Abstract
Recording devices are generally taken to present problems for the standard Kaplanian semantics for indexicals. In this paper, I argue that the remote utterance view offers the best way for the Kaplanian semantics to handle the recalcitrant data that comes from the use of recording devices. Following Sidelle (1991) I argue that recording devices allow agents to perform utterances at a distance. Using the essential, but widely ignored, distinction between tokens and utterances, I develop the view beyond the initial sketch given by Sidelle, and I answer the main objections raised against the view. The paper is structured as follows. Section 1 gives a succinct presentation of Kaplanian semantics and of the problem raised by the use recording devices, Section 2 presents the remote utterance view and Section 3 answers the objections put forward against the view and further develops it. I conclude that the remote utterance view can handle the data that comes from the use of recording devices with only modest modifications of the Kaplanian semantics.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
For example, the English sentence with the audible, or surface, form “Mary saw the boy with binoculars” will be matched with two sentences in the model:
(s1) [MaryN [sawV[theDET[boyCN[with binocularsAdjP]NP]NP]VP]S
(s2) [MaryN [[[sawV[theDET [boyCN]NP]VP[with binoculars]AdvP]S.
Kaplan was moved to make this claim by the observation that when embedded under modal, temporal or locational operators, indexicals do not shift their reference. This can be seen in sentences like: “It is possible that in Pakistan, in five years, only those who are actually here now will be envied” where the world, location and time referred to by “actually”, “here” and “now” are the world, time and location of the context of utterance, and not those determined by the modal, temporal and locational operators (Kaplan 1989: 499).
All these responses accept the data (i.e. the intuitions concerning the truth-values of (2) when used in recorded messages) as semantically relevant. Another type of response denies that the data is semantically relevant and tries to explain away the intuitions by appeal to pragmatic effects. According to this type of answer, the intuition that (2) is true when the message is played back is due to what is conveyed by that use of (2) and not because its propositional content is true at that context of use. On the contrary, when the message is played back, (2) is false even though the agent is not present at the time and place of the play back. It must be, then, that when played back (2) doesn’t expresses the content that the agent is present at the time and place of the playback. Rather, the content expressed is the false content that agent is not present at the time and place of recording. As Cohen and Michaelson 2013: 581-582) convincingly argue, the pragmatic answer fails for several reasons: (i) unlike conversational implicatures, answering machine recordings of (2) cannot be canceled as shown by the infelicity of “I am not here now, but I might be when you are listening to this.”; (ii) When (2) is played back on an answering machine, “understanding that the ‘here’ and ‘now’ refer to the place and time of playback is entirely non-optional, in that if one fails to interpret these indexicals in this way, one fails” to understand (2) Michaelson (2014: 535-536). There is a further reason to set aside the pragmatic response. The response assumes that when one records (2) on the answering machine, one performs an utterance. I’ll argue in this paper that this is not the case.
This paper has a rather limited scope: that of showing that Kaplanian semantics can handle the answering machine puzzle. Ever since the publication of Demonstratives many other objections, or proposed revisions, have been put forward for Kaplanian semantics. It is beyond the scope of this paper to deal with all of them. One other phenomenon that some authors take to require a revision of Kaplanian semantics, and is often discussed in conjuction with the answering machine puzzle, is the historical present: the use of “now” when narrating past events in order to refer to a moment in the past (e.g. as when during a history class about Napoleonic wars the professor says “Now, Napoleon prepares to cross the Niemen river into Russia”, “now” doesn’t refer to the time of the utterance, but it refers to 1812.) As Predelli (2005: 53) points out, the historical present is distinct from the challenge posed by recorded messages, but both are taken to motivate the same solution. This is debatable, though. Cohen (2013) and Corazza (2004) convincingly argue, in my opinion that occurrences of “now” in the historical present are not indexicals at all but bound variables. In the above sentence it is plausible that “now” is bound to the time of Napoleon’s crossing of Niemen by an immediately preceding discourse fragment. Kaplan (1989:489-490) was well aware that words like “he”, “now”, “here”, etc. have both an indexical use and a non-indexical use, and he explicitly limited his semantics to their indexical use. This fits well with the data presented in Nunberg (1993) that expressions like “that”, “he”, “here”, “now” can be used indexically, anaphorically and as bound variable.
