1. You've read Peter van Inwagen's chapter on philosophical failure.
2. Now you can* read John Martin Fischer's reply.
3. It is now up to you** whether to reply, however it is true that you should*** do so.
* 'can' in some sense profoundly indeterministic, ultimacy-bearing sense
** 'up to you' in some sense compatible with your deliberating about whether to reply
*** 'should' relative to some justified standards concerning your individual conduct given your aims and the Objective Value™ of the Irrepressible GFP Online Reading Group.
---------Fischer's comments on van Inwagen's "Philosophical Failure":
"I should begin by saying that I admire this book greatly. It is beautifully written, and very interesting and stimulating throughout. Not surprisingly, the book is filled with ingenious argumentation and penetrating insights. Here I shall focus solely on van Inwagen's fascinating suggestions about philosophical methodology in Chapter 3.
The chapter is entitled "Philosophical Failure," but it is really about philosophical methodology in general, or perhaps more precisely, about how best to assess and also to develop philosophical argumentation. Van Inwagen begins by rejecting a model of philosophical argumentation quite similar to what Robert Nozick has called "coercive philosophy"; on this model, one seeks to provide knockdown arguments from indisputable premises. As regards this sort of model, Van Inwagen points out (p. 36) that Nozick said that when he was young he thought that a philosophical argument is adequate only if anyone who understood the premises but not the conclusion would die! (One imagines the brain exploding. And note that even Nozick's characterization captures a less than optimally stringent model, insofar as it leaves it open that someone could reject the premises without fatality.)
Van Inwagen goes on to consider a second model. On this view, philosophical argumentation can be thought of as a kind of debate between two parties who have opposite views about the issue under consideration, where the goal of each is to convince the other to give up his position and adopt the competing view. (A weaker requirement would be that the other party to the debate switch from accepting the competing view to agnosticism; Van Inwagen doesn't explicitly consider this possibility. An even weaker requirement--or family or requirements--would be that the other party decreases to some degree his confidence in his view.) So suppose the issue is whether causal determinism is compatible with moral responsibility. Here we are to envisage an idealized debate (some of the conditions of idealization are discussed on pp. 42-3) between a compatibilist and an incompatibilist, where the compatibilist seeks to convince the incompatibilist to adopt compatibilism, and the incompatibilist seeks to convince the compatibilist to adopt compatibilism. The argument offered (say) by the compatiblist is deemed successful only if the incompatibilist is persuaded to adopt compatibilism.
Van Inwagen rejects this second model for the same reason he rejects the first, saying that it places the bar too high: "I very much doubt whether any argument, or any set of independent arguments, for any substantive philosophical conclusion has the power to turn a determined opponent of that conclusion, however rational, into an adherent of that conclusion." (p. 45)
Van Inwagen prefers a third model, according to which we understand philosophical argumentation as like an idealized debate between proponents of competing positions, where the goal is not to convert the other debater, but to convince an idealized "agnostic"--a person who is "neutral" in the sense that he has no particular antecedent inclination to accept the relevant position. Here the two debaters address a third party, rather than each other; and the goal is not "conversion", but securing "conviction", as it were. Van Inwagen contends that this weaker model is more reasonable, and that it has some advantages with respect to understanding issues about begging the question and the burden of proof.
I find Van Inwagen's proposal highly attractive, and I agree with him that it is preferable to the first two. I certainly have not offered such an explicit and helpful characterization of the model (as Van Inwagen has here), but in a way I have suggested something similar. That is, I have suggested that (for example) a compatibilist understand his arguments as directed to a fair-minded, reasonable person without a prior commitment to either compatibilism or incompatibilism. (In recent work by Haji and McKenna one sees a similar view.) In certain debates one often reads or hears the following sort of reply to a proposed argument or consideration: "but an incompatibilist would reject that…" It is as if a compatibilsit may only present considerations and arguments that would not be rejected by someone with an antecedent and iron-clad commitment to incompatibilism! But this really sets the bar too high; clearly, if we accepted this sort of requirement, no philosophical progress could be made with respect to substantive and contentious philosophical issues. I think it is quite helpful to point out that the arguments and considerations should be addressed to a neutral agnostic, rather than a committed opponent. As Van Inwagen points out, adopting this model might well result in different views about what is argumentatively permissible, as opposed to question-begging or otherwise illicit, and where the burden of proof lies.
Van Inwagen goes on to argue that even on the somewhat more permissive third model, not only is the atheist's argument based on evil a failure, but all arguments for substantive, interesting philosophical theses will turn out to be failures. (Here I won't assess either of these claims.) I wonder why Van Inwagen thinks it is a fatal criticism of the first two models that they would have it that no philosophical arguments are successes, but not a problem for his favored model that it also has this implication. Perhaps Van Inwagen believes that the first two models set the bar unreasonably high so that it would be unreasonably hard to meet their criteria for success, whereas the third model sets the bar at just the right height--but that even so no philosophical argument for an interesting, substantive thesis is successful. But perhaps it would be worth considering another model (or family of models), by reference to which it would turn out that some (but not all) philosophical arguments for the relevant sorts of theses are indeed successful (or at least successful to certain degrees).
