Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

What is meaning? What is its relationship to phenomenality?

  1. Mar 31, 2005 #1


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    The issue of meaning has popped up a couple of times recently in the Metaphysics & Epistemology forum, and I think it's substantial enough to merit its own discussion.

    What is meaning? As a first pass, it seems reasonable to define meaning in terms of representation. In turn, the notion of representation seems to imply a system for which something is represented. Here, then, is a tentative definition:

    An object or event X has meaning for an agent A just in case A takes X to represent some other object or event Y.

    The terms "agent" and "takes X to represent" remain somewhat ambiguous. The latter might be specified, for instance, either in terms of belief ('A believes that X represents Y') or action ('A behaves as if X represents Y,' or even 'A behaves as if X is Y.').

    I am particularly interested here to investigate the link, if any, between meaning and phenomenality. On the plausible definition given above, phenomenality is not necessary for meaning. A zombie's mental life would feature just as much meaning as ours, even though a zombie is not p-conscious by definition. For instance, a stop sign would have just as much meaning for a zombie as for a human. Just like a human, the zombie would take the stop sign to represent a command to stop a moving vehicle. Of course, a zombie would not have the attendent subjective experience of meaningfulness, but on the definition given above, the qualitative component of meaningfulness is not an essential aspect of meaning.
    Last edited: Mar 31, 2005
  2. jcsd
  3. Mar 31, 2005 #2

    I think you are absolutely correct in saying that phenomenality is not necessary for meaning. Instructions in a computer program have meaning, so much so that programming languages fail to translate or execute meaningless statements.

    That said, meaning in aesthetics is particularly complex. For instance, colors usually have a sort of aesthetic meaning that can't be expressed in language (unless you consider art a language in itself). You said a stop sign means the same thing for a zombie as it does for us, but I'm inclined to believe we have a reason for painting stop signs red, while zombies would not respond differently to a green stop sign. That is, there's something intrinsic to our perception of red besides linguistic and cultural conventions.

    Also, in music some tone intervals are clearly interpreted as "happy" or "sad" by most listeners, but no one knows why abstract mathematical relationships are capable of having that kind of meaning. Again, this is independent of cultural conventions, as far as I know.

    Because of aesthetics, I feel inclined to think zombies cannot be perfectly equivalent to sentient people, modern philosophers notwithstanding. The very notion of zombie in folk usage reflects that: zombies don't have aesthetic experiences, and they appear and behave as such, with their steely eyes and lack of moral values.

    So to answer your post, I think the full meaning of meaning is somewhat elusive. We can understand meaning from a logical/linguistic perspective, the same perspective which allows people to create computer languages. But the aesthetic aspect of meaning is quite far from being understood, and it's a very important component of it.
  4. Mar 31, 2005 #3


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    On the definition of meaning I gave above, a key component is the system or agent for which some object or event has meaning. I believe we must take into account the context afforded by a given agent to evaluate something's meaning, because it's not plausible on a definition of meaning that incorporates representation that things just have some sort of inherent meaning.

    With that in mind, I believe that computer instructions are certainly meaningful for us, as human cognitive agents. But we are also phenomenally conscious, so it's not clear that the fact that we view computer instructions as meaningful can tell us much about meaning in the absence of phenomenality.

    For the sake of argument, we can suppose that simple computer programs are not phenomenally conscious, and I think this is what you were trying to get at: that we can demonstrate that phenomenality is not necessary for meaning by appealing to computers or computer programs as agents that are not phenomenally conscious, but for which instructions nonetheless have meaning. However, I'm not sure that all computers/computer programs, as such, fit the bill.

    Your example of ill-formed computer instructions seems more like an issue of syntax than semantics, when regarded from the computer's point of view. Certainly, a program will only function 'properly' when it is given syntacticly 'well-formed' instructions (notice that the words in scare quotes are human-imposed norms). However, it seems implausible to suppose that a simple computer program would take its instructions to be representative of anything; it seems more like a type of 'dumb' causal chain that operates on certain human designed specifications.

