Abstract
Dromm’s Wittgenstein on Rules and Nature is a very welcome contribution to recent debates on Ludwig Wittgenstein’s later philosophical methodology. In contrast to more usual attempts to characterize Wittgenstein’s method generally or overall, Dromm’s book has a specific focus. The question he addresses concerns the role of appeals to facts of nature or natural history in Wittgenstein’s later writings, whereby Dromm concentrates on three types of passages: those that appear to explain (1) the development of natural language, (2) a child’s acquisition of language, and (3) our ability to follow rules. Thus, Dromm’s book might perhaps be described as discussing one of Wittgenstein’s many methods (cf. Philosophical Investigations [PI]; Wittgenstein 1953, §133), insofar as such appeals to facts do indeed constitute one unitary method or clarificatory technique (see below). But however that may be, a problem arises with regard to such appeals that has both an interpretational and a more broadly philosophical face, as Dromm explains. Prima facie, the appeals look as if they were meant to offer scientific-style explanations of the kind that Wittgenstein rejects as not belonging to philosophy in his methodological remarks. An example would be the assertion that the possibility of rule following can be explained (and skepticism about it can be avoided) by pointing out that there is an underlying uniformity in reactions that limits the possible interpretations of a rule or that rule following as a matter of fact is a practice. Another example would be the assertion that first-person pain expressions actually are extensions of primitive pain behavior, such as crying and moaning. (Explanations of this type are attributed to Wittgenstein in connection with the issue of rule following, for instance, by Kripke and Meredith Williams.) But if Wittgenstein rejects and still uses such explanations, there is a contradiction right at the heart of his work. The philosophical problem with this position is that, essentially, such assertions constitute hypotheses or speculations for which Wittgenstein offers no empirical justification and whose justification therefore remains unclear. Consequently, his philosophy would seem to be reduced to speculative armchair anthropology and/or psychology with little or no philosophical strength. What Dromm then seeks to do in response to these problems is to offer an alternative interpretation of Wittgenstein’s appeal to facts that avoids such problems. Or as Dromm also puts it, he seeks to provide us with an account of how to read relevant remarks of Wittgenstein.
So, what is the purpose of Wittgenstein’s appeals to facts? To explain this, Dromm makes use of Nozick’s notion of a philosophical explanation. The purpose of such explanations, as Dromm describes them, is not to increase our knowledge about anything. Rather, it is to enable us to understand better what we already assume or have established to be true—like Wittgenstein says that philosophy clarifies to us something we already know but have difficulty to understand. More specifically, what Nozickian philosophical explanations aim to do is to resolve apparent incompatibilities between different bits of our knowledge, whereby something we take to be true seems to exclude the possibility of something else we also take to be true. How such an explanation achieves its goal is by presenting us with a possibility that reconciles the different things we know so that they become compatible and comprehensible.
An example of a problem of relevant kind can be sketched as follows. We know or assume that there is rule following. But we also know or assume (in light of Wittgenstein’s arguments) that the terminus of the interpretation of rules cannot be anything mental, because any mentally entertained interpretation is itself open to various interpretations. So, how is the possibility of rule following to be understood? In response to this, Wittgenstein says that “there is a way of grasping a rule that is not an interpretation, but which is exhibited in what we call ‘obeying the rule’ or ‘going against it’ in particular cases” (PI §201). On Dromm’s reading (that also takes on board aspects of Gordon Baker’s later interpretation of Wittgenstein), the idea of rule following as a practice with which we are presented here is meant to constitute an alternative to the mentalistic account of rule following. We are offered a different possible way to think about rule following that dissolves the paradox of there being no right or wrong way to follow a rule that arises when we assume the rightness/wrongness of rule following to be a matter of interpretation. Here Dromm emphasizes that, for the purpose of such an alternative account, the mere possibility of what Wittgenstein says suffices. No assertion about how things actually are is needed for the explanation to do its job.