Consider a model that adopts the assumptions of Van Inwagen's third model--we have a debate addressed to a neutral agnostic (with respect to the view under dispute). But here success does not require that the neutral agnostic actually adopt the relevant view; what is required is simply that the agnostic's inclination to adopt the thesis increases (perhaps to some threshold amount). Here we can again imagine a family of different views; the weakest view would be that success simply requires some increase in the agnostic's inclination to accept the relevant view, whereas stronger views would require a range of more significant increases in the strength of the inclination to accept the view in question. On this sort of model, we presumably could say that some (but certainly not all) philosophical arguments for contentious, substantive positions are successful. It is arguably at least a strike against Van Inwagen's view that it seems (by his own admission) to imply that all philosophical arguments (of the sort under consideration) are failures, and a virtue of my suggestion that it does not have this implication. (I leave aside for now the question of whether the atheist's argument from evil is successful, on my proposed model.)
Van Inwagen is aware of the possibility of models similar to the one I have suggested above; indeed, he says:
I would suppose that most real agnostics, most actual people who do profess and call themselves agnostics, are not neutral agnostics. Most agnostics I have discussed these matters with think that it's pretty improbable that there's a God. Their relation to the proposition that God exists is very much like my relation to the proposition that there are intelligent non-human beings inhabiting some planet within 10,000 light-years of the Earth. And this consideration suggests a possible objection to my definition of philosophical success. Call those agnostics who think that it's very improbable that there is a God weighted agnostics. An argument for the non-existence of God, the argument from evil for example, might be a failure by my criterion because it lacked the power to transform ideal (and hence neutral) agnostics into atheists. But it might, consistently with that, have the power to transform neutral agnostics into weighted agnostics. If it does, isn't it rather hard on it to call it a failure? (p. 50)
Van Inwagen says that he would not object to revising his model to allow for success in the case described (in which the neutral agnostic becomes a weighted agnostic). But he goes on to contend that if the considerations he invokes in seeking to show that the argument from evil is incapable of turning neutral agnostics into atheists are persuasive, they would be equally persuasive in showing that the argument from evil is incapable of turning neutral agnostics into weighted agnostics. (p. 51)
But note that even Van Inwagen's permitted revision is significantly different from the model I sketched above, insofar as my model does not require for success that a neutral agnostic become a weighted agnostic; after all, a weighted agnostic, according to Van Inwagen, believes that it is "very improbable" that God does not exist. All that is required on my suggested model is that a neutral agnostic become inclined to some degree (which may fall considerably short of the threshold posited by Van Inwagen) to accept the relevant position (say, atheism). Given this important difference, even if Van Inwagen's conditional is true (that if the argument from evil is incapable of transforming a neutral agnostic into an atheist, then it is also incapable of transforming a neutral agnostic into a weighted agnostic), it would not follow that if the argument from evil is incapable of transforming a neutral agnostic into an atheist, it would also be incapable of increasing the inclination of a neutral agnostic to accept atheism (to the relevant degree).
Why exactly would we want a model by reference to which we could say that some (but not all) philosophical arguments for interesting, substantive thesis are successful? I suppose this may depend to some extent on one's temperament or philosophical personality. Some philosophers might well think that it is accurate to describe all philosophical arguments for substantive, contentious theses as failures, although perhaps they may invoke a model like mine to say that some arguments are "better" than others, or come closer to success than others, or perhaps are "partial" successes (to one degree or another). Others will be a bit less dour about the possibility of philosophical success. I suppose that my own view is that the interesting philosophical issues are so incredibly difficult, we should consider ourselves successful if we could change the inclination of an idealized neutral agnostic even a little bit. (And note that this change is in the context of a debate where the proponent of the opposing view can mount a vigorous argument of his own.)
There are of course various problems for the very sketchy proposal I have made. I simply note that it applies most naturally to highly contentious philosophical theses, where it is plausible that an ideal agnostic would in fact be neutral. If one considers theses which are in dispute but whose objective probability of truth is relatively high, then even a very uninspiring argument would presumably change (significantly) the inclination of an idealized unweighted agnostic. This suggests that perhaps in general the requirement should be that the argument move an ideal agnostic who has an inclination to accept the relevant view that matches (in a suitable sense) the objective probability of truth of the relevant view. But I will simply follow Van Inwagen in noting this problem without addressing it in detail (p. 49)."