    This is not to say that no computer could find anything meaningful. For instance, a suitably designed artificial intelligence computer could find things meaningful, according to the definition I provided. It's certainly a subtle and difficult question about where the boundaries and gradients between syntax and semantics lie, and one could come up with different proposals based on different interpretations of the initial definition of meaning that I provided. However, I think the case of instruction syntax for simple computer programs is extreme enough that we can plausibly say that instructions are not meaningful for simple programs, according to the definition of meaning I proposed above.

    Zombies are structurally and behaviorally identical to humans, though, by definition of what it means to be a zombie. So to the extent that humans have reason for painting stop signs red, zombies have equivalent reasons. If perceiving red triggers some sort of 'warning!' or 'caution!' mechanism in humans, it does the same in zombies. All that is different is that a stop sign qualitatively looks something like this for a human, but does not qualitatively look like anything for a zombie.

    Zombies would not have steely eyes or lack of moral values to any greater or lesser extent than humans, by definition. They would also value art and so on to the same extent that we humans do, again, by definition. The only difference is that in humans, art has some sort of phenomenal presentation, while it has none for zombies.

    So on balance, you are more or less claiming that zombies are metaphysically impossible; there is some sort inseperability between some qualitative experience and some behavior. This constraint of yours seems to necessarily push you either towards some sort of physicalism (i.e., it is logically impossible to have certain mental functions without some sort of illusory, qualitative experience arising) or some sort of interactionist dualism (i.e., it is logically impossible for certain mental functions to be instantiated without some sort qualitative experience to cause them).

    I can understand why you would want to consider some sort of contribution of aesthetics to meaning, but it's still not clear to me that such contributions could not be instantiated in zombies.
  5. Mar 31, 2005 #4
    In philosophy the notion of meaning is very wide and extensively controversial. Partly because meaning is a multi-disciplinary notion - that is, it cuts across several philosophical disciplines. In philosophy there are many papers by Russell, Frege, Strawson, Kripke, Donnellan and countless Language and anaylitical philosphers that directly confront this notion of meaning - e.g Russell's Theory of description . What I mean is that meaning cannot be defined in one context for there are many contexts that one has to examine in other to do justice to it. For example, there are some metaphysically vexing problems with meaning in Propositional Attitude context (such as in belief context). Philosophers have a problem with propositional attitude sentences with regards to meaning such as these ones that you specified: 'A believes that X represents Y' 'A behaves as if X represents Y. When you read their papers thoroughly you will understand why. You do not have to agree with them, but it is controversial in general.

    The question is whether consciousness can be progrmmed into a zombie, given that the zombie is scientifically improvable or programmable. And there is the issue of evaluating the evolutionary value of consciousness in the grand scale of things or life in general. If you ask me "can consciousness be programmed into machines?", my answer would be a capital 'YES'. Can we? Oh yes, we can!
    Last edited: Mar 31, 2005
  6. Apr 1, 2005 #5


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Right; I certainly don't pretend to have given a complete or non-problematic account of meaning. That's why I portrayed my effort as a tentative first pass.

    That's not the question at all. For one thing, we cannot program anything 'into' a zombie, because zombies do not exist. They are hypothetical entities that some philosophers believe are logically possible; certainly no one seriously considers them to be nomologically possible (possible in our world). Furthermore, if zombies did exist in our world, they would be physically identical to us humans (by definition), so I'm not sure it even makes sense to talk about programming anything into a zombie. What would it mean to program something into a human? Perhaps it makes sense on some level, but it doesn't seem to be the best choice of words.
  7. Apr 1, 2005 #6
    I believe your position is quite mistaken. It's not a given agent that affords context; if you don't have the context which provides meaning to a statement, you can't arbitrarily create that context yourself. Context can be given either by the language itself or by the phenomena the language refers to; it certainly can't be given by the conscious agent in the language, except in the few cases where language refers to subjective phenomena.

