Hostname: page-component-745bb68f8f-kw2vx Total loading time: 0 Render date: 2025-02-11T00:00:43.088Z Has data issue: false hasContentIssue false
  • English
  • Français

SUBJECT-SENSITIVE INVARIANTISM, HIGH-STAKES/LOW-STAKES CASES, AND PRESUPPOSITION SUSPENSION

Published online by Cambridge University Press:  31 October 2018

Michael Blome-tillmann
Rights & Permissions [Opens in a new window]

Abstract

It is a familiar criticism of Subject-Sensitive Invariantism that the view makes incorrect predictions about cases in which the attributor of knowledge is in a high-stakes situation and the subject of the attribution in a low-stakes situation. In a recent paper in this journal, Brian Kim has argued that the mentioned type of case should be ignored, since the relevant knowledge ascriptions are inappropriate in virtue of violating an epistemic norm of presupposing. I show, pace Kim, that the mentioned utterances do not carry factivity presuppositions. To this end I discuss a phenomenon known as presupposition suspension, which is widely associated with the presuppositions of epistemic factives such as know that p or discover that p. I argue further that the problem of unknown presuppositions discussed by Kim can be circumvented by slightly amending the cases at hand. In particular, I demonstrate that factivity presuppositions are unobjectionable in problem cases in which the high-stakes ascriber knows the presuppositions at issue to be true.

Type
Articles
Information
Episteme , Volume 17 , Issue 2 , June 2020 , pp. 249 - 254
Copyright
Copyright © Cambridge University Press 2018

1. THE PROBLEM

Subject-Sensitive Invariantism (henceforth ‘SSI’) is, roughly speaking, the view that whether one knows a given proposition p depends partly on one's practical situation or on what is at stake for one with respect to p.Footnote 1 It is a familiar criticism of the view that SSI seems incompatible with data stemming from cases in which the attributor of ‘knowledge’ is in a high-stakes situation and the subject of the attribution in a low-stakes situation. While SSI predicts that Lo – the subject in the low-stakes situation – knows p, Hi – the subject in the high-stakes situation – can seemingly truthfully assert ‘Lo doesn't know p’ – contrary to the truth-value predictions made by SSI. Jason Stanley (Reference Stanley2005: 5) has produced an elegant example illustrating the problem, which is based on Keith DeRose's (Reference DeRose1992, Reference DeRose2005) widely-discussed bank cases:

High-Attributor/Low-Subject Stakes (HALSS)

Hannah and her wife Sarah are driving home on a Friday afternoon. They plan to stop at the bank on the way home to deposit their paychecks. Since they have an impending bill coming due, and very little in their account, it is very important that they deposit their paychecks by Saturday. Hannah calls up Bill on her cell phone, and asks whether the bank will be open on Saturday. Bill replies to Hannah, ‘Well, I was there two weeks ago on a Saturday, and it was open.’ After reporting the discussion to Sarah, Hannah concludes that, since banks do occasionally change their hours, ‘Bill doesn't […] know that the bank will be open on Saturday’. (Stanley Reference Stanley2005: 5)

The example is problematic for SSI because Bill is in a low-stakes situation and thus, according to SSI, knows that the bank will be open on Saturday. Intuitively, however, Hannah is speaking truly when asserting (1):

  1. (1) Bill doesn't know that the bank will be open on Saturday.

Thus, the objection goes, SSI makes mistaken predictions with respect to High-Attributor/Low-Subject-Stakes (henceforth ‘HALSS’), for it ascribes an incorrect truth-value to Hannah's utterance of (1).

2. KIM'S ARGUMENT

In a recent paper in this journal, Brian Kim (Reference Kim2016) aims to defend SSI against the above objection. He argues that the defenders of SSI have a simple but powerful response to the challenge posed by HALSS. According to Kim, Hannah's utterance of (1) is infelicitous or ‘inappropriate’, because it carries an unknown presupposition – that is, a presupposition that Hannah herself doesn't know. Here is Kim:

In [HALSS], it would be inappropriate for [Hannah] to assert that the bank will be open because she does not possess the evidence that would license both her and [Sarah] to take for granted that the bank will be open the following day. And since asserting that [Bill] does not know that the bank will be open triggers the presupposition that the bank will be open, it would also be inappropriate for [Hannah] to assert that [Bill] does not know. Given the presupposition triggering of factive verbs, straightforward third-person denials of knowledge are inappropriate in high stakes conversational contexts. Therefore, the contextualist's main argument against SSI is based upon a false premise. (Kim Reference Kim2016: 235)

Let us reconstruct the argument in a little more detail. Kim's argument begins with a general claim about presuppositions and occurrences of factive predicates such as ‘know(s) p’ when embedded under negation – what I shall call the Presuppositionality of Epistemic Factives under Negation (PEFN):

PEFN An assertion of the form ‘x doesn't know p’ presupposes p.

