Abstract
This paper considers the connections between semantic shiftiness (plasticity), epistemic safety and an epistemic theory of vagueness as presented and defended by Williamson (1996a, b, 1997a, b). Williamson explains ignorance of the precise intension of vague words as rooted in insensitivity to semantic shifts: one’s inability to detect small shifts in intension for a vague word results in a lack of knowledge of the word’s intension. Williamson’s explanation, however, falls short of accounting for ignorance of intension.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
The identification of vagueness with the capacity for borderline cases is contentious: see Weatherson (unpub-a, b) for considerations on criteria for identifying vague terms. I take this platitude from Williamson (1997b, p. 945).
There is a certain awkwardness in trying to formulate the relevant thought as we typically think of error in terms of beliefs and beliefs as relations to propositions. Williamson (1996b) glides over this complication:
Someone who asserts ‘Everyone with physical measurements m is thin’ is asserting a necessary truth, but he is still lucky to be speaking the truth. He does not know the truth of what he says. Although he could not have asserted the proposition he actually asserted without speaking truly, he could very easily have asserted a different and necessarily false proposition with the same words. (Williamson 1996b, p. 230)
I take it this means that the speaker does not know which proposition he is believing—the considerations should hold just as much for belief as they do for assertion. I prefer not to pick on this and just assume for now that a plausible way to express the safety condition can be found that meets the requirements for the account. I am interested in testing plasticity’s role in the explanation of vagueness.
Following Hawthorne (2006), we should distinguish semantic plasticity from a close cousin: the ease with which an object can fall in or out of the extension of a term at a world. A person who narrowly avoided falling off a cliff falls under the extension ‘didn’t fall off a cliff’ but could easily, in similar cases, have been in the extension of ‘fell off a cliff’. This isn’t sufficient for the predicate’ fell off a cliff’ to be plastic, since no implications about the shift in intension follow. My use of the term ‘plasticity’ differs from Hawthorne’s in one way: Hawthorne suggests that a term is plastic when “(a) slight differences in usage make for differences in semantic value and (b) we are insensitive to the ways in which difference in usage make for difference in semantic value” (Hawthorne 2006, p. 186). In effect, I am only requiring clause (a) as part of the definition of ‘plasticity’. This change is motivated: I am here trying to isolate the semantic component of plasticity from the epistemic.
At least, not vague on account of my ignorance of the relevant physics.
The less consistent issue is that the latter but not the former text includes methods used to make the judgment.
There is a bit of a tension here that I won’t exploit but that is worth thinking about. Williamson’s hypothesis about vagueness involves ignorance and lack of sensitivity to shifts in similar cases that makes one prone to error. This, however, presumably can vary person to person just as ignorance of a tree’s height can be subject to greater or smaller margins for error depending on the quality of one’s eyesight. Vagueness seems to be a property of a word relative to a community and so subject to community-wide features of use rather than individual uses.
For example, the function could associate nearly all the GFs with 20,000 ft tall for ‘tall’ but take a certain class of GFs and give them a highly sensitive pairing. This is improbable but not ruled out a priori.
I’m ignoring context sensitivity in the experiment for simplicity, and factors such as inflation that would plausibly result in people rich with some amount of money being poor with that same amount later. I assume this does no violence to the point I am trying to make.
The case suffers from a defect: I have not said anything about the effects of keeping the intension of ‘rich’ fixed on other related terms. One might think, for example, that one couldn’t do what Barney does to ‘rich’ without doing something similar to ‘poor’, since there are presumably valid entailment patterns from ‘x is rich’ to ‘x is not poor’. Perhaps apparent synonyms have to be similarly fixed (‘wealthy’ may be an example). I don’t think this complication spoils the point but it is worth keeping in mind. We could imagine a tribe with fewer words and fewer analytic connections between them.
Establishing vagueness is no easy task: one way to tell is by looking at whether or not there are borderline cases. It’s hard to see why the existence of borderline cases should depend on plasticitya. If we suddenly were to be governed by a semantic God like Barney and we became non-susceptible to instantiating GFs that shifted the intension of ‘rich’, it’s hard to believe that our margins for error governing ‘is rich’ would simply disappear.