    The point I was trying to make is that meaning can exist in the absence of phenomenality. Once a language becomes sophisticated enough, you can create concepts that refer to the language itself rather than to anything the language refers to (let's call those 'meta-concepts'). Any computer is perfectly capable of understanding meta-concepts. Moreover, we understand meta-concepts in exactly the same way computers do. Words like "not", "and", "more", and so on, do not bring any images to mind. We reason about those concepts completely in the dark, so to speak.

    Not at all. A computer will not refrain from executing a statement such as 2/0 not because it contains a syntax error, but because it has no meaning.

    I think what you are failing to see is that computers did not come out of nothing, they were created by people. Computers are the embodiment of our ways of thinking, just like voice recorders are the embodiment of our ways of speaking. As such, computers provide a wonderful opportunity to look at our own minds from an objective perspective. The fact that some people don't recognize themselves in the workings of a logical machine only means they don't understand how the machine was created.

    That is not the issue. The issue is, does a statement need to represent anything in order to be meaningful to us? And the answer is a clear negative. Gosh, even the question "does a statement need to represent anything" does not represent anything other than meta-concepts.

    I see a huge problem with this zombie thing. Zombies are thought not to have subjective experiences, yet they are also thought to behave as if they have subjective experiences. But you can't simply say "that is the way it is by definition" without justifying it, without addressing complaints that the definition is bogus; that would be sophistry.

    Of course not, because the definition of zombies is bogus. The problem is that without subjective experiences there is no language and therefore no meaning. Defined the proper way, a zombie is exactly what we are when we are unconscious: we can still move our muscles and talk gibberish, but we cannot possibly behave the same way or have meaningful conversations. Your line of thinking leads one to conclude that subjective experience is just a frill, an unnecessary ingredient of brain activity. I find that notion ridiculous.
  8. Apr 1, 2005 #7
    An agent is not the ultimate origin of context..

    ..but "langauge itself" is nothing without an agent.

    "syntax error" and "meaninglessness" are both interpretations of something causal.
  9. Apr 1, 2005 #8


    User Avatar
    Homework Helper

    First of all, zombies are logically possible, unless you can find a way to prove the people around you are in fact conscious. You might believe some behavior requires consciousness, but this can only be based on your own subjective data which isn't infallible when trying to describe the objective world. All logically possible means is that there is no logical contradiction in their defintion. For example, an eight headed unicorn is logically possible, but an eight headed unicorn with seven horns is not.

    As for the notion of defining meaning. We're essential trying to find the meaning of the word meaning, and it seems self reference might be involved. The problem is that we take a certain intellectual ground for granted, and all our meanings derive from that. For example, if we ran into an alien intelligence, how could we convey information to them? We would need some symbolic system. But first we would need some intellectual grounding to work from, maybe prime numbers. From there we would move up to math, physics, and eventually we might be able to communicate with them as we do with other humans. But how could this be possible without finding some common ground? In other words, there may be some necessary concepts that cannot be defined in any way, but are taken for granted and from which all other concepts are built. The question is what these are, and whether they are unique, or if there are many possible basis sets.
  10. Apr 1, 2005 #9
    It's perfectly possible to prove that people are conscious, since a person must be conscious to behave in a conscious way. By definition.

    What is impossible to prove is that people have the kind of ineffable subjective experience zombies are supposed to lack. But how do we know we have those, if they are truly ineffable? If someone tells you "I have this thing in my head which I cannot describe to you", how would you know what they were talking about, and how would you know you also had it?

    Clearly you know you are conscious not because you discovered it from introspection, but because someone told you so. So the claim that it's impossible to tell if people are conscious has no substance: you certainly need other people to know if you are conscious to learn about that fact yourself.

