceptual schemes, like languages, are translatable, and therefore share a common framework which one employs when moving from one scheme to another. In the end, Davidson believes, “we must say much the same thing about differences in conceptual scheme as we say about differences in belief: we improve the clarity and bite of declarations of difference, whether of scheme or opinion, by enlarging the basis of shared (translatable) language or of shared opinion. Indeed, no clear line between the cases can be made out” (OVI, 197).

So thinkers like Kuhn, who suggest that two conceptual schemes can be incommensurable (which in Davidson’s view means “not intertranslatable”), are talking rubbish. Just as languages are intertranslatable, so are conceptual schemes. Consequently, talking about opposing schemes does nothing for us, at least nothing that talk of differing languages or beliefs cannot handle on its own. Davidson points this out specifically when considering what it would mean for a new conceptual scheme to appear. “We get a new scheme out of an old scheme when the speakers of a language come to accept as true an important range of sentences they previously took to be false. We must not describe this change simply as a matter of coming to view old falsehoods as truth…a change has come over the meaning of the sentence because it now belongs to a new language” (OVI, 188). Changes of scheme rely on changes in language, and so “that truth is relative to a conceptual scheme – has not so far been shown to be anything more than the pedestrian and familiar fact that the truth of a sentence is relative to…the language to which it belongs. Instead of living in different worlds, Kuhn’s scientists may, like those who need Webster’s dictionary, be only words apart” (OVI, 189).

For Davidson, the whole idea of translatability between conceptual schemes is merely a translation of language, and consequently the very idea of a conceptual scheme does not give us any new perspective or information. All of Kuhn’s talk in Structure and later texts is merely talk of language, and the idea of two languages being incommensurable is necessarily unintelligible. Davidson’s proof of this lies in his two major criticisms of Kuhn.

Kuhn distills Davidson’s comments against him in CCC.

Most or all of the discussions of incommensurability have depended on the… assumption that, if two theories are incommensurable, they must be stated in mutually untranslatable languages. If that is so, a first line of criticism runs, if there is no way in which the two can be stated in a single language, then they cannot be compared, and no arguments from evidence can be relevant to choice between them. Talk about differences and comparisons presupposes that some ground is shared, and that is what proponents of incommesurability, who often talk about comparisons, have seemed to deny. At these points their talk is necessarily incoherent. A second line of criticism cuts at least as deep.  People like Kuhn, it is said, tell us that it is impossible to translate old theories into a modern language. Then they proceed to do exactly that, reconstructing Aristotle’s or Newton’s or Lavoisier’s or Maxwell’s theory without departing from the language they and we speak every day.  What can they mean, under these circumstances, when they speak about incommesurability? (CCC, 34-35)

By showing that the idea of non-intertranslatable languages is unintelligible, Davidson believes that we can safely say that Kuhn’s idea of incommensurability is incoherent. This conclusion helps prop up Davidson’s claim that any talk of differing conceptual schemes is a waste of time.


Kuhn’s Response:
Kuhn’s aim in CCC is to adequately answer these two criticisms, and hopefully prove that Davidson has misunderstood the claim that two languages can be incommensurable. His response relies on what he takes to be two mistaken assumptions Davidson makes in OVI, each of which corresponds to one of the major criticisms. The first assumption is that incommensurable should be equated with “not translatable”, or “not comparable”. The second assumption is that translation and interpretation are the same thing. The first is a misunderstanding of Kuhn’s work, while the second is a misunderstanding of how we communicate between languages.
“The claim that two theories are incommesurable” Kuhn explains, “is then the claim that there is no language, neutral or otherwise, into which both theories, conceived as sets of sentences, can be translated without residue or loss” (CCC, 36). Nowhere here do we have a statement concerning comparability. Also, in terms of translatability, this claim does not suggest that it is impossible, only that it operates at a loss. Davidson’s understanding of the term incommensurable is incorrect. The first major criticism, that proponents of incommensurability continually compare languages, leading to incoherence in their theory, seems almost put to rest.
But this does not dispel Davidson’s point that the idea of incommensurability is unintelligible. To do that, Kuhn must explain what “operating at a loss” means in terms of translation. His explanation, which appears after he addresses the second mistaken assumption in CCC, is better understood in light of his discussion on the difference between interpretation and translation. Following Kuhn’s lead, let us return to exactly what is meant by “at a loss” once we have briefly examined his response to the second assumption.
In response to Davidson’s second criticism, Kuhn argues that once again Davidson, among others, has made a faulty assumption.
All three (Davidson, Kitcher and Putnam)[4] sketch the technique of interpretation, all describe its outcome as a translation or a translation schema, and all conclude that its success is incompatible with even local incommensurability…The argument or argument sketch I have just supplied depends critically upon the equation of interpretation with translation….I believe it is wrong and that the mistake is important…The confusion is easy because actual translation often or perhaps always involves at least a small interpretive component. But in that case actual translation must be seen to involve two distinguishable processes. (CCC, 37)
So how are these processes, interpretation and translation, distinguishable? In translating, “the translator systematically substitutes words or strings of words in the other language for words or strings of words in the text in such as way as to produce an equivalent text in the other language” (CCC, 38). Translation is the mere switching of symbols. The translator has done nothing but exchange one word or group of words that represents a concept for another (such as moving from “blanc” in French to ‘white” in English). No substantive work has been done beyond this switching. “The fact of translation has not…change[d] the meanings of words or phrases….The translation consists exclusively of words and phrases that replace (not necessarily one-to-one) words and phrases in the original” (CCC, 38). To call it merely symbol switching may be a bit misleading, but it is essentially what Kuhn’s depiction of translation is. When I translate “la neige est blanche” to “the snow is white,” I have done nothing except exchange some words for others. There is no loss of meaning, as Kuhn notes can happen in “actual translation,” and each of the words in French has the same meaning as the corresponding words in English. I do not need to give a gloss on the meaning of neige, it simply means snow. No extra work in moving between the French and English phrases is required.
But of course, this is not always the case. There are words in French that do not have equals in English, and where symbol swapping is not possible. What then is the translator to do? This is where interpretation comes into play. Kuhn describes the historian’s act of interpretation when moving from an older language to his own idiom,