Sidelle (1991:535) gives this as an example of an action performed at a distance: one plants a bomb which much later on, when the agent is thousands of miles away, is detonated. The detonation is an action she performs at a distance.
Or, to borrow an example from Searle (1978:209), consider the case of someone who holds up the same sign inscribed with “Stop here” on several occasions. Each time she holds up the sign she utters the sentence “Stop here” and she does it with the very same token.
The distinction between tokens as concrete particulars composed of ink, pixels, sound waves, hand movements, etc. and utterances as actions executing communicative intentions is found in other places in the philosophical and linguistic literature: Searle (1978:209), Bromberger ( Bromberger 1992;191-193), Millikan (2012: 221), Perry 2001: 38-39), O’Madagain (2014: 75).
As Sidelle (1991: 553) puts it “If you lie when you are home, you must be telling the truth when you are not. It can’t be because one is at home at the time of the machine; one can use the machine to utter “I am not here now”. One being home at the time of the call simply provides one with the opportunity to use the sentence to lie”.
What counts as a tokening (i.e. use of the token) is less clear when we deal with written messages (like post-notes) than it is when we deal with answering machine. There seem to be a consensus, though, that each instance of the post-note being read constitutes a tokening (i.e. use of the token). See Michaelson (2014: 251-253), Cohen (2014: 24)
Michaelson (2014: 531) does raise this question for the remote utterance view.
Corazza et al. (2004) argue that there are conventions associated with each type of recording device which determine whether (5) is to be interpreted at the context of encoding or at the context of decoding (i.e. context of reading, hearing, etc): if (5) is heard on an answering machine, then the relevant time and place for its interpretation are the time and place of the playback; if it is read on a postcard or letter the relevant time and place are those when the postcard or letter was written. But this is doubtful. At best certain usages are more common than others. And certainly the remote utterance view need not adhere to it. For sure, both answering machines and postcards/letters can be used in ways that go against the conventions stipulated by Corazza et all, and these uses are neither linguistically nor socially deviant. Imagine that one records the following sentences on an answering machine: “I decided to give up everything. I’m signing the donation agreements right now. From now on this house no longer belongs to me”. The time relevant for interpreting these sentences is the time of the recording and not the time of decoding. As far as letters are concerned, imagine that one writes the following in a letter: “You are about to read my side of the story. Today, you’ll finally understand why I decided to leave”. The time relevant for interpreting these sentences is the time of the decoding/reading and not the time of the encoding/writing. Moreover, people can successfully communicate with the help of novel recording devices which are not governed by social conventions, as when one video records their will. To borrow an example from Sherman (2015), one can video record either “Today I met with my lawyer to go over all the details before making this videotape” and “Today you all received a phone call telling you to come to my lawyer’s office.” The time relevant for interpreting the first sentence is the time of the recording, while the time relevant for interpreting the second sentence is the time of the decoding. The remote utterance view can easily account for this and it doesn’t need to appeal to conventions. Both sentences recorded on the video will are evaluated at their respective contexts of utterance. Video wills, like all recording devices, can be used both to record and to perform utterances. When one video records “Today I met with my lawyer to go over all the details before making this videotape” on her will she records her utterance for a future audience. When one records “Today you all received a phone call telling you to come to my lawyer’s office” on her video will, she makes preparations so to utter the sentence when the video is being played to a future audience.
I should point out that not all authors accept Predelli’s intuitions. For example, Cohen (2013) whose position is that recorded expressions are semantically interpreted always at the context of decoding (i.e. context of reading, hearing, etc) claims that the indexical “now” in the note refers to 8 P.M. and not to 5 P.M. I take this to be a serious drawback of Cohen’s position.
See Cohen and Michaelson (2014) for an overview.
To make his case Predelli must show that it is metaphysically impossible for John to utter (6) at his home at 5PM by means of the note, and that yet we have intuitions about the truth-value of such an impossible utterance. But surely there are possible worlds in which John does utter (6) at 5PM at his home by means of the note: those in which someone is home at 5PM and reads the note.
