Back    All discussions

Truth and abstraction
I would like to address an issue which seems to me to be of considerable value as to the relation between language and the external world. The observations to follow are about negation, but the argument may well be extended to cover both disjunction and possibly conjunction.

Negation (the act of negating something) in natural languages seems in many cases, but not in all, to be an abstract description of the world out there. We say, for example, that 'John is not in the office now'. Someone would claim that indeed this is what matters in the world out there, namely that John (the same person) isn't in the office (in the same office and at the same time).  Well, where in the external world can someone spot a state of affairs of the sort 'John is not in the office' -and generally 'X is not in a specific place at a time Z'? The only thing the world can identify is that an individual X is somewhere at a specific time. Not that he isn't somewhere at the same, or whatever, time. That is, in the world out there, there is only presence, not absence.   It is only, when language as a means of reference, or thought as a means of representation, come, that absence can be expressed, or conceived.

As I mentioned, there are cases, though, where negation does not seem to 'behave' in the above-mentioned way.  Take as an example the sentence 'John cannot open the door', when uttered as a description of an individual named John, being physically disable to open a heavy or locked door. In that case, i think that the description has (more or less) its correspondence with reality, whereas in the initial sentence such a correspondence doesn't hold, or, if it holds, this is so in a much indiscernible way. The difference (in my opinion) is that in the first case, there is absence, whereas in the second one there is presence (either we see the man who is not able to open a door, or we imagine of him). To generalize, truth seems to me to be one thing, and abstraction quite another one.  I.e. something may be true in a possible world, albeit being an abstract description of a state of affairs in the same world.

The argument may be extended to cover disjunction (as expressing in many cases an uncertainty which nowhere in the world exists), and, why not, conjunction, something I won't discuss here.

Truth and abstraction
Reply to Nash Kats


Here's a "naturalized" stab at the origin of "negation" (although I suspect it is already implicit in the bivalent logic of reward and punishment, and perhaps even in neural excitation/inhibition). An organism is hungry. It tries to eat a mushroom, but the mushroom makes it a little sick; it tries another mushroom, and that satisfies the hunger. Through a long series of trials, errors, and corrective feedback from having eaten mushrooms that make it feel worse and mushrooms that make it feel better, the organisms learn to distinguish the features reliably predicting which mushrooms are edible and which aren't. Fast-forward to when the organisms evolve into hominids and eventually homo sapiens, with the advent of language. They now know not only which kinds of mushrooms to eat and which kinds not to eat, but that the ones to eat are called "edible" and the ones not to eat are called "inedible" (not edible).

The original affirmation/negation came from the internal consequences of eating things that made you feel better or worse. The internal wiring impelling you to persist in doing the things that made you feel good and desist from doing the things that made you feel bad made it possible to sort the external things into the ones to eat and the ones not to eat. And then, with the advent of language came the possibility of making the proposition "This is edible" and "This is not edible", with truth values "True" and "False" (i.e., not true), confirmable (for empirical propositions) from the negative and positive consequences (and deducible by formal symbol manipulation alone from the combinations of predicates into different true and false propositions). 

Harnad, S. (2000) From Sensorimotor Praxis and Pantomine to Symbolic Representations The Evolution of Language. Proceedings of 3rd International Conference. Paris 3-6 April 2000: Pp 118-125.   

Cangelosi, A. & Harnad, S. (2001) The Adaptive Advantage of Symbolic Theft Over Sensorimotor Toil: Grounding Language in Perceptual Categories. Evolution of Communication 4(1) 117-142  

Harnad, S. (2005) To Cognize is to Categorize: Cognition is Categorization, in Lefebvre, C. and Cohen, H., Eds. Handbook of Categorization. Elsevier. 

Harnad, S. (2007) From Knowing How To Knowing That: Acquiring Categories By Word of Mouth. Presented at Kaziemierz Naturalized Epistemology Workshop (KNEW), Kaziemierz, Poland, 2 September 2007.