Most of the words in the older language are identical in form and function with words in the language of the historian and the historian’s audience. But others are new and must be learned or relearned. These are untranslatable terms for which the historian or some predecessor has had to discover or invent meanings in order to render intelligible the texts on which he works. Interpretation is the process by which the use of those terms is discovered… (CCC, 45)
Interpretation is the invention or discovery of meaning. Unlike translation, where words and substituted for other words in a general manner, the process of interpretation involves a specific understanding of certain concepts and contexts in the language we are translating from and a similar understanding in the language we are translating to[5]. Kuhn uses the word ‘doux’/’douce’ to underline his point. The word in French can mean ‘sweet’, ‘soft’, ‘bland’, ‘tender’, or ‘gentle’ depending on the context of the French sentence. A translator would not be able to give an English equivalent word, for there is none that carries all these meanings. Instead, we have equivalent English words depending on the context in which we find ‘doux’/’douce’, like ‘soft’ when we are speaking about wool. To actively render the meaning of ‘doux’/’douce’ into the English language (if that is possible), one has to understand the context in which the word is used, and the underlying concept that the word represents in this context.
This is not the same as an act of translation, for the interpreter must make judgments about the context and what the word means in this context. It is harder to see in an example such as ‘doux’/’douce’, since it is well known that it means soft in the context of wool. But on many (if not most) occasions, it is not so easy to identify how the usage of the word changes in context, and it is up to the historian or interpreter to “discover or invent” a reasonable guess about the usage. Interpretation involves an attempt to figure out the contextual usage of ambiguous words. That attempt involves understanding “the interrelated words in some local part of the web of language” (CCC, 44), getting a grasp on words connected to the word in question. With interpretation, one cannot focus on one word at a time. One must see how one ambiguous term relates to the terms around it.
This is Kuhn’s answer to how he manages to explain older scientific theories in a modern idiom, and it also sheds light on how things might be translated “at a loss.”  When historians of science enlighten us about earlier conceptions of physics or biology, they give an interpretation, teaching us to see how certain words were once connected. How the words fit into “the web of language” has changed, and so we need an understanding of how each word works in relation to others in the previous language. Kuhn does precisely that in Structure, explaining that words such as “principle” were attached to difference concepts and used in different contexts than they are now[6]. Kuhn defends himself from Davidson’s second criticism by illustrating that his explanations of older sciences are not merely translations. He cannot merely swap older words for ones in our idiom, he has to figure out the relationship between words in the old language, and explain that relationship to us in the new.
From this we can get a better understanding of what Kuhn means by “at a loss.” The word ‘doux’/’douce’ is once again a good example. ‘Doux’/’douce’ cannot be simply translated, but must be interpreted for its meaning to be transferred from one language to another. Thus it is an example of

…terms that can be translated only by part and by compromise. The translator’s choice of a particular English word or phrase for one of them is ipso facto the choice of some aspects of the intension of the French term at the expense of others. Simultaneously it introduces intensional associations characteristic of English by foreign to the work being translated. (CCC, 48)

