In On Silencing and Systematicity: The Challenge of the Drowning Case, McGowan et al. intend -among other things- to challenge the idea that silencing is a "systematically generated, illocutionary-communicative failure (of a very specific sort)" (McGowan et al. 74). Their argument goes as follows: 1) for a communicative failure to be taken under this account of silencing (which they call the Hornsby-Langton account, or H&L), it needs to meet two requirements: it should involve uptake failure, and this uptake failure "must be brought about in a systematic manner" (id. 76); 2) There is a case -the drowning case-that meets these two requirements and, however, it does not appear to be a case of silencing; 3) Thus, if the drowning case is a good counterexample, the H&L account of silencing is insufficient as it is. In this very brief comment, I will try to show that the drowning case is not a good challenge to the H&L account of silencing. If my argument is sound, then the need of a different counterexample will become evident if we are ever to challenge what I also consider a troublesome account of silencing.

I will begin by reproducing the authors' characterization of the drowning case: Suppose that Sally is drowning and Peter, who is walking by, notices this.

When Peter attempts to save her, Sally says "no" with the intention of refusing Peter's assistance. Peter fails to recognize her intention to refuse, however, because Peter believes that drowning people want to be saved, and so he interprets Sally's utterance of "no" as an expression of denial; he believes that she is expressing the belief that drowning is just too awful to actually be happening to her. (McGowan et al. 79)

According to the authors, the drowning case appears to meet both of H&L'S requirements for silencing. It involves uptake failure, for Peter is not just ignoring Sally; instead, he fails to understand her intention of drowning as a whole. But this failure to recognize her intention is not due to Peter's idiosyncrasies: it does not seem dependent on his particular beliefs, for if we were facing the same situation we would (probably) also fail to recognize Sally's intention of drowning. As the authors put it, we would probably believe that Sally's "no" is an expression of denial, "that she is expressing the belief that drowning is just too awful to actually be happening to her" (McGowan et al. 79). And if Peter's uptake failure is not idiosyncratic, then it must be brought about in a systematic manner. Therefore, the drowning case meets both of H&L'S requirements for silencing.

However, McGowan et al. are not so keen to admit that Sally's communicative failure is a case of silencing. "It just seems plain wrong" (McGowan et al. 79), they say, but they do not give reasons other than common sense and intuition for why it is so hard to admit this as a case of silencing. I will put aside any considerations regarding the rest of their article to focus exclusively on the relationship between silencing and the drowning case, for I, with them, consider that it is not a case of silencing; the difference being, however, that I do not admit that it necessarily meets both of H&L'S requirements in the first place.

As I said before, the first requirement of H&L'S account of silencing is uptake failure. This can happen for a lot of reasons, and many of them are perfectly compatible with the drowning case: suppose, for example, that Peter is cognitively disabled. If this were the case, we can assume that he would probably fail to recognize Sally's intention of drowning, given that he could think that she is playing a joke on him even if she did other things that usually accompany acts of refusal: she could try to break away from his grasp or push him away, but he would consider it all as part of the joke. Therefore, there are possible situations in which the drowning case could involve some forms of uptake failure, so I do not find this point of their argument particularly troublesome when it is taken by itself.

But it seems that the drowning case presented by McGowan et al. supposes that Peter is not cognitively disabled, nor that his uptake failure depends on particular circumstances and beliefs which the authors call "idiosyncratic." On the contrary, his uptake failure is supposedly dependent on the widely-held belief that drowning people want to be saved. It is this belief which leads Peter to fail to recognize Sally's refusal, for it renders him unable to understand that she wants to drown. And given that his belief is presumably shared by the majority of ordinary people, it must be systematic rather than idiosyncratic, for which reason the drowning case would meet the second requirement of H&L silencing.