    Ah, but there is exactly such a contradiction in the definition of zombies, only it's not as clear to see as in your example. But it's there nonetheless, and the fact that some people can't see the inconsistency doesn't mean it's not there.
  11. Apr 1, 2005 #10
    Though I am of the opinion that you have not thought out the consequences of your position as carefully as you should, I agree with most everything you have said. :approve:

    To discuss the meaning of meaning, we must give meanings to the terms used in the discussion. It should be clear that "meaning" is a component of language. Without meaning, any language is meaningless and without language, meaning does not exist. The purpose of language is communication and the purpose of meaning is communication. So long as any part of that structure is vague in any way, communication fails. What is important when I use a word, symbol or any reference (and yes, art is a language, one of the vaguest languages in existence), that reference brings to the mind of another the same thing it brings to my mind. Once you see that, you begin to see the problem. That problem even occurs within one's own mind; sometimes internal references bring alternate things to mind. That's why we can't even understand our own comprehensions. Actually, I think the advantage of conscious is that we can exist as a separate entity in our own minds and can thus take advantage of this very phenomena. :cool:

    To quote myself:
    That is, accidental discovery of new interpretations of available information is essential to learning. So the imprecise communication between the subconscious and the conscious mind may very well be the source of the great value of having a conscious mind. Now, that's a theory! :rofl:
    First, what the devil does "NL" stand for? Secondly, I don't think you comprehend my starting point. You should look at my post on that issue. o:)

    The only reason I use mathematics is that I am fairly sure that I will obtain almost universal agreement with any conclusions I can deduce consistent with that field of endeavorer. My problem is that I cannot get anyone to look down that rabbit hole. I have no idea what they are afraid of. :confused: At least my approach begins with a set of meanings almost universally agreed upon. Starting from there, I have a procedure which yields exact definition after exact definition perfectly consistent with everything at every point. It is my contention that my approach is the only rational approach to meanings and I am resisted by those who would rather have no approach at all. :biggrin:

    Have fun -- Dick
  12. Apr 1, 2005 #11

    Les Sleeth

    User Avatar
    Gold Member

    I think it's just the opposite; that is, meaning, as an experience, is so simple it defies the complexity needed for any sort of explanation. To clarify a bit more, I don't believe the experience of meaning has any "parts" to it, but rather is a sort of "experiential singularity," and so like its sister qualia, can't be segmented into the parts an intellect requires to formulate a proper definition.

    It is my opinion that all the current models of consciousness are missing this “singularity” feature. Out of that unified field emerges offshoot singularities which are likewise indivisible and have forever evaded precise analysis by the intellect (e.g. love, happiness, interestedness, compassion, appreciation, etc.). Logically, the integrated facet would be the seat of individuality and subjectivity, the unique “me” at the core of one’s being.
  13. Apr 1, 2005 #12


    User Avatar
    Homework Helper

    Don't mix up conscious as in "awake" with conscious as in "experiencing." You are experiencing while you dream, so these obviously can't be exactly the same thing, but their similarity can cause confusion.

    We can tell others experiences exist, but we can do nothing to convey what they're like. You know red looks like something, but how can you put it into words without simply listing red objects or emotions that accompany red, both of which are completely independent of what the experience itself is like? A zombie would say all the same things, but there would be no accompanying experience. When he says red looks like something, he would be wrong. But when he tries to describe what it looks like, he would do no worse than you.

    In a way, but I think we all know there is something it is like to be us and experience what we do. When we hear the definition of consciousness, we know other people have singled out this phenomenon as well and have invented a word to talk about it. There's no paradox here, if you accept that we can say that we experience but not what it is like.

    What is it then? Logical, or a priori, possibilty is just that there is nothing in the defintions of the words involved that is contradictory. A zombie is an entity that behaves identically to us but does not experience. All experience means is that there is something it is like to be them. Where is the contradiction in these defintions?

    Now, it may turn out down the road that we find some law of nature that proves they are impossible. If one did turn up, that would mean they are a posteriori impossible. I believe they are impossible in this sense, but science has yet to provide any evidence either way.