‘Doux’/’Douce’ is translated at a loss, because when choosing a certain word to represent it in the English translation, certain aspects of the word disappear. For example, if I choose to translate it into ‘soft’, my reader will never have the knowledge that the word used also has a connotation to ‘sweet’[7]. There are plenty of cases when the multiple meanings of a given word are important, as in wordplay or irony, and to choose a single word to represent multiple meanings is to lose something of what the word is in its original language. This is Kuhn’s sense of incommensurability. The only true way to understand ‘doux’/’douce’ is to understand the concepts it represents and the contexts in which it is used. That requires interpretation, not merely symbol swapping.[8] And in turn, interpretation is required because the word in French does not have the same connotations as any word in English, and any word in English we would use as a representative has its own connotations, its own ‘intensional associations.’ ‘Doux’/’Douce’ is an incommensurability between English and French, because it cannot fit into the English web of language.
Kuhn gives us a depiction of two different types of movement between languages, interpretation and translation, and it is on the former that his arguments rely. Davidson has neglected to see that we do not merely swap words when moving from one language to another, and that the “discovery or invention” of the meaning of words must come into play. Incommensurability has a place in talk of languages, for it does not mean that they cannot be compared, but only that one must be interpreted for the other.
Tacit Knowledge
Unfortunately, Kuhn’s response in CCC is flawed as well: though his points are correct, his scope is too narrow. By keeping with Davidson’s analogy between conceptual schemes and languages, he betrays one of the most compelling points in Structure. His arguments for Davidson’s misunderstanding of incommensurability are sound, and there is no need to explore them further, save for when they must be integrated in a more complete, Kuhnian response to Davidson (to be discussed later). Kuhn himself seems to forget that language is not the be all end all of a conceptual scheme, at least not his understanding of it in Structure. There is a further, and equally powerful, attack to be made on Davidson, from Kuhn’s own work, that should be explored.
When Davidson claims “We may accept the doctrine that associates having a language with having a conceptual scheme…(since) where conceptual schemes differ so do languages” (OVI, 184) why is Kuhn so ready to assent? We can challenge this claim, using Kuhn’s own appraisal of conceptual schemes. The trap that Kuhn has fallen into is that Davidson’s connection between languages and conceptual schemes is correct, at least generally. We do not want to push the (absurd) idea that language and concepts are not heavily entwined, so we must grant Davidson this connection. But we can also point out that something is missing.
Kuhn uses Newton’s Second Law of Motion, f = ma, in the Postscipt of Structure to give us an idea of what that something is.
The sociologist, say, or the linguist who discovers that the corresponding expression (f = ma) is unproblematically uttered and received by the members of a given community will not, without much additional investigation, have learned a great deal about what either the expression or the terms in it mean, about how the scientists of the community attach the expression to nature…How have they learned, faced with a given experimental situation, to pick out the relevant forces, masses, and accelerations? (Structure, 188)

Already hinted at here is the idea that although we can read the statement, f = ma, and give a language definition for it, our understanding of it is somehow lacking when placed beside the understanding of a scientist.  Scientists can “pick out the relevant forces, masses, and accelerations” where we generally cannot. The sociologist or linguist would only know the definitions of the words, not how to use the concepts behind those words in practical situations. Kuhn explains further,
…While learning to identify forces, masses and accelerations in a variety of physical situations not previously encountered, the student has also learned to design the appropriate version of f = ma  through which to interrelate them, often a version for which he has encountered no literal equivalent before[9]. How has he learned to do this? A phenomenon familiar to both students of science and historians of science provides a clue. The former regularly report that they had read through a chapter of their text, understood it perfectly, but nonetheless had difficulty solving a number of the problems at the chapter’s end. Ordinarily, also, those difficulties dissolve in the same way. The student discovers… a way to see the problem as like a problem he has already encountered. (Structure, 189)

Can this new discovery be described in language? Kuhn has given some sort of description of it, and when asked the student could probably describe how a new problem is analogous to an old one, but there is an aspect that has not been covered in either of these cases. What the student has done ultimately lies outside the realm of language[10]. He has seen one problem as like another. Although the textbook has taught him new concepts and how they interrelate, he has trouble solving the practice problems until he begins to have an independent grasp on how these concepts fit into each problem. Giving the definition of a scientific concept is not the same thing as knowing how to use it in a problem, and this is the trouble the sociologist or linguist has. They have the definitions, but do not know how to use the concepts in the practical ways that scientists do. The definitions are the language aspects to the concepts, how they are used in differing situations is another aspect altogether.

The only way to grasp these non-language aspects is to move on from the text of a chapter and do the accompanying problems, slowly learning (usually by analogy) how the concepts are to be employed. “…What results from this process is ‘tacit knowledge’” Kuhn notes, “which is learned by doing science rather than by acquiring rules for doing it” (Structure, 191). This tacit knowledge, knowledge of the non-linguistic aspects of a scientific concept, is something totally ignored in both OVI and CCC.

We can rightly criticize Davidson’s equation of language with conceptual scheme by saying that it seems to do away with tacit knowledge entirely. We can grant that where languages differ, so do conceptual schemes (at least generally), but also note that a conceptual scheme should not be represented merely by the language it is attached to. Look at the example of the sociologist failing to truly understand f = ma. It appears that the sociologist and scientist share linguistic definitions of the concepts of force, mass, and acceleration, but as Kuhn points out, the actual concepts themselves are different to the scientist. This seems to be an example where language remains constant but conceptual schemes differ[11]. The sociologist and the scientist are not “words apart,” but “ways of looking at the world[12]” apart.

Davidson is wrong to dissolve talk of conceptual schemes into merely talk of language, because tacit knowledge does in fact play an important role in the development of and use concepts. We can see this in the example of the student of science. In the Postscript to Structure Kuhn spends a fair amount of time illustrating the idea of tacit knowledge because it plays an important role in how students of science become scientists in their own right, and also sheds some light on how scientists view the world. Furthermore, since Kuhn’s notion of a paradigm is not a theory but a body of work, tacit knowledge also must clearly play a role there as well. One aspect of working under a paradigm is seeing and solving problems in the way of the paradigm, which means “picking out the relevant forces, masses, and accelerations” for example.

This seems a good place to dig our heels in to begin a defense of conceptual schemes, and it is terribly frustrating that Kuhn seems to entirely ignore it. Tacit knowledge is never once referenced in CCC, and Kuhn even proclaims “If I were now rewriting The Structure of Scientific Revolutions, I would emphasize language change more and the normal/revolutionary distinction less” (CCC, 17). I am not arguing that Kuhn’s defense in CCC is a total failure, or that language is unimportant in his clash with Davidson, but language is not the only aspect to a conceptual scheme. To give a truly Kuhnian response is to merge the different aspects together. Just as a conceptual scheme is neither only linguistic or tacit knowledge, but both, and our response should involve both kinds of knowledge.