Truth and abstraction
Reply to Stevan Harnad
Ok, thank you for your illuminating comments. I don't think, though, that they contravene my observations. After all, the emergence of affirmation and negation through heuristics related to reward and/or punishment, isn't in contradiction, or even in contrast, with what is the essence of my observations, namely the fact that negation -this is what we are interested in here- comes through symbolic representations, describing something which is not primarily out there, that is, absence. This seems so, at least when preceding 'existential' predicates , but not in all cases of its use. Your observations are about phylogenetic aspects of negation, whereas my observations are about deeper, semiotic aspects of negation, and about the relations between the external world, and language (or thought) as systems of mediation and representation. 

Truth and abstraction
Reply to Nash Kats
NK: "negation... comes through symbolic representations, describing something which is not primarily out there, that is, absence. This seems so, at least when preceding 'existential' predicates, but not in all cases of its use."
It seems to me that when I say "This is not-P" I am no more saying something about an absence than when I say "This is Q" (if Q means not-P or vice versa).

And what about internal and external negation, even in an existential statement: Ex(not-Px) vs not(Ex(Px))?

In other words, I don't think negation is semantically linked to absence, even in existential statements -- though it is probably linked to falsity.

Truth and abstraction
Reply to Stevan Harnad

My initial observation is neither merely about the meaning of such a proposition, nor about its truth. It is rather about what language as a means of reference adds in comparison to the external facts. Specifically,It expresses a 'non-fact', i.e. something which can't be spotted out there, namely the 'non-presence' of someone (or something).In other words, it is about language or thought   as a vehicle of enriching the stimuli of the external world, or of making abstract observations of a somewhat second-order nature....

Truth and abstraction
Reply to Nash Kats
Nash said: "...That is, in the world out there, there is only presence, not absence.   It is only, when language as a means of reference, or thought as a means of representation, come, that absence can be expressed, or conceived."

I once had a dog. Her name was Mutthilda. Mutthilda was very attached to my wife (the animal held me in distinctly lesser esteem). My wife left for a few weeks once to visit her family. For that entire time, Mutthilda was the very picture of dejection. She moped around, refused to eat, and was atypically lethargic. Clearly, Mutthilda was able to notice the absence of her Most Favored Person, even though she couldn't talk. (Her writing was pretty illegible, too.)

I have a lot of difficulties with your statement that I quoted above. What you call "the world out there" seems like a very uncanny place. I'm also confused by your saying that absence  can only be "expressed" or "conceived" when language has "come". Well, it is true that one who doesn't have language can't say very much. But, like Mutthilda, one can express oneself without words. Both animals and pre-verbal children clearly notice the absence of loved ones--or the failure of a meal to appear on time. Then again,  no dog can speak of either presence or absence, nor of walruses and sealing wax, etc. So if your point is that language gives us the special ability to speak (to ourselves or others) about "absence", then I must observe that the same holds true of every other topic. Being able to speak or understand language seems to have no special relation to "presence" or "absence".

You seem to be saying that negation is "an abstraction" that can only be expressed via language, but not found in "the world out there". Am I to assume that "presence" is not an abstraction, but something else? Concrete, perhaps? And presence can be understood (or whatever) without language? And surely you can't be saying that absences cannot occur in the "world out there". (Well, I have already said that I suspect you are speaking of a very quirky world, so perhaps you are saying that.) I find these distinctions completely opaque.

I am not sure whether I have ever been to "the world out there". Is it the kind of place one has to buy a ticket to enter? When I sit here at my computer, am I in "the world out there", or does my office not count? Do I have to go somewhere to be in "the world out there"? Well, wherever this world may be, you describe it in startling terms: it is full of "presence"...and nothing is ever absent. This reminds me a bit of Wittgenstein's opening sentence in the Tractatus (if one could manage to misunderstand it completely): "Die Welt ist alles, was der Fall ist".  You seem to think that "what is the case" has a very different standing from that which is not, but I can't make out what the difference you are attempting to point out might be. Other than the obvious one, of course.

