# Why is "then" deictic?

Then is mentioned as deictic in many papers but I couldn't find a sufficient explanation for that. Every example I could think of involves then acting as anaphora, but how decctic? If I say:

• I am going to eat at 1pm, I’ll see you then,

then is the anaphora and at 1 pm is the deictic.
I would like to see a reference too, if possible.

• "Then" can refer deictically to a time in the very recent past, for example: A: Did you hear a scream? B: When? A: Just then. Or if I was watching a film or video, or looking at a photo depicting past events I might say We were happier then. Commented Jun 14, 2016 at 11:48
• The term you're using is spelled deictic, the adjectival form of deixis. The Wikipedia article may be helpful. Commented Jun 14, 2016 at 13:58
• As BillJ points out, the deictic uses of now//then (cf. here/there) are 'demonstrative'. Since the 4th dimension is relativistically compressed into just 'now', it takes an abstract representation (like a memory or a photo) to let us point to 'then'.
– amI
Commented Jun 15, 2016 at 19:57
• Read Fillmore's Deixis Lectures to see what's deictic and what's not. Commented Jun 16, 2016 at 2:13
• The anaphora/cataphora distinction is independent from the concept of deixis. Commented Jun 16, 2016 at 18:27

It is fairly possible to make an account of the anaphoric words as deictic (or indexical; I consider deixis and indicality plainly synonymous). I personally share this approach. In my formulation (published here in Italian; summed up here in English), being deictic, for a word, means that the context disambiguates it deterministically.

Such words as horse, dog, table etc. are not deictic since they might denote whatever real object the speaker decides to denote thereby. This means that the hearer has also the "right" to misunderstand, or not to know, what the speaker is speaking about. There is always a bit of ambiguity in such words (what horse/dog/table was effectively intended?).

On the other hand, when we use such words as I, now, yours, here etc. we don't admit any ambiguity. The hearer does not have the right not to know who is meant by an utterance of the word I (and other deictics). That is, the rules of language do not let us ignore the meaning and the reference of the deictic words, which are, therefore, completely unambiguous.

Why are they unambiguous? Because they rely on the elements of the communication circuite which the speakers cannot avoid knowing. In other words, when we are engaged in a conversation, we have to know who is speaking, where and when does the conversation occur, and what was said so far (plus some other elements that I don't mention here). All the language apparently have means of referring to such obligatorily known elements. Pronouns denote the speakers, the so-called "shifters" denote space and time, and anaphoric words denote the bulk of information communicated in the preceding co-text and considered already disambiguated.

To sum up, if we explain deixis as a reference to an obligatory element of the communication circuit, then anaphors are plainly deictic since they refer to something that has been said in the previous utterances, and the speakers are supposed to have assimilated, if they want to continue speaking.

Therefore, when you say then you mean "at the times has been mentioned right now, and you cannot avoid remembering". The reference of this adverb is rigid, not unlike the reference of the word I, whoever utters it in a discourse.

A similar approach is to be found also in a paper by Fr. Recanati.