A Truly Kuhnian Response

Let us offer a quick response to Davidson, one that effectively counters his criticisms as well as notes that he has made a glaring mistake in dissolving talk of conceptual schemes into merely talk of language. Firstly, we can probe the latter, arguing that tacit knowledge serves as a prime example of how talking only of language loses something of the scope of conceptual schemes. We might go on to suggest that there may be conceptual schemes that differ where languages do not, considering the linguist versus the scientist when they use the words force, mass and acceleration[13]. Opening the scope of the conversation to nonlinguistic aspects of conceptual schemes leaves room for many possibilities for exploration.

Secondly, we still must address the criticisms laid against Kuhn. One might ask, “why do this? Isn’t it enough that we have shown that Davidson’s entire argument stands on a flawed assumption?” But we have not quite done that. We have only argued that to talk of language is not enough. Kuhn’s arguments for the incommensurability of languages has not yet been defended, and furthermore, we must not make a mistake that mirror’s Davidson’s, speaking only of tacit knowledge and ignoring language altogether. This is where the arguments in CCC must be presented. Kuhn himself provides a robust defense of his position, so it is in our interest to borrow his when defending ourselves against Davidson.

Lastly, we must plant Kuhn’s talk of language in CCC into the overall talk of conceptual schemes. Kuhn already has done most of the work for us. We may make a brief analogy between his ‘web of language’ and what we might call a ‘web of concepts.’ Incommensurability between conceptual schemes is similar to one in language. Connections that exist between concepts in one web do not exist in another, and the only way to move from one to another is to understand the relations between these anomalous concepts, and when/how they appear contextually, to interpret.

It might appear that I have fallen into Davidson’s trap myself, suggesting an analogy between conceptual schemes and languages. Yet there is no reason to suppose that analogies do not exist. I have only argued that analogies should not imply equation. Words and concepts have similar problems of incommesurability because they both exist within a conceptual scheme. Language is a part of a conceptual scheme, so of course there exist similarities. Not all aspects of a conceptual scheme are those of language, but at least some are.

---

[1] From now on I will refer to On the Very Idea… as OVI and Commensurability… as CCC.

[2] From now on I will refer to it as Structure.

[3] The page numbers correspond to the printing of OVI in Davidson, “Inquiries into Truth and Interpretation.”

[4] Kitcher and Putnam level criticisms against Kuhn that are very similar to Davidson, and in CCC Kuhn addresses these criticisms en masse.

[5] Obviously translation involves knowledge of concepts and contexts as well, but how the relationship between the two is brought forth in the act of interpretation is different than in interpretation. The interpreter need not spend time deciding how the context of a given word should be represented in another language, to do so is by definition to interpret.

[6] Of course, the words might also appear in the same context with a different meaning, but what I have said does not challenge that.

[7] Obviously barring that they one day choose to learn French, or a French speaker explains this relationship to them.

[8] One may also point out that I have not truly done ‘doux’/’douce’ justice. I have listed a number of English equivalents and spoken a little about the context, but a reader who does not speak French could hardly claim to know how to use the word. If I have done any interpretation, it has not been very good.

[9] Italics are mine.

[10] Obviously the problem itself is language-based, etc, but to quibble on that is to miss my point.

[11] This is debatable, one could argue that the sociologist and scientist in fact speak a different language. Even if we grant that, the force of the point is not lost, there is a difference in tacit knowledge between the two men, and therefore the major difference between the two is not really a linguistic one.

[12] Or at least ways of looking at the practical problems of science.

[13] Once again, I am not trying to offer that this argument is correct, but I believe it is an option to be explored.

The Terror of a Hypothetical: Blocking the Intuitions of the Ticking Bomb

Noah Cruickshank 3/8/10

Dismantling the ticking time bomb hypothetical[1] has proven to be an arduous task. In the face of empirical objections to the scenario - such as those found in Bob Brecher’s book Torture and the Ticking Bomb or Darius Rijali’s Torture and Democracy - the proponents of the TTB believe that they need only modify their examples to retain the hypothetical’s force. I shall do my best here to show that even with the strictest modifications to create the most likely scenario, the TTB still gives rise to the kind of empirical objections Brecher and Rijali pose. Thus, even the most airtight case of the TTB will still be riddled with both empirical and practical issues that suggest torture should not be used.

Still, a determined proponent of torture might accept those objections and still be committed to torturing in my particularly defined case of the TTB, arguing that there is still a chance that torture might work. And if that chance still exists it is one that we are morally required to take. I would like to meet that position head on, and argue that even if we accept that we should torture in the deeply modified TTB, this does not commit us to anything in the real world. We can grant Charles Krauthammer that there is a hypothetical where it would seem that torture is required, without granting him that this hypothetical modifies the discussion of torture entirely, so “that the argument is not whether torture is permissible, but when…”[2] Even if the hypothetical is airtight (which, I will show, even in the extreme case it is not), that does not mean that it necessarily has practical application. In the end, I believe that we can be always against torture for all intents and purposes even if we grant the TTB proponent my “nearly airtight” case, since we can deny that the hypothetical has any bearing on our practical positions on torture. To put it bluntly, just because torture might work in an extremely defined hypothetical is a woefully pitiful reason to use torture in real life.