If Wittgenstein can be understood as providing this kind of Nozickian philosophical explanations, his disinterestedness in the empirical justification of his statements about natural facts or history is explained. For such an explanation, the truth of what is said is irrelevant, and the crucial thing is our ability to imagine possibilities. Thus, we also have a way to understand what goes wrong with interpretations such as Kripke’s and Williams’s that take Wittgenstein’s account of rule following to be based on factual claims about human practices. And, indeed, as Dromm points out in support of his interpretation, we do often find Wittgenstein making relevant qualifications in connection with his statements about natural history. For example, his conception of first-person pain expressions as an extension of primitive behavior is prefixed by saying that it is “one possibility.” Similarly, Wittgenstein often qualifies his statements about the development of language from more primitive forms with phrases such as “I would like to say” or “I want to say,” indicating that what is expressed is a particular conception, not necessarily the only possible one. Similar reservations are also made in the context of the rule-following discussion (see PI §198).
I entirely agree with Dromm that such qualifications are very important and that ignoring them, or the details of how Wittgenstein expresses himself, is problematic for the interpretation of his philosophy. More abstractly, we must try to understand the kind of explanation he is offering, and to this his painstakingly drafted and redrafted formulations are a clue that we cannot ignore. Nevertheless, in the end, I also feel uneasy about where Dromm leaves things. That is, I believe that Dromm is right that Wittgenstein’s philosophical purposes are met without him having to make any assertions about facts of nature and that the absence of empirical justifications therefore does not constitute a defect or a weakness. But still, it does not seem quite satisfactory to say that Wittgenstein is only inviting us to imagine possibilities. The problem I have can be put like this: Would any possible account of rule following do just as well as the explanation of rule following as a practice, for instance? What about the possibility that the correct interpretation of the rule is determined by extraterrestrials who have programmed humans to act in relevant ways so that no interpretation of rules is required? Would our having been thus programmed do equally well as the explanation of rule following? More generally, if there is more than one possible way to explain the possibility of a phenomenon, is any account as good as any other? If not, how do we choose between accounts? What is the criterion of adequacy for a philosophical clarification, as Wittgenstein conceives them? Notably, if any possible account is not as good as any other, this means that there is something more complicated going on in Wittgenstein’s texts than the mere presentation of alternative possibilities with the purpose making compatible what seems incompatible. And if, from the point of view of Nozickian philosophical explanations, any possible account that can dissolve apparent tensions is as good as any other, then maybe this is not, after all, an adequate model for what Wittgenstein is doing.
In other words, although I would agree with the general thrust of Dromm’s criticism of “factualist” interpretations of Wittgenstein that make him vulnerable to the accusation that he fails to justify his alleged factual assertions, I did find the positive side of Dromm’s interpretation somewhat underdeveloped. In this regard, his strict focus on Wittgenstein’s use of natural facts or history seems problematic in that perhaps comparing Wittgenstein’s employment of natural history with his other methods, such as the tabulation of grammatical rules, would be helpful at this point. To spell out this suggestion, I take it that the role of such rules in Wittgenstein (for example, the definition of meaning as use) is to function as objects of comparison that constitute tools for ordering, and thus making perspicuous, complex and messy uses of natural language. This means that grammatical rules are essentially modes of presenting language use with the purpose of philosophical clarification. But it seems that something similar could be said about Wittgensteinian natural history too—at least sometimes. For example, it seems that one might well understand the purpose of his account that first-person expressions of pain are an extension of natural primitive pain behavior and his story about how children learn the use of linguistic pain expressions in the following way. What we are offered here is a model for the functioning of relevant expressions. This model has its basis in natural historical considerations about language learning, but when used as a grammatical model for the functioning of relevant expressions, the role and status of the model are not those of an empirical hypothesis about language. (Hence, worries about its empirical justification would be misplaced.) Rather, what this account of language learning provides us with is a mode of presenting the functioning of relevant expressions entirely comparable to Wittgenstein’s use of grammatical rules in that such models based on natural history also serve the goal of perspicuous representation in the capacity of objects of comparison. Notably, taken as such, a model Wittgenstein’s account of how first-person pain expressions are learned can then also be further extended to other first-person expressions of sensations. Certainly, such an extension would be illegitimate if what is at stake is an empirical claim about language. But things are different with grammatical models with regard to their generality and justification. What is crucial is whether and to what extent the model can make the functioning of relevant expressions comprehensible to us with a view to resolving associated philosophical problems. And importantly, not just any possible model or way of conceiving the functioning of language can render it clear and solve relevant philosophical problems. (Some create further problems, as when bringing in extraterrestrials to explain rule following.) Thus, we now also have an answer to the question raised above about the criteria of adequacy for philosophical explanations of language. Every possible explanation is not as good as any other, because not all of them can solve our philosophical problems.
