The statement "I am here now" seems to have many peculiar properties that I am not sure how to express. It appears analytically and equally true for any speaker anywhere at anytime. Yet its truth is also contingent upon a particular, physical speaker in all cases.

It appears altered when spoken through a telephone or on a recording, and obviously false when written down. It is also technically false or unprovable according to Heisenberg's uncertainty principle, insofar as "here now" cannot be physically determined. Its truth seems thus to require some physical fact, yet cannot be determined as such, hence false.

I am also not sure about its Kantian status, since time and space are intuitions constitutive of the transcendental subject yet "prior to" any possible experience of a "self" in the physical sense, I believe.

Does this statement occupy some peculiar niche or limit in logic? Is it a fallacy of self-reference? Is it an antinomy of reason? Is it entirely redundant, in that "am here now" is not a proper predicate?

  • I changed your title to match the question in the body ("statement" instead of "proposition"). The reason for this is that, with indexicals, a statement expresses different propositions in different contexts. By contrast, non-indexical statements express the same proposition across contexts (or, at least, the degree of variation is much less; you might want to allow some variation of proposition expressed due to considerations from pragmatics). – Dennis Oct 14 '15 at 17:40
  • Russell distinguished knowledge by acquaintance and knowledge by description:"We shall say that we have acquaintance with anything of which we are directly aware, without the intermediary of any process of inference or any knowledge of truths". The only knowledge by acquaintance according to Russell is expressed by "I", "this", "here" and possibly "now", everything else is by description only. plato.stanford.edu/entries/knowledge-acquaindescrip – Conifold Oct 14 '15 at 19:28

The sentence "I am here now" does have some peculiar properties, at least insofar as it contains indexical expressions, which themselves are somewhat peculiar. Within the philosophy of language, the proper semantic treatment of indexicals is an area of live debate.

The Basics

First off, what's an indexical expression? An indexical expression is, roughly, an expression whose reference can shift across contexts. Your sentence contains three such expressions: "I", "here", and "now". How does reference shift across contexts? Well, consider what the referent of "I" would be when you utter "I am here now" (hint: it's you). When I utter the same sentence, the reference of "I" shift to me. In usual circumstances (e.g., not in indirect discourse), "I" refers to the speaker.

We can see that this is an example of the reference of "I" shifting across contexts, where the role context plays is to specify the speaker of the sentence. This is why "I" is an indexical.

"Here" works similarly, except the thing referred to is (usually) the location of the speaker (fixed by the context). "Now" as well, with the thing referred to being the time of utterance (fixed by the context). David Lewis famously proposed an indexical analysis of "is actual", claiming that it functioned like "here", except that the location specified is the possible world inhabited by that individual (rather than a region within a possible world occupied by an individual).

Formally this all gets modeled using centered possible worlds. These were popularized (first discussed?) by David Lewis in his 1979 "Attitudes De Dicto and De Se". Centered worlds are possible worlds which come with privileged individuals and times. You can see why they'd be useful: context supplies some centered world as the world relative to which the utterance is to be evaluated (the "evaluation world") and this centered world privileges certain individuals and times in a way that makes them the default referents of indexical expressions (in that context). Location gets fixed because the privileged individual (assuming they aren't scattered individuals) occupies a location at the time of utterance in a way that makes them the default referents of indexicals (in that context).

So, back to your sentence "I am here now". When I utter that sentence, context selects a centered world where the privileged individual is me, the privileged location is where I am at the time of utterance, and the privileged time is when I am at the time of utterance (which is just a convoluted way of saying "it's the time of utterance"). When you utter the sentence the same thing happens, mutatis mutandis.

Contrast this with a sentence like "Dennis was in Massachusetts on October 14th, 2015". This sentence is true, but note that the reference of the terms "Dennis", "Massachusetts", and "October 14th, 2015" is (plausibly) the same no matter the context or who utters it (this, strictly speaking, depends on some plausible but controversial assumptions about proper names, e.g., that they're "rigid designators"). To evaluate it, you need only specify a possible world (by default, the actual world), and so you don't need the privileged individuals/times/locations supplied by centered worlds(/contexts).

The Answer to Your Question

So, to answer your question, what's peculiar with the sentence "I am here now" is that the the worlds relative to which you evaluate the utterance are centered worlds as opposed to the more familiar possible worlds of modal logic and formal semantics in the style of Heim and Kratzer (obviously excluding the sections on interpreting indexical expressions).

Further Reading

  1. Indexicals: The Stanford Encyclopedia of Philosophy has (as usual) a great article on the subject of indexicals. The locus classicus here is David Kaplan's "Demonstratives". Kaplan's Theory of Indexicals is covered in the SEP article, but you can also read about it in the Internet Encyclopedia of Philosophy's entry on Demonstratives and Indexicals.

  2. Possible Worlds in Formal Semantics: the standard text in formal semantics is Irene Heim and Angelika Kratzer's "Semantics in Generative Grammar" (copies of which can be found all over the internet). That's a good general reference on all of these topics.


The sentence is not a huge problem to treat logically, but it does require a stipulated way of treating indexicals to ensure that it will behave logically the way it seems it should (to be true whenever uttered, yet contingently so). So, contrary to what others have said, it is a sentence that holds some interest in logic as a paradigm of an indexical sentence whose propositional content varies across circumstances and yet remains true whenever uttered.

For example, Kaplan's influential treatment of indexicals and demonstratives is known as a "two-dimensional" account of indexicals. On such an account, you can guarantee that the sentence "I am here now" will come out true, while still preserving within your logical representation of the proposition expressed the sense of contingency and the structure of that contingency. More generally, I would offer, looking at the logical properties of such a sentence can teach us about how indexicals work in the language, and, depending upon your view of the relation between semantics and ontology, may even hold metaphysical lessons (of course this would require argumentation).