The Ticking Bomb

 I will give my own detailed account of the TTB below, but let me first give a rough overview of how most examples of the TTB work. In a given essay on torture, we are told to imagine a case where a bomb is due to go off sometime in the near future. Someone with information about the bomb’s whereabouts or deactivation code has been taken into custody. After attempted interrogation, the person is unwilling to give up the needed information. If the bomb goes off, there will be a high number of casualties. Proponents of the TTB now want to insist that since torture might be a viable option in forcing the person to give up the information, it is a morally required option.

The account above is incredibly vague, though as it turns out, it is not much worse than most accounts of the TTB in the literature on torture. There are a number of problems that occur in most presentations of the TTB, generally falling into 4 separate types: time constraints; the specific techniques that would be used and the efficacy of those techniques; who is victim of the torture; lastly, who is the torturer. To give brief descriptions of these problems now before looking at them more closely, writers generally do not define time constraints on the event specifically, do not explain what torture could or would be used and what should be expected to work, do not say who actually is being tortured, and do not say who is doing the torturing. While each of these problems bleed into the others, I will discuss each problem specifically and show how my example of the TTB attempts to counteract them specifically.

Firstly, most proponents of the TTB have no idea how long torture usually takes to work. Krauthammer, for example, suggests that the bomb will go off in one hour. This relies on a complete misunderstanding of how torture is actually practiced. Rijali notes, “Physical interrogation methods (of torture), like psychological methods, take time, time that interrogators do not have in emergencies. Real torture…takes days, if not weeks…”[3] In the face of Krauthammer’s one hour time limit, there would be no reason to torture the person withholding information, because there would not be enough time for the torture to be effective. Any example of the TTB that gives us only a few hours to interrogate a suspect already rules out the chance of torture actually working.

A TTB proponent might then expand the time frame, giving the kind of example that Fritz Allhoff does, where there is “a bomb in a crowded office building that will likely explode tomorrow”[4], but this runs into trouble as well. First of all, will the bomb “likely” explode tomorrow, or will it actually explode? But more importantly, if the authorities in this hypothetical city are aware that an office building may explode in a day, that gives them enough time to make sure people who work in office buildings could be alerted and begin a thorough search of office buildings in the area. It seems that the authorities could very easily tell everyone with an office job to take the next day off for their own safety. So expanding the time frame is problematic as well; too little time, and torture is useless, too much time and other means to save lives are available.

In light of this, I will begin to introduce my version of the TTB. Suppose that a four-man cell of Al-Qaeda has planted hydrogen bomb somewhere in New York. Let’s say the men are caught at 12:30 pm on a Monday and it is set to go off at 12:30 pm the next day, so the authorities have a day to interrogate, and possibly torture the subjects. Remember that the blast radius of a regular nuclear device is about two miles, and a hydrogen bomb is substantially more powerful than that. This means that if the bomb goes off, it is likely that not only New York city is leveled, but the greater New York area as well. New York is huge and populous, which means it would be nigh impossible to evacuate in a day, and also nigh impossible to effectively search for the hidden bomb. This does not mean that the authorities should not try to evacuate and search, but it is understood that unless they are granted a miracle, these methods alone will not stop a substantial loss of life. A day is still not a lot of time to torture with full efficacy, but it would be long enough to begin the torture process. It seems, on the face of it, that the time constraints are dealt with as best they can be. In this scenario there is not enough time to effectively use non-interrogational methods, but enough time for torture to maybe start having an effect on the captured terrorists.

But within this time frame, what torture methods are actually viable? Aside form Alan Dershowitz, who suggests that a sterilized needle be placed under the fingernails of the torture victims, most people who discuss the TTB never mention what kinds of torture can or should be used. This vagueness also stems from a lack of understanding of how torture works, and to alleviate this problem I will now suggest what might be the best bets to try in the scenario I am constructing. Rijali notes, “Short time changes a torturer’s preferences. Torturers cannot use techniques that take time, like forced standing and sleep deprivation. They must push to maximal pain fast with techniques like whipping, harsh beating, violent shaking, and electroshock.”[5] With the time frame being 24 hours, it is possible that we might be able to incorporate techniques such as forced standing that place the terrorist in an uncomfortable position, but unless those worked relatively quickly (within twelve hours, say), we would have to move on to the more brutal forms of torture that Rijali lists. We might perhaps throw waterboarding in as well, just as another method at our disposal, but it is unclear that waterboarding would provide the kind of maximal pain needed to perform adequate torture quickly. This problem also seems to apply to Dershowitz suggestion of needles.