Suppose that Tom's wife calls me, and asks me to deliver a message to him in his office (she does this all the time, Tom hates phones, and refuses to answer them). So I go to Tom's office; sometimes he is there, and sometimes he is not. Do you truly want me to think that there is some important philosophical distinction between an occuppied office, and an unoccupied one? If so, I'm going to need a very great deal of explanation.

Truth and abstraction
Reply to Nash Kats
I just really enjoyed Peter Cash's response. To some extent one might focus upon credibility: for instance, which philosophers make it credible that there are interesting things that can be said -- which is just the point -- about a world and which are distinctively philosophically/theoretically interesting in a way illuminated through use of example?

Descriptive, referential or quantificational torturing is a sort of thing that philosophers do in support of that, I don't see why that shouldn't be something real in itself, so that (when it's done well) it feels like a kind of qualitative or even perspectival finessing. And if it is a finessing then there will be linked dependencies ...

As an example of the kind of thing I'm thinking of, taking something from a recentish debate, some of the support that participants seem to feel that they can derive from intuitions or injunctions about semantic and pragmatic partitions giving one some core systematic dependencies, appears to be a matter of what intuitions can be said to discretely support; with just a slight switch in terminology and in significance, those kind of backing-up intuitions might support abstraction at the level of explanation at least.

Truth and abstraction
Reply to Nash Kats
What i think is that you have misunderstood my thoughts. First of all, i am not only referring to language, but -something that you ignored- to thought as well, as vehicles of enriching the external stimuli or adding to them. And i am talking about either a prime thought, or about a much more elaborated thought, as is the case with us, human beings
Second, can you imagine of a visible world consisting of ''absent'' things at a moment? At that moment, if something is absent, or if it don't exist, then it isn't out there. Not the reverse: for example, my desire to dring a cup of beer, isn't something visible, but it exists as a disposition, as a mental state. And this is what my thoughts entail.

Yes, even when you work on your pc, you are doing something visible, something which belongs to the external world, the world of observable things. Have i ever said -as you misconceived my thoughts- that when you are indoors you belong to an ''internal'' world? No, on no account. When i am talking about 'absent things, i am merely talking about things that are considered as such, only when observers of what is out there come into ''play''. And, since they are absent, they just either nowhere exist, or somewhere don't exist. For instance,  when i'm saing that 'dinosaurs don't exist', i mean that dinosaurs are absent, or more precisely non-existent at all - at least at the moment i utter it. Whereas, when i'm saying that 'John is not in the office', i mean that John-the same individual-is absent from the place where his office is. Not that he is absent, as dinosaurs are: he is simply somewhere else.
If there is a difficulty in conceiving my saying that ''absences can't occur in the world out there'', it is due to the fact that in trying to conceive it, you are in any case an observer, and, thus, your thought, or the language you can use to describe it, function as ''mediators'' as to the relation between language and the world.

Truth and abstraction
Reply to Nash Kats
I am certain that I have not understood your thoughts, and I have probably also failed to understand what you wrote.

I can only wonder what possessed me to post my comment in the first place. We are talking here about "language and the world", and apparently "thought" (whatever that is) enters into it also. We are talking about "the external world", about "observation", "existence" and "non-existence"...well it completely baffles me why I thought this was an appropriate forum to discuss such issues. What hubris! What folly! How could I understand what you write in those few paragraphs about some of the greatest, most complex, most muddled topics of philosophy? How could I hope to say anything serious and useful in the same format? My entire life would not suffice to do justice to the analysis of even one of the topics I have put in quotes.

I apologize for my irruption into this forum, and humbly withdraw. If the moderator will kindly delete this and my previous post, I will perhaps have made one minimal contribution: it could then be truly asserted that "Peter's postings are absent here." I will, regrettably, be unable to participate in the analysis of this negation, but wish you well.

