"I am not here now, and I miss you": Written Messages and Indexical Reference

Both indexicals and written inscription pose interesting problems to philosophers of language. In this paper I am concerned with two things. Firstly I will explore an issue philosophers often neglect, that is, the differences between spoken or verbal utterances and written inscriptions. Secondly, and related to the previous issue, I will be concerned with indexicals, especially written instances of indexicals. All of this will stem from an examination of how the utterance “I am not here now” may be uttered truly. In the end, I will arrive upon a theory of how to evaluate written messages which proves to be particularly useful in instances of messages containing indexical.

Predelli, at the start of his paper “I am not here now,” briefly outlines a theory of how to evaluate sentences containing indexicals that is basically Kaplan's. Without wading through the complexities of the argument that Kaplan makes in his paper “Demonstratives,” let us put forth Kaplan's general view as pertains to our present purpose thusly: The content that an utterance u of expression e expresses is the content that the character of e yields when we “plug in” the context that pertains to the utterance u. If we simplify and apply this to a relevant example, that is, an expression containing an indexical—such as “I am not here now”—we see that, to Kaplan, “now” refers to t in a situation s if and only if s is the situation in which “now” is uttered, and t is the time of s.

We find that Kaplan's line of reasoning supports our intuitive assumptions about indexicals. If we take the utterance “I am not here now” and evaluate it along Kaplan's line of reasoning, we arrive at the conclusion that the indexicals “I,” “here,” and “now” refer, respectively, to the utterer, the place of utterance, and the time of utterance. It seems to follow from this that “I am not here now” may never be uttered truly, for logical reasons. But, as many have noted, we can find true instances of “I am not here now,” such as written messages. Kaplan's theory does not account for this problem. Predelli and others have offered their own solutions to the problems that written messages pose to the way we usually evaluate sentences containing indexicals. In the first portion of this paper I intend to first discuss some of these solutions and why they are insufficient, then to offer my own solution to this problem.

In his aforementioned paper, Predelli first considers a theory of Sidelle's. According to Sidelle, past theories have failed in solving the problem we are presently concerned with because they confuse what the referents of the indexicals in a given expression actually are. Sidelle claims that, in the case of the written message “I am not here now,” the indexicals “I,” here,” and “now” refer, respectively, to the encoder, the place of decoding, and the time of decoding. When one writes a message, Sidelle claims, one is actually deferring an utterance. That is, one is preparing to make an utterance at a later time. This utterance, which occurs at a later time, Sidelle calls the “genuine” utterance. Predelli challenges Sidelle's theory by offering the following scenario: Jones, expecting his wife home at six o'clock, writes a note at four o'clock which says “I am not here now,” with the intention of informing Mrs. Jones that he is away from home at six o'clock. However, Mrs. Jones is late arriving home, and does not read the message until ten o'clock. In this case, Predelli claims, “now” does not refer to the time of decoding, but must refer to the intended time of decoding. Indeed, Predelli spends the latter half of his paper expounding his belief that what is most important in evaluating written instances of sentences containing indexicals is the utterer's intention.

“Written and recorded messages,” Predelli concludes, “are to be evaluated with respect to the intended context of interpretation, which need not coincide with the context of utterance.” This conclusion is certainly unacceptable. If we follow Predelli's line of reasoning, we arrive at the outrageous conclusion that the referent of an utterance of “now” is whatever the utterer intends it to be. We simply cannot allow for a theory of reference to permit this kind of arbitrary assignation. If I were to put on my door the message “I am not here now” with the intention of my roommate seeing it at three o'clock, but she does not see it until five o'clock, the message is as true—if I am still not at home—at five o'clock as it was at three o'clock, because “now” does not refer to my intended context of interpretation. Any theory that suggests that for any utterance of u, u refers to whatever the utterer intends u to refer to must be rejected, as it makes language, in the end, arbitrary. We will have cause to return to the matter of utterer intention later. First, though, let us approach this problem a different way.

For reasons that will become clear, out of the above theories, Sidelle's gets us the furthest in our understanding of written instances of sentences containing indexicals. But his theory, like Kaplan's and Predelli's, fails, in part, because it fails to consider fully the differences between verbal utterances and written inscriptions. Written inscriptions are different than verbal utterances in that written inscriptions have the unique quality of existing over time. Verbal utterances occur at an isolated point in time. It is easy to conclude that when one verbally utters “I am not here now,” “now” refers to the time of utterance. Written inscriptions, however, do not exist only at an isolated point in time. But let us examine another aspect of verbal utterances and written inscriptions before we further explore the differences between the two.