This need to perform adequate torture in a relatively short time frame brings out the third problem that usually arises when confronting the TTB: who will be doing the torture? This is a particularly distressing point that most proponents of the TTB seem to overlook. Generally when posing the scenario, the question “Would you do it?” - as in, “Would you torture the terrorist?” – is asked.[6] But from the outset, that is a problematic question, since most people – including writers of the TTB scenarios – have no idea how to torture. If the you in this question is an average person, we can expect the torture to be brutally misapplied and deeply ineffective. In a situation as dire as the TTB, it would be ludicrous to call on a neophyte if torture was the chosen option to extract information. As Brecher puts it, “The ticking bomb scenario requires us not to imagine what we would do, but to imagine what we would require someone else – a professional torturer – to do on our behalf; and…as the practice of their (sic) profession. The institutionalization of the profession of torture is a necessary condition of the example’s even getting off the ground…”[7] The only way that torture might work in a TTB scenario is if the person torturing knows what he is doing – he must be a professional. So to keep my scenario as airtight as possible, I will have to admit to there being professional torturers, and their use in this case. Lucky for us, we know the US has these kind of men working for them, so let us assume that one of them is in New York city with us, ready to torture these four Al-Qaeda members if the authorities say the word. This stipulation already deeply compromises the scenario – since by assenting to torture we would be implicitly assenting to the profession of torturer – but it seems the only real way to “get if off the ground,” as Brecher says.

Lastly, there is the problem of who the victims are. I have already said that in this TTB, our possible victims are a four Al-Qaeda operatives working together. Let us suppose there are no others in their specific group. Let us also suppose that the spy that brought them in is trustworthy, and has provided ample evidence to show that these are the men who planted the bomb. On occasion the TTB examples are so vague it is unclear who is to be tortured, and thus an easy way to reject the hypothetical is that we could be torturing an innocent person. I want to do away with such objections, so let us stipulate that we know for sure – due to the outstanding work of our intelligence agencies, that we have the men. Furthermore, they have been trained in interrogative techniques, and will not speak.[8] Any question posed to them is responded to with silence. Thus their interrogators cannot get any of their techniques going to try to glean some information about the location of the bomb.

Let me recapitulate the scenario for clarity. There is a hydrogen bomb under New York city and we have a day to find and defuse it. This is not enough time for a full evacuation of the city or a comprehensive search for the bomb. Due to some very good intelligence, we have all the members of the terrorist group (Al-Qaeda) who planted the bomb in our custody. The suspects will not talk, thus interrogational tactics are useless. We also have a professional torturer on hand, who would be able to make the best use of the brutal techniques needed for such a small time frame to get results. There is still no guarantee that torture will work to loosen these men’s tongues, but of all the TTB examples I have seen this gives us the best shot of torture being effective. Let us also suppose that this is top secret, and will not be known by the public, thereby quelling objections on the ground that this act of torture would have a devastating social impact. So now the question can be posed: should we tell our professional to go ahead and torture these men? I now want to give an empirical and practical case for why we should say, “No.”

Empirical and Practical Problems with the TTB

Even with the TTB as airtight a case as we can have it, there are still a multitude of problems that arise, all of which suggest that we should not torture. I want to go through the general problems of the TTB again and show that even in my scenario there is still a lot of evidence to show that torture would be ineffective, and furthermore the practical consequences of getting the hypothetical “off the ground” are also deeply worrying.

Let us begin again with the time constraints. While managing to allot twenty-four hours for interrogation does give our professional torturer some time to work effectively, it is still not clear whether this would be enough. Rijali notes that “Hardcore believers, including presumably the common terrorist, do not break quickly.”[9] This usually occurs for a few reasons: firstly, “For decades, guerilla organizations have made ‘torture contracts’ with their members: if you get arrested, keep the interrogators busy for twenty-four hours and let us change the passwords and locations…’”[10] In our case of the TTB, the tortured Al-Qaeda operatives know that if they hold on for a day, the bomb will go off and the torture will end – either because they have been killed in the explosion or there is nothing for him to give up. By stipulating twenty-four hours to give time for the torture to work, we have also given the terrorists a goal – just get through the day. With something like this in their minds, the terrorists will be harder to break. Secondly, the CIA Kubark manual notes, “Persons of considerable moral or intellectual stature often find in pain inflicted by others a confirmation of the belief that they are in the hands of inferiors, and their resolve not to submit is strengthened.”[11] An Al-Qaeda operative will surely view the American torturer as an inferior, and since the operative believes he is on a mission from God, there is a good chance torturing him will only strengthen his resolve to not speak.

Concerning the efficacy of torture tactics in the TTB, there is a direct problem with attempting to achieve “maximal pain,” immediately. Suppose the torturer, through electroshock or harsh beatings, manages to reach the pain threshold for the victims, and yet they still do not break. Many torture victims describe a numbness that sets in once they reach a certain level of pain.[12] If maximal pain is reached, and the victim is not broken, then what is the torturer to do? Even the best method to break the victim in a short amount of time comes with a cost, in that torture will lose any possibility of efficacy if it is not effective from the get go.

Furthermore, if these men are terrorist operatives in the United States, they should be prepared for the possibility of capture. We stipulated that these men were keeping silent, and trained in interrogation tactics as a way to force our authorities into torture as an option. Knowing that, it is ludicrous to believe that they were somehow not prepared for torture as well. Either the men have not been trained to undergo any interrogative techniques, or they have been trained to undergo all kinds of them. In the case of Al-Qaeda, the latter is far more likely. Thus even with a day we probably would not have enough time for the torture to work, since our victims would already be prepared for such measures and could resist them.