One immediate consequence of giving up the restriction of semantic evaluation to proper context is that the logic of indexicals developed by Kaplan (1989) must be slightly rewritten. But Stojanovic (2011) argues convincingly there are independent reasons for a revision of Kaplanian logic of indexicals, one in which (2) doesn’t come true at every context of utterance.
References
Bianchi, C. (2008). Indexicals, Speech Acts and Pornography. Analysis, 68(4), 310–316.
Bromberger, S. (1992). On What We Know We Don't Know: Explanation, Theory, Linguistics, and How Questions Shape Them. Chicago: University of Chicago Press.
Cohen, J. (2013). Indexicality and the Puzzle of the Answering Machine. Journal of Philosophy, 110(1), 5–32.
Cohen, J., & Michaelson, E. (2013). Indexicality and the Answering Machine Paradox. Philosophy Compass, 8(6), 580–592.
Corazza, E. (2004). On the alleged ambiguity of 'now' and 'here'. Synthese, 138(2), 289–313.
Corazza, E., Fish, W., & Gorvett, J. (2004). Who Is I? Philosophical Studies, 107(1), 1–21.
Kaplan, D. (1989). Demonstratives: An essay on the semantics, logic, metaphysics, and epistemology of demonstratives and other indexicals. In J. Almog, J. Perry, & H. Wettstein (Eds.), Themes from Kaplan (pp. 565–614). Oxford: Oxford University Press.
Kölbel, M. (2009). The evidence for relativism. Synthese, 166(2), 375–395.
Martí, G. (2012). Empirical Data and the Theory of Reference. In P. Kabasenche, M. O'Rourke, & M. Slater (Eds.), Reference and Referring. Topics in Contemporary Philosophy (pp. 63–82). Cambridge: MIT Press.
Michaelson, E. (2014). Shifty characters. Philosophical Studies, 167(3), 519–540.
Millikan, R. (2012). Are there mental indexicals and demonstratives? Philosophical Perspectives, 26(1), 217–234.
Nunberg, G. (1993). Indexicality and deixis. Linguistics and Philosophy, 16(1), 1–43.
O'Madagain, C. (2014). Indexicals and the Metaphysics of Semantic Tokens: When Shapes and Sounds become Utterances. Thought: A Journal of Philosophy, 3(1), 71–79.
Perry, J. (2001). Reference and Reflexivity. Stanford: CSLI Publications.
Predelli, S. (1998a). I am not here now. Analysis, 58(2), 107–115.
Predelli, S. (1998b). Utterance interpretation and the logic of indexicals. Mind and Language, 13(3), 400–414.
Predelli, S. (2005). Contexts: Meaning, Truth, and the Use of Language. Oxford: Oxford University Press.
Predelli, S. (2011). I am still not here now. Erkenntnis, 74(3), 289–303.
Recanati, F. (2005). Indexicality, Context, and Pretense: A Speech-act theoretic Account. In N. Burton-Roberts (Ed.), Advances in Pragmatics (pp. 213–229). Basingstoke: Palgrave-Macmillan.
Searle, J. (1978). Literal meaning. Erkenntnis, 13(1), 207–224.
Searle, J. (1986). Meaning, communication, and representation. In R. Grandy & R. Warner (Eds.), Philosophical Grounds of Rationality: Intentions, Categories, Ends (pp. 209–226). Oxford: Oxford University Press.
Sherman, B. (2015). Constructing Contexts. Ergo, an Open Access Journal of Philosophy, 2(23), 581–605.
Sidelle, A. (1991). The Answering Machine Paradox. Canadian Journal of Philosophy, 21(4), 525–539.
Smith, Q. (1989). The Multiple Uses of Indexicals. Synthese, 78, 167–191.
Stojanovic, I. (2011). Context, Meaning and Logical Truth. In A. Gupta & J. van Benthem (Eds.), Logic and Philosophy Today, Studies in Logic Series, vol. 30 (pp. 11–24). London: College Publications.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Briciu, A. Indexicals in Remote Utterances. Philosophia 46, 39–55 (2018). https://doi.org/10.1007/s11406-017-9909-x
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11406-017-9909-x