The thing that I find troublesome about this argument is the very idea of "systematicity." There is no unequivocal definition of systematicity in the article; after all, part of the authors' purpose is to show that several possible definitions of systematicity that one could intuitively deem reasonable fail to capture completely the matter at hand. However, the authors seem to agree that -in this particular context- uptake failure is systematic when it "is brought about in a non-idiosyncratic [...] manner" (McGowan et al. 79). If this is the case, then we can begin our analysis by comparing the "systematic belief" ofthe drowning case with other beliefs that we intuitively consider systematic.

What does it mean to say that the belief "drowning people want to be saved" is systematic in the same way that other beliefs such as "women are bad drivers" are? There are at least two possibilities which I will now address: either 1) a belief is systematic merely when it is shared by a group of people (or, in other words, when it is not-idiosyncratic), or 2) a belief is systematic when there are some features in society (i.e. cultural, political, etc.) that bring it about in groups of people and that constantly reinforce it.

Both the beliefthat drowning people want to be saved and the belief that women are bad drivers are shared by groups of people, but I do not think that this is enough for us to consider them systematic. If this was taken to be the case, then the authors' example of an inside joke could not be idiosyncratic, as they think (cf. McGowan et al. 76). Actually, it would be very difficult (if not impossible) to find an example of an idiosyncratic belief if we take seriously the first possibility that I point out, for the vast majority of beliefs (if not all of them) are usually shared by groups of people, not exclusively by one person. The fact that a belief is shared or common is probably a necessary condition for systematicity, but it is not a sufficient one.

The second possibility, however, seems to solve this problem. If we take it to be true, then it is not possible to consider the inside joke of a group of friends as systematic, for it is not necessarily brought about or reinforced by cultural or political features of society. On the other hand, the belief that women are bad drivers could certainly be understood as a systematic one under this definition, for it is not hard to identify several depictions of women in mass media as bad drivers which could bring about and constantly reinforce this (obviously false) belief. This definition of systematicity is even compatible with the theater case and the dinner party guest case that McGowan et al. point out -following ^{Maitra's (2004}) characterization of silencing as context-specific and rule-governed interpretive mistakes- (cf. 84-85), for one could admit that the beliefs that cause uptake failure in both of these cases could be brought about and reinforced by society. However, it differs slightly from Maitra's characterization, insofar as it avoids the problem of deciding whether there is a certain rule in place that is being followed or a false belief that a certain rule is in place (cf. McGowan et al. 86), given that it emphasizes the reinforcement itself of the belief, independently of whether it is brought about by rule-following or not.

With all this in mind, let us return to the drowning case. If we are to consider the drowning case under the scope of the second definition of systematicity I presented above, we must ask ourselves the following question: "Is there something in Peter's failure to understand Sally's refusal that is brought about and constantly reinforced by certain features of society?" If the answer to this question is affirmative, then one would have to show what these features are and how they operate; if it is negative, then one would have to admit that Peter's failure is idiosyncratic,¹ even though the belief that causes his failure to understand Sally's refusal could be shared by other people. For I do not see how Peter's belief that "drowning people want to be saved" could be brought about and reinforced by social features of society in the same way that the belief "women are bad drivers" is,² my answer to the above question is negative. And if the second definition of systematicity I mentioned above is sound, then the drowning case is not a good enough challenge to H&L silencing for it does not meet this particular requirement of systematicity.

So far I have tried to show that McGowan et al.'s challenge to H&L silencing is not a good one since there is at least one reasonable definition of systematicity that is not compatible with the drowning case. And even though I must recognize that this definition of systematicity, when applied to the H&L account of silencing, is also unable (by itself) to "do the remaining discriminatory work" (McGowan et al. 86) between problematic and non-problematic types of silencing, this probably gives us a clue as to what could be the matter at hand here: namely, that there is no relevant illocutionary difference between these two types of silencing, for which one could say that what makes silencing problematic is not how it is brought about, but rather what consequences it has in some particular contexts (such as the sexual refusal one).

I would like to conclude my comment by giving these authors credit for putting forward the fact that the H&L account of silencing is troublesome, at the very least, for it is not evident what systematicity is in situations of silencing. Hopefully, discussions of this sort will eventually lead us to an exhaustive answer to this difficult yet interesting problem.