Again, if the victims of our intended torture are prepared – and we should expect that they are – it is very hard to see how effective torture would be. Even if we are using techniques such as electroshock that strive for maximal pain immediately, if they are trained to handle great pain it will doubtful that these techniques will work at all. A more likely way to break our victims would be something like threatening to torture or in fact torturing their loved ones – this is how Khalid Sheikh Mohammed was finally broken, for example. But this gets us into dangerous territory, both empirically and practically. Unless these men have friends and relatives living in New York, a day is not a lot of time to find their families and set up an effective scenario where we would be able to – or at least make it appear that we could – torture them. It is not impossible, only very unlikely to work within the confines of a day.[13] Also, this would mean that we would have to accept torturing – or even murdering – innocent people. That sets a dangerous practical precedent, even if the authorities never publicly disclose their actions. Many people who agree to the TTB might be happy to torture the terrorist, but how would they feel about his innocent sister in Jordan? If we want the best torture techniques available to us to defuse the bomb, torturing innocents is right up there.

This of course connects to the question of who is being tortured, and I have little else to say about the subject, expect to reiterate that if one assents that torturing the terrorist is morally required because it might be effective in stopping the bomb, it is hard to see how torturing his family is not required as well, since there is evidence that it too might be effective. If in response the TTB proponent says, “Well, we don’t torture innocents on principle,” it seems reasonable to say “Well, we shouldn’t torture anyone on principle.” If we limit our torturing practices based on principle, why not eliminate them altogether on principle as well?

Lastly, as I noted before, having a professional torturer to do the work for us in this TTB will be the only possible way that the torture might work, but that leaves us open to the practical problems of having the professional torturer around. First of all, to have professional torturers means we would have to train people to become professionals in torture. As Brecher puts it, “To admit the profession of the torturer to the range of legally recognized professions would require that we recognize torture training in the same way that we recognize, for example, legal, medical and teacher education and training….”[14] To have this man at our disposal in the TTB case means we must create a system where that man could learn his trade. Brecher specifically speaks of legalizing the torturer’s profession, but we need not go that far: perhaps we have a secret torture program that is officially illegal, but in place nonetheless. Yet having a secret system to teach torture does not seem much better than having an open, legal system to do it. To assent to the TTB means to assent to having a cadre of trained torturers at our disposal, whom we pay and train.

Brecher points out a further issue concerning the problem of “who does the torturing?” “The necessary role of medical personnel (in torture) is well documented,” thus if torture became a practical possibility, it would make “medical cooperation, advice, and training in interrogational torture on a par with the provision of medical services in prisons.”[15] Not only would we have professional torturers, but professional torture doctors. Brecher and others[16] tease out the many implications of having torture as an institution, but for the sake of space I will not discuss them further. Suffice it to say, the professionalization of torture – the creation of a entire system of practices where torture is taught and refined – means that we are not limiting ourselves to torturing only in the case of the TTB. If we have these professionals, what would stop the authorities from using them in less dire situations, or situations where torture would be more effective?[17] Furthermore, it is hard to see how these torturers could become skilled without practice. It is conceivable that they only train in role-playing exercises – perhaps each torturer in his training must be tortured so that his classmates can have practice, etc. – but even then it seems that we are obliging them to torture innocent people as a part of learning their trade. Either they need to torture possibly guilty people outside of the TTB to learn how to torture, or they torture their innocent companions. This is already problematic without the other repercussions that Brecher and others suggest.

Even with this “mostly airtight” TTB, there is still no guarantee that torture would be effective – indeed, no guarantee that torture might be effective; there also appear to be dangerous practical repercussions that appear from stipulations we must make to have the hypothetical get off the ground in the first place. If it is unclear that torture is even a viable option in the TTB, there seems plenty of reason to answer “no” to the proponent of torture. If it will not work, then why subject the victim, and ourselves, to this degradation? Furthermore, we might be unwilling to create the kind of system needed for torture to even be a viable option in the TTB – having professional torturers and the like. The TTB usually pulls at our intuitions – the feeling that we must do something, anything, to stop the bomb from exploding – but I do not think that having a professional system of torture has much intuitive pull for anyone. Assenting to the TTB is assenting to everything the professional torturer must do to become effective when the TTB case emerges[18] - that probably means a whole lot more torture.

But, a proponent of the TTB could still be adamant that torture is a moral requirement in this case because it might work. No matter how improbable the chance, the proponent might say, there is a chance, and it is one we must take in order to save lives. A proponent of torture in the face of a ticking bomb might very well accept the fact that we would have to have professional torturers, and all that they entail. The proponent might say it is an unfortunate necessity, a lesser evil in the face of this possibility – even though this possibility is beyond remote. Richard Posner seems to be of this opinion, suggesting that we let the government do terrible things in secret for the greater good. When considering my TTB and the problems with it, I believe Posner would still say we should keep torture on the table, and everything that comes along with it, because there is a chance for success. The question now is, what do we say to such a person? How do we respond to the person who accepts the large empirical evidence against torture and the practical implications that professional torture has, yet still thinks that in the case of the TTB torture is morally required because it has a chance of working? I want to take the next section to see how we can grant the proponent of torture my TTB while still keeping our hands clean.