    Do you have evidence they are impossible, or is it just your belief? I don't think you do, but even if so, their logical possibility remains.
    Last edited: Apr 1, 2005
  14. Apr 1, 2005 #13


    User Avatar
    Homework Helper

    While typing my last post, I thought of something that actually connected it to this topic. Could it be that the only reason we think of experiences as ineffable is that we totally lack even a fundamental language for describing them? When we ask what it means for something to "look green", we dismiss the question as unanswerable ("you have to experience it to see"). Is this really true, or are we giving up too easily?
  15. Apr 1, 2005 #14


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    The point is, any given statement does not have meaning unless there is an agent around to interpret it as having meaning, or as signifying/representing something else. The words in a book do not have intrinsic meaning; they are only meaningful when regarded by an agent who understands the language. (Or, they could even be taken to mean something by an agent who does not understand the language; for instance, perhaps the patterns of shapes and colors resembles a visual image of some object for the agent.)

    Alright, you seem to be arguing for a more wide-ranging notion of meaning than the one I intended. In my definition, we need an agent to be aware, in some sense, that some sort of representation is going on. (Note that awareness, as I'm using it, does not require phenomenality.) That's a pretty ambiguous issue, and depending on how one interprets the various terms, it might amount to something like you're proposing here. However, I still think it's absolutely essential that we build into the definition some reference to a system for which a given object/event/process/etc. has meaning. Things do not have intrinsic meaning; they acquire meaning in the proper contexts, which must include some sort of system that 'interprets' things in some way or another.

    Actually, computers refrain from dividing by zero because human programmers explicitly give them instructions to avoid attempting such operations, so it amounts to an ill-formed request. Without such instructions, a program will straightforwardly go about trying to calculate 2/0 according to its particular algorithms (and will probably crash).

    The point is, the kind of computers we are accustomed to do not have a concept of what numbers are, or what operations on numbers are. When we feed 2+2 into a calculator and get back 4, it's just a blind causal chain, not very much different in character from dropping a ball or knocking over a set of dominos. Now, obviously it's more complex than that; one could characterize the brain in a similar matter, and obviously we do take things to have meaning. It's not clear exactly what happens in the brain to make this occur, but I do not think it's going out on a limb to say that simple computer programs do not have it, but more complex, AI inspired computers could.

    I explicitly acknowledged that point several times, and in fact it was the reason I objected to your example. I thought we should consider things from the computer's point of view (if it even makes sense to attribute a point of view to the kind of computers we use today), not from our human point of view, to evaluate the question of the meaning of instructions for a computer.

    Please produce a meaningful statement that does not have representational content.
  16. Apr 1, 2005 #15


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    I'm responding to this post in two parts: the last to address meaning, and in this response to address the issue of zombies.

    The definition is what the definition is. You can argue that zombies are logically impossible, or in other words that the definition is incoherent, which is fine. But if you want to talk about hypothetical beings whose behavior resembles humans' in some respects, but who do not appreciate art and have no moral values and the like, you are not talking about zombies. You are talking about something else.

    Obviously, accepting the logical possibility of zombies presents problems. But rejecting the logical possibility of zombies also presents significant problems. As I said before, if you reject the logical possibility of zombies, then you are essentially claiming that human behavior and human subjective experience are inseparable in a very strong sense. You must either claim that it is logically impossible that certain mental functions/brain activities occur without subjective experience arising, or you must claim that certain mental functions/brain activities cannot occur without subjective experience playing a role in their causal production. The former amounts to some form of physicalism, the latter interactionist dualism. But there are strong arguments that physicalism is not a sufficiently powerful ontology to account for subjective experience, and interactionist dualism rejects the well-founded principle of the causal closure of the physical. So, if you reject zombies, which of these impalatable options are you going to choose?

    By the way, it is not the case that accepting the logical possibility of zombies forces us into a kind of epiphenomenalism where subjective experience is just an unncecessary frill. The "A Place for Consciousness" discussion is about a book that demonstrates this notion.
  17. Apr 2, 2005 #16
    it seems that in this flow of thought about the idea of meaning, zombies play a key role.
  18. Apr 4, 2005 #17
    So many interesting replies, too bad I don't have time to address all the issues.