It is uncontroversial that the specific referents of indexicals vary depending on the context or situation in which they are uttered. Furthermore, for each utterance of a sentence containing indexicals—and even those that do not—there is a set of possible contexts in which the utterance is true. By virtue of having uttered “I am not here now,” one establishes the set of possible contexts in which the utterance is true. That is, those contexts in which the referent of “I” is not at the referent of “here” at the referent of “now.” The set of possible contexts in which an utterance is true, can be found in the syntax of the utterance, and is often not necessarily specific. Establishing the set of contexts in which “I am not here now” is true, does not actually tell us much, such as, what the reference of words are. It simply tells us that, regardless of what “I” refers to, it must not be at whatever “here” refers to at whatever “now” refers to. The sentence “I am not here now” may never be verbally uttered truly—and let us ignore instances of quotation or imitation—because the actual context in which it is uttered is never a part of the set of contexts in which the utterance is true.

Written inscriptions differ from verbal utterances in the following way: written inscriptions have an initial utterance, that is, when they are initially written, but they are then uttered continually until they are physically destroyed; the inscriptions themselves become proxy utterers. When one writes “I am not here now” it is initially false, as the referent of “I” is at the referent of “here” in order to write the note in the first place. But, when one does write this, one establishes the set of possible contexts in which it is true. When one leaves the message, the referent of “I” is still the initial utterer, the referent of “here” is the place of utterance, and the referent of “now” is the time of utterance. When we consider that a note, as a sort of proxy utterer, continually utters the sentence of the initial utterance, we see that the place of utterance is the place of the note, and the time of utterance constantly varies. Note that above I did not say that the note constantly utters the initial utterance, but that the note constantly utters the sentence of which the initial utterance is a token. This is an important distinction, and perhaps not obvious nor intuitive. Of course, the syntax of the written message does not change. But, the note utters not the same utterance continually, but a constant “stream” of new utterance. Of course when the referents of the words of an utterance change, a different proposition is expressed, and thus is a different utterance. As we saw, “now,” in this case, refers to the time of utterance and time is constantly variable. So, we see that the note utters a different utterance each instant, as the referent of “now” constantly changes, and thus a new proposition is expressed by the utterance which is distinct from that expressed by the previous utterance. This leads to the perhaps noncontroversial yet still important claim that one utterance expresses exactly one proposition or none at all (such as in cases where utterances contain words with no referents). This is a claim I will have cause to defend more as I further develop my argument.

So far, we have considered a theory of how to evaluate written messages which seems to deal with the problems posed by indexicals better than the alternatives offered by Kaplan, Predelli, and Sidelle. Kaplan's theory was insufficient because it fails to recognize contexts of utterances in which the actual human body is absent at time of utterance. Predelli's theory which claims that utterer intention determines reference of indexicals carries some very unattractive implications and I have illustrated thus far—and will illustrate further later on—that intention does not have anything to do with indexical reference at all. Sidelle's idea of utterance deferral was a step in the right direction but is ultimately not sufficient in explaining written instances of indexicals, such as “I am not here now.” My examination of this single sentence has lead to a theory which presently seems more appealing than those just mentioned. My development of this theory has led us to some general statements about language which will be important in the remainder of this paper: firstly, utterer intention does not determine (nor even influence) indexical reference, and, secondly, one utterance expresses no more than one proposition. But it seems my working theory of written messages—and all of its implications—ought to be tested more rigorously in order to make sure it is a considerable theory and not just a clever way to explain the sentence “I am not here now.” I intend to devote the rest of this paper to a more in-depth exploration of my theory of the evaluation of written inscriptions (especially written inscriptions containing indexicals). I will begin this exploration with an attempt to explain other instances of written inscriptions containing indexicals within the framework I have heretofore established.

Consider the sentence “I miss you.” Of course “I miss you” does not pose the exact problem “I am not here now” did. In our consideration of “I am not here now” we first had to determine how it is possible to even utter the sentence truthfully at all. We may by-pass this first step in considering “I miss you.” Regardless, we ought to be able to evaluate the sentence along the same lines as the one in the first part of this paper. This requires us to do several things, as we have seen. To evaluate the truth value of an utterance of “I miss you,” we must first do two things: 1) determine what the set of possible contexts in which the utterance is true is, then 2) determine if the actual context is part of the set of possible contexts in which it is true. Before we even consider some examples, we can determine what the set of possible contexts in which utterances of the sentence “I miss you” is: all of those in which the referent of “I” has the relation of missing the referent of “you.” Of course when we consider actual examples of utterances of this sentence, we will need to decide what the referents of “I” and “you” are.