PEFN is initially plausible and convincing in virtue of being an instance of a more general claim – namely, that presuppositions are preserved under negation: sentences containing presupposition triggers usually retain their presuppositions when embedded under negation.Footnote 2 From PEFN Kim then infers, by instantiation:

  1. (i) Hannah's assertion of (1) in HALSS presupposes that O.

    And from SSI it follows that:

  2. (ii) Hannah in HALSS doesn't know O.

Kim furthermore assumes, implicitly, what we may call the Knowledge Norm of Presupposition (KNP):Footnote 3

KNP Presuppose p only if you know p.

It then follows from (i) and (ii) that Hannah's utterance of (1) in HALSS is in violation of KNP: Hannah's utterance of (1) presupposes a proposition that Hannah doesn't know – namely, the proposition that O. In other words:

  1. (iii) Hannah's assertion of (1) in HALSS represents a violation of KNP.

    Given this violation of KNP, we may conclude with Kim that:

  2. (iv) Hannah's assertion of (1) in HALSS is inappropriate.

As Kim's argument suggests, Hannah's assertion of (1) is inappropriate, because it violates the Knowledge Norm of Presupposition.

3. CONTEXTUAL SUSPENSION OF PRESUPPOSITIONS

To begin the discussion of Kim's argument, note that its conclusion (iv) is rather surprising and shouldn't be taken as uncontroversial. This is so because Hannah seems, intuitively, perfectly well positioned to assert that Bill doesn't know O. In fact, the assertion at issue simply doesn't seem to be suffering from a violation of an epistemic norm of presupposing. From a purely phenomenological point of view, Hannah's utterance is unassailable – a fact that is widely reflected in the literature on the topic. For it is the very fact that Hannah's utterance appears natural, felicitous, and true that is taken to constitute a problem for SSI and its predictions in the first place. If Hannah's utterance of (1) appeared dodgy or inappropriate, why did epistemologists in general and the defenders of SSI in particular take it seriously as a potential counterexample to their views? On the face of it, Kim's conclusion that Hannah's utterance is inappropriate is thus in conflict with the data.

Let us therefore assume, for the sake of argument, that Kim's conclusion is mistaken. At least one of his premises must then be false, too. A likely candidate for controversy amongst Kim's premises is PEFN – the Presuppositionality of Epistemic Factives under Negation.Footnote 4 How problematic is PEFN? In its generality, PEFN is likely to be mistaken. To see this note that epistemic factives tend to suspend their presuppositions in certain contexts. As Barbara Abbott (Reference Abbott, Birner and Ward2006: 1) puts it:

Some presuppositions seem to be weaker than others in the sense that they can be more easily neutralized in some contexts. For example, some factive verbs, most notably the epistemic factives like know, be aware, discover are known to shed their factivity fairly easily….

To illustrate further the phenomenon Abbott refers to, consider the following example from Abrusán (Reference Abrusán2016: 167):

  1. (3) I have no idea if Mary is cheating on John. But if he discovers that she is, he will be sad.

‘Discovers that p’, if embedded in the antecedent of a conditional, usually retains its presupposition that p. However, note that if it is part of the common ground that the speaker doesn't know whether p, then the embedded presupposition that p is suspended or neutralized. (3) is such an example: in (3) the presupposition that Mary is cheating on John is suspended (or cancelled) because it is part of the common ground that the speaker doesn't know whether Mary is cheating on John.

Similar phenomena can be observed with respect to the epistemic factive ‘know(s) p’. Consider a context in which it is common ground that the speaker doesn't know whether Tim's lottery ticket is a loser, despite the fact that the probability of it being a loser is exceedingly high. In such a context, one can then truthfully and felicitously assert (5), without presupposing (6):

  1. (5) Obviously, Tim doesn't know that his lottery ticket is a loser.