You might look at the following SEP articles to dig deeper into the study of indexicals and two-dimensional treatment of them if that interests you:




Short Answer: There is nothing peculiar with the propositions arising from that sentence.

Long Answer...

First, being universally true while requiring an individual is not a problem; in fact, universal quantification is defined in those terms. That is...


Is a universal statement by virtue of the fact that P(x) holds for each individual x.

Second, if someone were to encounter "I am here now" after the fact, it would still be true. For instance, if I was in New York 3 years ago and wrote "I am here now", and you read it 3 years later, you have 2 valid interpretations:

Interpretation 1

  • Here = New York
  • Now = 2012
  • Statement = I was in New York in 2012

Interpretation 2

  • Here = wherever I am
  • Now = the time now
  • Statement = I am where I am right now

You'd only run into a problem if you fixed the value of Here to New York and left Now to mean now. But then there still wouldn't be a problem with the statement, just your invalid interpretation of it.

Third, Heisenberg's uncertainty principle is irrelevant in this case because my statement is about the macro world of unaided awareness. Put another way, when I say I'm at a location, I never set the precision of my claim to be that tiny.

Finally, this statement makes no claim about the ultimate status of time, place or person-hood. It simply uses the conventional notions of such to make its claim.

Ok, with that said...

Language is a tool used in context, so you'll often run into problems if you try to interpret it out of context. That people say things like "I am here now" all the time without a problem seems to hint that the issue isn't the statement, but how you are trying to analyze it.

For instance, if I tell a friend "I am here now", it could mean that I'm standing in a specific spot (as part of a game) where "here" is only true if I remain within a foot of the spot. Or it could mean I'm at a cafe where we agreed to meet, in which case the "here" is true with a precision of a hundred feet or more. The radically different interpretations and thresholds for truth for that statement shows that there is a lot of information that's left out and ignoring it leads to the kinds of pseudo-problems you list above.

The real problem here is in naively trying to analyze a linguistic statement. Don't do that. Rather, spend some time converting the statement into something specific -- like symbolic logic -- and then analyze that. The process of conversion will reveal assumptions, contexts and varying interpretation that would have caused you problems. Once you've resolved all those and have an unambiguous logical formula, then you can analyze it -- if there's anything left to analyze. Often you'll find that the "problem" goes away once you precisely formulate it in logical terms, for the very disambiguation required to do so got rid of the source of the pseudo-problem you set out to solve in the first place.

  • upvoted +1, notably for your last passage, which elaborates in a clear way one fundamental goal of analytical philosophy. – Jo Wehler Oct 14 '15 at 17:31
  • I think OP's previous formulation of the title (prior to my edit) was somewhat misleading. The body asks about the statement. You're definitely right that the proposition is not at all peculiar (except that the definite article the seems inappropriate as there are many propositions that statement could express; though, bracketing ambiguity, only one in a given context). But the statement is peculiar for reasons tied to the semantics of indexicals discussed by me and transitionsynthesis. Definitely a good point in the last paragraph. – Dennis Oct 14 '15 at 17:47
  • Thanks, these are all good, informative answers. True, I should learn some symbolic logic, just a matter of time, priorities, and eyes glazing over. Still, I do think something interesting happens when the statement is written down, if not a "logical" problem. Perhaps the statement is, at least, unusually or even uniquely revealing of its various possible contexts. – Nelson Alexander Oct 14 '15 at 19:25
  • 1
    @NelsonAlexander yes, you can make progress by writing the sentence down, listing possible meanings of the sentence and then studying the context that goes with the meaning. You could also use functions. For instance, Here(me) is always true. NewYork(me) may be true or false (I was in New York 3 years ago, but not now). So some info is missing. This info is time. Now NewYork(me,time) is unambiguous as whether I was in New York at a particular time has a truth value without ambiguity. In this way, context is explicit and you can make progress. – R. Barzell Oct 14 '15 at 19:37
  • 1
    @NelsonAlexander "Phenomenology of Spirit", the chapter on "Sense Certainty". There's a good outline/commentary on that chapter here: publishing.cdlib.org/ucpressebooks/… – R. Barzell Oct 15 '15 at 12:55

In the Kantian sense, to have experience it's not enough for there to be a stream of raw sensations; there must be an I that provides the concepts to make intelligible this stream.

It's also notably the conclusion of Descartes cogito; and the foundation of his epistemology; which was later subverted by the observation that there to be an I there must be We preceding it.

It's also constitutive of the observation that there is something - me - rather than nothing.

And also the first step in self-consciousness, when one is no longer merely an unreflective being; but a being that reflects on being: being here; and then later, not being here (mortality) and possibly never being here.

When unpacked of its sense (Sinn) rather than its form and reference (Bedeutung), it's a sentence that leads in many usefully different directions.

  • 1
    The statement does strike me as close to a "Kantian Cogito," we have the founding intuitions of space, time, and continuity of being in a nearly analytical tautology... but still no content. Then, the "I am" suddenly becomes or "synthesizes" its own content or "truth." Someone must "actually" be somewhere at sometime for the statement to be a tautology. Sorry, this probably makes no sense. The statement is a "rabbit hole" unto itself. – Nelson Alexander Oct 14 '15 at 19:46
  • @Nelson Alexander: it makes perfect sense to me; as a 'tautology' it's Kants transcendental syllogistic. – Mozibur Ullah Oct 14 '15 at 22:51

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.