We could, I suppose, imagine Barney simply telling her. But I take it that in our case we could imagine an omniscient being telling us what amount of money divides the rich and the poor and that we would be immune from the effects of plasticitya were this to be the case. Presumably we need to focus on methods where plasticitya could plausibly make the difference. I’m not entirely sure how to characterize this class of methods.
Thanks very much to Daniel Nolan for invaluable discussion of this point.
See Lewis (1997). For what it’s worth, I always have been a little puzzled by the weight philosophers put on the intuition that finkishly protected vases are fragile. If I owned several of those vases and wanted to sell them, I would presumably advertise their non-fragility as a key selling point: I’d bounce them around in front of putative customers to demonstrate it.
Thanks to John Hawthorne for discussion of this point.
Thanks to Dave Chalmers, who suggested this to me in conversation.
There are other explanations as well. Cody Gilmore suggested (p.c.) that the semantic laws might prefer to assign extension of at least 200 people to predicates. Take a world where the line between the tall and the not-tall was 6′7 and there were exactly 200 people that were 6′7, none taller. In this case, the line would at least be protected from shifting downward at nearby worlds where all else was held fixed (the number of 6′7 people). This hardly makes a case for 6′7 cutting out a natural line despite the reduction in plasticity.
These results dovetail nicely with those of Hawthorne (2006), where it is argued that plasticitya has fairly devastating results if it leads to ignorance in all cases. After all, it looks like such a priori-isms as vague instances of the T-schema become unknown if semantic terms are allowed to be plastic. Suppose that ‘true’ is plastic in the way envisioned. If so, then:
‘a is bald’ is true iff a is bald.
are unknown, since in nearby worlds the sentence ‘a is bald’ does not fall under the extension of ‘is true’ but easily could have since ‘is true’ would have a slightly different intension than it does at our world. This means that if ‘is true’ is vague and plastica we aren’t safe with respect to the T-schema since at nearby worlds we would express a falsehood by uttering its counterpart. ‘is true’, however, is clearly vague and so an alternative explanation (in Hawthorne’s case, of our knowledge rather than our ignorance) is required.
See Kearns and Magidor (2008) who argue that there are no plausible conditions governing knowledge that mix safety and plasticity that result in the ignorance characteristic of vagueness.
References
Hawthorne, J. (2006). Epistemicism and semantic plasticity. In J. Hawthorne (Ed.), Metaphysical essays (pp. 185–210). Oxford: Oxford University Press.
Kearns, S., & Magidor, O. (2008). Epistemicism about vagueness and meta-linguistic safety. Philosophical Perspectives, 22, 277–304.
Lewis, D. (1997). Finkish dispositions. The Philosophical Quarterly, 47(187), 143–158.
Williamson, T. (1996a). Wright on the epistemic conception of vagueness. Analysis, 56(1), 39–45.
Williamson, T. (1996b). Vagueness. London: Routledge.
Williamson, T. (1997a). Reply to commentators. Philosophy and Phenomenological Research, 57(4), 945–953.
Williamson, T. (1997b). Reply to commentators. Philosophical Issues, 8, 255–265.
Williamson, T. (2000). Knowledge and its limits. Oxford: Oxford University Press.
Weatherson, B. (unpub-a). Three objections to smith on vagueness.
Weatherson, B. (unpub-b). Vagueness without toleration: Reply to Greenbough.
Acknowledgments
Thanks to David Chalmers, Sam Cumming, Greg D’amico, Michael Glanzberg, Robbie Hirsch, Mandana Kamanger, Ernest Lepore, Timothy Williamson and an audience at Rutgers University for comments and advice on various parts of this paper. Extra gratitude is due to Stewart Cohen, Cody Gilmore, John Hawthorne, Daniel Nolan and Jonathan Weisberg for providing excellent advice and suggestions on various drafts of this paper.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Sennet, A. Semantic plasticity and epistemicism. Philos Stud 161, 273–285 (2012). https://doi.org/10.1007/s11098-011-9734-1
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11098-011-9734-1