  2. (6) Tim's lottery ticket is a loser.

Here is a second example: at the time of writing, the chances that Real Madrid will not top the Primera División league table this year are very low indeed – Real Madrid not winning the league is what pundits sometimes call ‘a mere theoretical possibility’. In a context in which it is part of the common ground that the speaker (and everybody else) doesn't know whether (8), one can then truthfully assert (7), without presupposing (8):

  1. (7) It's very likely that Real Madrid will top the league table this year. But (even) Natalia doesn't know that they will.

  2. (8) Real Madrid is going to top the league table this year.

Again, in the envisaged context, an utterance of (7) doesn't carry the presupposition that Real Madrid is going to win the league this year, because it is part of the common ground that that proposition is not known by the speaker.

With the above data at hand, we have a plausible explanation of why Hannah's utterance of (1) in HALSS isn't in violation of the KNP, and thus doesn't appear inappropriate: Hannah's utterance of ‘Bill doesn't know O’ doesn't trigger the presupposition that O, because it is part of the common ground, in HALSS, that Hannah doesn't know whether O, despite the fact that it is quite likely that O. The presupposition that O is, accordingly, cancelled or suspended by context and Kim's assumption (i) – that Hannah's utterance of (1) in HALSS presupposes O – is mistaken. PEFN is false.Footnote 5

4. KNOWLEDGEABLE ASCRIBERS

While there are good reasons for the view that Hannah's utterance of (1) in HALSS doesn't trigger an unknown presupposition, and thus doesn't violate an epistemic norm of utterance presupposition, there is another response to Kim's argument that grants Kim the controversial claim that (1) triggers the mentioned presupposition in HALSS. To see what I have in mind note that the problem allegedly caused by Hannah's utterance of (1) can be circumvented by amending the example slightly. Consider a variant of HALSS – HALSS* – in which Hannah has just been inside the bank and has spoken to the bank tellers about their opening hours. In addition, she has checked the opening hours on the company's website and found that the bank will be open on Saturday. According to SSI, Hannah then knows O, even relative to her high-stakes situation, and she can, therefore, truly assert (9):

  1. (9) I know that the bank will be open on Saturday.

Moreover, Hannah can, intuitively, truly and without inappropriateness assert (1), reproduced here for convenience:

  1. (1) Bill doesn't know that the bank will be open on Saturday.

Now, even if Kim were right and all utterances of (1) were to carry the presupposition that O, Hannah's utterance of (1) in HALSS* clearly wouldn't violate any knowledge norm of utterance presupposition: Hannah does, after all, know the presupposition that O in the envisaged case. In fact, note that Hannah can, in the envisaged case, make this very fact explicit by – entirely felicitously – asserting (10):

  1. (10) I know that the bank will be open on Saturday, but Bill doesn't.

While Hannah has, in HALSS*, enough evidence to satisfy the higher standards prevalent in her own high-stakes context, Bill doesn't. But Bill nevertheless has enough evidence to satisfy the standards in his own, low-stakes context. Thus, according to SSI, Hannah's utterance of (10) is false, because, according to SSI, Bill knows that the bank will be open on Saturday. But Hannah's utterance of (10) doesn't appear false or inappropriate at all. To the contrary, it appears entirely felicitous, appropriate, and true – just as her utterance of (1) in HALSS did.Footnote 6

5. CONCLUSION

I have argued that Hannah's utterance of (1) in HALSS is not in violation of an epistemic norm of presupposing, because it doesn't trigger a presupposition in the first place. The phenomenon of contextual presupposition suspension at play in the example is shared by all epistemic factives and is well-documented in the literature on presuppositions. I have furthermore argued that Hannah's utterances of (1) as well as (10) in HALSS* are not in violation of an epistemic norm of presupposing either, because Hannah knows, in HALSS*, the presupposition in question. Thus, the relevant presupposition is either not known and suspended (as in HALSS) or known and not suspended (as in HALSS*). Either way, no epistemic norm of presupposing is violated. As a consequence, the argument against SSI based on high-ascriber/low-subject cases retains its full force. It cannot be defused by appeal to an epistemic norm of presupposing.Footnote 7

Footnotes

1 See Fantl and McGrath (Reference Fantl and McGrath2002, Reference Fantl and McGrath2007, Reference Fantl and McGrath2009), Hawthorne (Reference Hawthorne2004) and Stanley (Reference Stanley2005).