Truth and abstraction
Reply to Nash Kats

The problem I have always had with possible world semantics is that it blurs the line between possibility and actuality (or as NK put it, between abstraction and truth (but this is by no means the same distinction).  Our conceptions of language are thus limited to representation and we forget that the meaningful is much broader than the true.  

<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /> 

Your comments on abstraction relate to the origin of language and the “insight” that language begins with representation, or as Wittgenstein saw it, language cannot begin with lying.


But to follow-up with the Wittgenstein theme, language does not begin with representation – it cannot because if it did it would presume the “thought” that we are aware of something and then choose to re-present it and that is to assume (as Jerry Fodor did) that there is a language of thought which precedes language.  But that does not answer the question, rather it is a confused shifting of the problem back a level.


The way we are in the world is not the way a match is in a matchbox and likewise language is not a mere compendium of words that go proxy for things or sentences whose only value is determined by truth or assertibility conditions.  I am now off to read some poetry.


P.S. I too found Peter T. Cash’s comments amusing and also interesting.

Truth and abstraction
Reply to Nash Kats
I don't see the relevant difference between the two cases.  If you
think all truths have truth-makers, then 'John is not in the office'
is made true by the world rather than by a negative fact or state of
affairs.  But so is 'John cannot open the door'.  John does not have a
disability to open the door.  Rather, he fails to have an ability to
open it.  The statement is not made true by an ability or disposition
he has.  If the statement has a truth-maker, then the truth-maker is
the world, not a negative fact about John.

Truth and abstraction
I have trouble seeing the difference also. As far as I can tell, Nash thinks that the important difference is that the first proposition expresses what might be called an "existential negative": it says something is not there. He thinks that the second sort of proposition is only similar in form (or grammar) to the first; in this case, we are not denying the existence of anything, but stating that John is (somehow) incapable of opening a door.

Why is "John is not in his office" so different, and why is this important? I can only guess, but I suspect that certain philosophical views underlie Nash's assertions:
  1. The Tractatus view that the world "is everything that is the case"
  2. The notion that language derives its meaning through its relationship to the world
  3. The notion that propositions are true or false depending on on whether they describe "something that is the case"1
Note that Nash starts out by saying that "John is not in his office now" is an "abstract" statement. Why is it abstract? I can think of no reason whatever; it could be a perfectly ordinary thing to say, given some circumstances in which it could be said. However, if we think of language and the world as described in items 1-3 above, then we might be led to think that if a proposition describes something that is not a fact, then it is abstract. That is, of course, a strange use of "abstract".

Nash's point appears to be that propositions of this form are not related to the world in the manner that we (supposedly) think they are: the proposition describes a state of affairs that is not the case. Because it is counterfactual, it is somehow more ephemeral, more "abstract" than other statements. It seems to me that there is a great deal of confusion here.

Can we never say that "the refrigerator is empty"? Or, "All the air has been pumped out of the glass bottle--it contains nothing."? If we think that language gets its meaning by referring to the world, why should we think that every bit of language must refer to something that actually exists in the specified time and place? Isn't this a bit like saying that it is impossible to utter certain falsehoods--such as lying about someone's whereabouts?

Surely, there is a mistake about what the proposition "John is not in the office now" says. There's nothing abstract about it; it asserts something that can be checked by referring to the world. Indeed, the only way we can verify it is by checking John's office. There is nothing peculiar here (well, who knows what he keeps in his office). And you know--perhaps Nash is wrong, and John is in his office after all. Does this mean that the proposition retroactively becomes "concrete" instead of "abstract"?

One of the difficulties here is that no one really cares about poor John. Nash does not care enough to go looking for him, and no one else has launched an investigation. Would anyone assert that the claim "John is not in his office" is mere abstraction if his absence was a sign that something dreadful had happened to a real human being--perhaps even a beloved department head? No! We'd all go searching for him! No talk about whether this claim is "abstract" or whether it has anything to do with the "world out there".
1. I do not subscribe to these views.