Another difficulty with Dromm’s interpretation comes to view here. As noted, the account of Wittgenstein’s employment of natural history just sketched might not work for every case of Wittgenstein’s appeal to natural history. Talk about natural history may serve different purposes in his discussions. By contrast, however, I get the impression from Dromm’s book that he takes Wittgenstein to be always employing natural history for the purpose of Nozickian philosophical explanations. I find this very narrow, too uniform for how Wittgenstein generally operates, and therefore implausible. (In fairness to Dromm, in relevant introductory contexts, he only says that Wittgenstein’s explanations are akin to Nozick’s. But then no explanation is given for how Wittgenstein’s explanations might deviate from Nozick’s.) Thus, for example, when Wittgenstein talks in PI §143 about a pupil’s inability to learn how to follow a rule (certainly a point about the natural history of human learning capacities), it is hard to see how this would be intended to resolve any tensions in the style of Nozickian explanations. Rather, the point seems to be this: Wittgenstein uses this as a way to remind the reader that there is not always a sharp distinction between following a rule and not following it. There is also the case of following a variant of a rule (as in the case of the pupil with a somewhat limited capacity to learn), but the distinction between following such a variant rule and not following one at all is not sharp. If this is the way to read the remarks around §143, then I do not quite see how it would fit the general pattern of the employment of natural historical considerations that Dromm detects in Wittgenstein.
There are also other lesser points of criticism. As part of his discussion of Wittgenstein in the light of Nozickian explanations, Dromm compares such explanations to explanations about art and music, saying that the purpose of such explanations is not to provide us with new information either but to enhance our understanding in a different way. This may well be so and is perhaps an interesting point of comparison. But Dromm leaves this at the level of a mere suggestion that is not properly spelled out. The same goes for his comparison between what Wittgenstein is doing and literature (fiction, storytelling) that comes up here and there. However, if something like this is taken up at all, it would be desirable to have it properly explained. In the same way, I found Dromm’s comparison between Wittgenstein’s approach and those of Socrates, Freud, and Nietzsche very thin and consequently unhelpful. Rather than trying to undertake such comparisons in the space of a few pages, it would seem better to have dropped them and extend discussions elsewhere to the benefit of those discussions whose pace on more than one occasion seemed a bit too quick. At times, Dromm’s manner of quoting Wittgenstein is also somewhat unmethodological: remarks from the early 1930s are quoted alongside later remarks without attention to any possible developments of Wittgenstein’s views in the 1930s and without explaining the context of the quotations. Although Dromm is by no means the worst example of this among writers on Wittgenstein, one would hope for greater care.
Nevertheless, my earlier characterization of Dromm’s book as a very welcome contribution was not made out of politeness. Despite the issues raised above, I do think that Dromm makes some very interesting points, and I found the book thought provoking in a good way. It made me think, and it forced me to try to spell out my reservations and disagreements. In this way, it definitely had positive value for me. I also found much to agree with in his criticisms of “factualist” interpretations of Wittgenstein. For regardless of my reservations about Dromm’s positive account of the role of natural historical considerations in Wittgenstein, the problem that the “factualist” interpretations of Wittgenstein face is well explained. So, I will recommend this book to students and would recommend it to colleagues who are concerned with relevant issues. That I cannot say about all the Wittgenstein books I have read.