2 This is a familiar and widely studied phenomenon. See, for instance, Karttunen and Peters (Reference Karttunen, Peters, Oh and Dinneen1979).

3 Kim formulates his norm of presupposing in terms of evidence rather than knowledge (Kim Reference Kim2016: 235). For the sake of simplicity I shall proceed the discussion here with KNP, since the only salient shortcoming of Hannah's evidence in HALSS is that it is not strong enough for knowledge.

4 Some will, of course, also worry about KNP (see, for instance, Brown Reference Brown2008, Reference Brown2010; Lackey Reference Lackey2007, Reference Lackey2016). However, as mentioned in fn. 3, Kim does not commit to KPN, which I only use as a convenient placeholder for Kim's evidentialist norm of presupposing.

5 It is an interesting question what the exact mechanisms are that bring about the contextual suspension (or cancellation) of the presuppositions of epistemic factives. But this is a matter for another occasion. What matters here is the fact that ‘knows p’, as a paradigm case of an epistemic factive, displays contextual suspensibility under the conditions described. For further discussion of this phenomenon and further references see Abrusán (Reference Abrusán2016).

6 Kim (Reference Kim2016: 240, fn. 215) claims to address examples such as HALSS* in Section 2.3.2 of his paper, but then doesn't.

7 I am indebted to two anonymous referees for this journal for providing fruitful feedback to earlier versions of this paper.

References

REFERENCES

Abbott, B. 2006. ‘Where Have Some of the Presuppositions Gone?’ In Birner, B. and Ward, G. (eds), Drawing the Boundaries of Meaning, pp. 1–20. Amsterdam: John Benjamins.Google Scholar
Abrusán, M. 2016. ‘Presupposition Cancellation: Explaining the ‘Soft-Hard’ Trigger Distinction.’ Natural Language Semantics, 24(2): 165–202.CrossRefGoogle Scholar
Brown, J. 2008. ‘The Knowledge Norm for Assertion.’ Philosophical Issues, 18(1): 89–103.CrossRefGoogle Scholar
Brown, J. 2010. ‘Knowledge and Assertion.’ Philosophy and Phenomenological Research, 81(3): 549–66.CrossRefGoogle Scholar
DeRose, K. 1992. ‘Contextualism and Knowledge Attributions.’ Philosophy and Phenomenological Research, 52: 913–29.CrossRefGoogle Scholar
DeRose, K. 2005. ‘The Ordinary Language Basis for Contextualism, and the New Invariantism.’ Philosophical Quarterly, 55(219): 172–98.CrossRefGoogle Scholar
Fantl, J. and McGrath, M. 2002. ‘Evidence, Pragmatics and Justification.’ Philosophical Review, 111(1): 67–94.CrossRefGoogle Scholar
Fantl, J. and McGrath, M. 2007. ‘On Pragmatic Encroachment in Epistemology.’ Philosophy and Phenomenological Research, 75: 558–89.CrossRefGoogle Scholar
Fantl, J. and McGrath, M. 2009. Knowledge in an Uncertain World. Oxford: Oxford University Press.CrossRefGoogle Scholar
Hawthorne, J. 2004. Knowledge and Lotteries. Oxford: Oxford University Press.Google Scholar
Karttunen, L. and Peters, S. 1979. ‘Conventional Implicatures.’ In Oh, C.-K. and Dinneen, D. A. (eds), Syntax and Semantics 11, Presupposition, pp. 1–56. New York, NY: Academic Press.Google Scholar
Kim, B. 2016. ‘In Defense of Subject-sensitive Invariantism.’ Episteme, 13(2): 233–51.CrossRefGoogle Scholar
Lackey, J. 2007. ‘Norms of Assertion.’ Noûs, 41(4): 594–626.CrossRefGoogle Scholar
Lackey, J. 2016. ‘Assertion and Expertise.’ Philosophy and Phenomenological Research, 92(2): 509–17.CrossRefGoogle Scholar
Stanley, J. 2005. Knowledge and Practical Interests. Oxford: Oxford University Press.CrossRefGoogle Scholar
Williamson, T. 2000. Knowledge and its Limits. Oxford: Oxford University Press.Google Scholar