    Language is used for communication, but it's also used for thinking. In fact you can't use language to communicate if you don't use it first for thinking, although this is a bit tricky to explain. What needs to be understood is the difference between words and what words refer to, the major difference being that words are arbitrary while their referents are not. For instance, a child may not know the English word for "sun" but she knows a word probably exists, so the child can think about "sun" without even knowing if the word actually exists. So we have those "placeholders" in our minds which are filled with whatever it is that a word stands for, and when we learn the word all we have to do is attach it to the existing placeholder.

    I said all that to get to the point that your problem may or may not exist, it really depends on the context you're talking about. If you talk about "the decline of Western society in posmodernistic times", you may have more words than placeholders for them (that is, you learn words without really knowing what they mean); on the other hand, when talking about "the sun rised in the East and sets in the West", the words may be vague but the meaning, at least in my head, is absolutely clear.

    So I don't fully acknowledge the existence of the problem you're talking about when it comes to descriptions of the physical world, although I certainly agree it exists in the more abstract domains of human knowledge.

    This is really what I was trying to point out when I mentioned aesthetics. I'm glad you seem to agree. We can't create this artificial dichotomy between the physical world and the picture we have in our head, they are the same thing. Stop signs are not simply abstract concepts, they are red hexagonal things which can only have meaning when taken in their entirety. I don't know about others, but I have never seen an abstract stop sign, the kind of which Hypnagogue says zombies are capable of seeing.

    Of course we can, that's why we have language.

    The contradiction is embedded in the meaning of "behavior" and "experience". Just because you can put a sentence together without offending the rules of grammar doesn't mean the sentence doesn't harbor a contradiction. The problem is that the contradictions can be very hard to see. For instance, Newton's laws of mechanics were contradictory, but those contradictions only showed up 300 years later when they gave rise to the paradoxes which could only be solved by relativity and quantum mechanics. And those, as everyone knows, also give rise to paradoxes of their own.

    But physics is a lot easier than philosophy, because we cannot lie to the world. If an equation describes a real physical phenomenon as an undefined mathematical entity, we know the equation must be wrong. Philosophy however doesn't have that luxury; we can go on and on for centuries making logical mistakes and nature will never correct us.

    Actually, we need to make some distinctions before we can clearly discuss this issue. Words do have some form of meaning; that meaning is given by its relationships with all other words in the language. That kind of meaning is what allows computers to translate many statements from one language to another.

    But there's more to words than their relationships to other words, and that is the part of meaning which only exists in the minds of conscious agents. That kind of meaning is what prevents computers from translating most statements from one language to another.

    I contend that the second kind of meaning can only exist together with what I'm loosingly calling "aesthetic experiences". A computer, or your zombies, can know that "happiness" is the opposite of "sorrow", but they cannot know why one must be sought and the other avoided.

    Try that yourself. Try explaining to your favorite zombie why is it humans spend so much time and energy seeking happiness when sorrow is a lot easier to find.

    Any statement about language itself qualifies. To take a simple example, "all meaningful English sentences must include a subject and a verb"

    By the way, that is the kind of sentence a computer/zombie is perfectly capable of understanding. The grammar checker in your word processor software certainly understands it.
  19. Apr 4, 2005 #18
    Newtonian mechanics is not self-contradictory, it is just not
    in line with the behaviour of matter.

    Not it doesn't. It *implements* it. A falling rock does not "understand" gravity.
  20. Apr 4, 2005 #19
  21. Apr 4, 2005 #20

    Les Sleeth

    User Avatar
    Gold Member

    I took Hypnagogue's topic to be about the experience of "meaningfulness" that can accompany literal meaning. A zombie could identify the meaning for a word, just like it can recognize the color red. The zombie isn't capable of the phenomenal element which human consciousness does rather effortlessly.

    If your link applies, then I've posted to the wrong thread.
    Last edited: Apr 4, 2005
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?

Similar Discussions: What is meaning? What is its relationship to phenomenality?