The Failure of the Hypothetical
Even if a person accepts the fact that in this highly specified TTB scenario, he or she would be morally required to condone torture, it is not clear that this gets the proponents of torture anywhere. As noted in the introduction, Charles Krauthammer believes that if we grant him the TTB, it changes the discussion, so “that the argument is not whether torture is permissible, but when…”[19] I do not believe that we need to agree with Krauthammer on this point. Agreeing on a nigh-fantastical hypothetical has no bearing in the real world – it has no bearing on our practical discussions concerning whether we should torture or not.
The proponent of the TTB wants to first place us in a hypothetical where we might agree to torture, and then expand the conditions of the hypothetical, arguing that if we agreed to the TTB we have little ground for denying other, similar cases. So once we’ve agreed in the case of a ticking bomb, the proponent of torture then shifts the discussion to when, as Krauthammer says, arguing that there are similar cases to the TTB, and by the same logic we should agree to those as well. This is brilliant sophistry, but sophistry nonetheless.
The only reason the TTB I posed above made any sense was due to its highly specified constraints: there needed to be an exact window where methods other than torture would not be as effective; there needed to be a number of incredibly improbable occurrences – we knew that we had those responsible for the attack, we knew exactly when the bomb would go off, etc. And even with those specificities, we still had to allow for a professional system of torture, and had to grant that there was seemingly good reason to torture innocents if we agreed that torture was required. The moment any of those specificities are expanded, the hypothetical collapses, and thus the claim that torture is morally required also collapses. Contracting the time frame leaves torture to be ineffective. Expanding the time frame means there are other, better techniques at our disposal. Changing the bomb from something catastrophic like a Hydrogen bomb means that things like evacuation become possible. Even moving out of New York city means that there is a far greater chance to search for the bomb and evacuate the city. The TTB only works – and it hardly works, at that – with these specificities. Taking them away means torture is no longer viable.
So we can respond to Krauthammer by saying, “No, the discussion is still about whether. If there is only one possible occasion[20] where torture might work and so would be morally required, then there is no reason to do it in any other case.” Krauthammer believes that if we give him an inch, he can take an ell, but we need not let him do so. I personally do not believe that we should even believe that torture is morally required in my TTB, but even if it is, the hypothetical has no practical application, because the moment you try to apply the TTB, you have to change its structure, and by doing that you turn it into a case where torture is ineffective in comparison to other methods, and thus can be dropped. This means that we can hold a position that is not absolutist, per se, but for all practical purposes has the same import. If there is only one possibility where torture would make any sense, there seems no reason to prepare for that particular possibility over the innumerable others, most of which are more likely.
The truth is that men like Krauthammer have won a battle in a fantasy world, and they want this victory to have a direct bearing on real life. But consider any other deeply improbable hypothetical – Martians land and in order for them to not make war on us you would have to let them subject you to some sexual degradation. You might be convinced that you were morally required to do it, but that has no bearing on real life. It might seem that my analogy is far fetched – the proponent of torture might say “we know that there are terrorists, we don’t know that there are Martians willing to kill us.” That is very true, but try to guess chances of four men planting a hydrogen bomb under New York City, our authorities finding all of them, knowing we have a day, knowing for sure that they are the ones who did it, and that torture would work on them. That seems just as improbable, or not much less improbable than alien life wanting to exchange sexual degradation for peace.
I have been a bit facetious above, but in the hopes of showing that we need not let a ridiculous hypothetical hold sway over our practical considerations of torture. Let the Charles Krauthammers of the world have their TTB, just because he has managed to show that we are not absolutists does not mean that we should believe that torture is justifiable in the real world. Furthermore, if we spend our time and interests trying to create a world where people do not feel that they should blow each other up, the more we reduce the chance of any ticking bomb. Torture is not a practical solution, and the ticking bomb should not trick us into thinking that it is.

---

[1] For the sake of space, I will generally refer to the scenario as the TTB.

[2] Krauthammer, The Truth About Torture, 309 from Torture: A Collection, Ed. Levinson.

[3] Rijali, Torture and Democracy, 474.

[4] Fritz, On the Permissibility of Torture, International Journal of Applied Philosophy, 17, p.129

[5] Rijali, 474.

[6] Jacobo Timmerman, himself a victim of torture, ends his TTB hypothetical with exactly this question. (From The Torture Debate in America, edited by Karen J. Greenberg, 18)

[7] Brecher, 24.

[8] As I note below, if these men are trained in interrogation techniques they will also be trained to resist torture, leaving us with another problem with the efficacy of torture.

[9] Rjali, 476.

[10] Rijali, 475

[11] As quoted in Rijali, 476

[12] Henri Alleg, for example, says so in his personal account of torture.

[13] Imagine that one of the terrorists has family in Egypt. We would have to get into contact with our men on the ground there (or the Egyptian government), find the family members (which could take a lot of time if they are in hiding), set up an interrogation, and then provide proof to the terrorist that his family is in danger. This would be quite the process, and if he nonetheless stands firm, it would be a lot of work for nothing.

[14] Brecher, 70.

[15] Brecher 70-71.

[16] David Luban being another notable example.

[17] If the torturer had weeks and not a day to break his victim, for example.

[18] If it ever will, considering that it is highly improbable.

[19] Krauthammer, The Truth About Torture, 309 from Torture: A Collection, Ed. Levinson.

[20] And again, this possibility is so small as to be non-existent.