Consider that Smith is going out of town and expects he will miss his partner, who is staying home for the weekend. On Friday, Smith is at home getting ready to leave for the airport and his partner is still at work and will not be home until later that evening. Smith, before walking out of the door writes the message “I miss you,” on a piece of paper and leaves it on the counter top. Now let's say that Smith's mother decides she is going to surprise her son by coming to his house for the weekend. She shows up, not knowing that Smith is going out of town, and of course no one is home. She does, however, see the note on the counter, and she is touched. She thinks, for whatever reason, that her son must have known she was coming, and so left her the note, anticipating not being at home when she arrived. So Smith’s mother stays for the weekend with Smiths partner and when Smith finally comes home his mother shows him the note and says “do you miss me? Is this true?” Consider that Smith does happen to miss his mother, whom he never gets a chance to see, surely he would say something like “Well, I actually didn't intend for you to read that note, I didn't know you were coming, but yes it is true, I do miss you.” Of course the claim I am trying to get at right away is that utterer intention does not determine indexical reference in this case, just as it did not in the case we considered in the first part of this paper. In this scenario, and others we will examine, the referent of “you” turns out to be the decoder, or reader, of the utterance. Of course an utterer often does have in mind something he intends for “you” to refer to when he writes the message “I miss you,” and the note is usually false unless the utterer’s intended referent of “you” is the one doing the decoding. So it is tempting to conclude that the referent of “you” is determined by utterer intention. But this is untrue. It is merely coincidence that the utterance is usually false unless the decoder is the utterers intended referent of “you.” Of course when Smith's mother shows him the note and asks him “Is this true?” he would not say “No,” and scribble a new note that says “I miss you,” and, giving the new note to her—the one he intended for her to read—say “But this is.”

When Smith first wrote the note “I miss you” he began a “stream” of constant utterance. That is, there was an initial utterance—when Smith first wrote the note—but then the note itself became a proxy utterer for Smith, continually uttering tokens of the sentence “I miss you.” As we saw with “I am not here now,” the note does not continually utter the same utterance, but, because referents of indexicals are variable, and this variability causes variation in the proposition expressed by an utterance, the note, all though the syntax does not change, sometimes utters a different utterance. “I miss you” differs from “I am not here now” in the following way: the proposition expressed by “I am not here now” is continually variable, as the referent of “now” is continually variable. Each new instant, the note that says “I am not here now” utters an utterance distinct from the previous one and expresses a proposition distinct from the previous one. “I miss you” is not quite as variable. Rather, the referents “I” and “you” are not as continually variable as such an indexical as “now.” We saw in the above scenario that the referent of “you” happens to be the decoder of the note. So, the note depends on a reader to change the proposition that is expresses. Each time the note is read by a new person, the referent of “you” changes and thus the proposition expressed changes. If we consider the note when no one is reading it (and I am not sure we really need to) we see that when no one is reading it, it fails to express a proposition at all, as one of the words, “you,” has no reference.

Let us consider another scenario which varies slightly from the first one we looked at. As before, Smith is going out of town and leaves a note for his partner that says “I miss you.” Again, Smith's mother decides to drop in for a surprise visit but finds nobody there, so she makes herself at home until Smith's partner arrives. It just so happens that Smith's mother does not notice the note on the counter before Smith's partner comes home. When he does come home though, Smith's mother comes into the kitchen to greet him and they both see the note at the exact same time. Per what we have already illustrated, the note, when read by Smith's partner expresses the proposition “I miss you” where the referent of “you” is he, Smith's partner. When Smith’s mother reads the note the proposition it expresses is “I miss you” where the referent of “you” is she, Smith's mother. This seems to perhaps conflict with one of the conclusions we arrived upon in our consideration of “I am not here now,” that is, that one utterance may express one proposition. In this current scenario, a single utterance seems to express two propositions at the same time. Surely if one utterance did express two propositions at the same time, it would create some holes in my present argument. This is not what occurs in the scenario in which Smith's mother and Smith's partner read the same note at the same time though. The very act of one reading the note “I miss you” creates a new utterance of the sentence written on the piece of paper. When one reads the note, the referent of the word “you” changes. This is difficult to come to terms with perhaps because we are not used to thinking of syntactically identical utterances occurring in almost identical contexts as separate utterances. We can think of this in the following way: When Smith's partner reads the note (at exactly the same time as Smith's mother) the utterance the note makes is this “I miss you1” and the utterance the note makes when Smith's mother reads it is “I miss you2.” This is not to initiate some ontological debate about what entities are actually present in the syntax, but to show that change of reference entails a change of utterance. And though human utterers seem not to be able to make two separate utterances instantaneously, we have no reason to hold that notes, as proxy utterers cannot do so.

So it seems the framework I developed in the first part of this paper is not merely a clever way of explaining written instances of “I am not here now” but succeeds in explaining other similar instances of written indexicals. Our methodology has remained useful: In evaluating written inscriptions, we must determine two things: what the set of possible contexts in which the utterance(s) that the inscription makes is true is, and whether or not the actual context is part of that set. My theory of written inscriptions as proxy utters has proved useful in explaining how the methodology above succeeds completely in cases of written inscriptions that contain indexicals. Further, this methodology also allows us to preserve the general conclusions we have arrived upon about language, specifically, that one utterance may express one proposition, and that utterer intention is irrelevant to indexical reference.