5

This relates to the physical nature of matter, not to metaphysics. I cannot find a robust, universal definition of consciousness, so perhaps the question is meaningless. I am thinking of the behaviour of molecules in arriving at equilibrium within a system, for example. Does the system exhibit consciousness? Is consciousness an incrementally emergent feature?

13
  • 3
    It is hard to answer, mainly due to the fact that we do not know what consciousness is for living creature. Oct 23, 2023 at 15:44
  • @Mauro ALLEGRANZA To come to terms with consciousness is to ask nature. To pose questions to nature means to design experiments about mental processes. Such experiments check and correct our prejudices. And that’s what neuroscience does with some success. - I know that’s only a rough picture of scientific progress :-)
    – Jo Wehler
    Oct 23, 2023 at 18:26
  • 13
    Define 'living', & 'consciousness', & you'll have your answer.
    – CriglCragl
    Oct 23, 2023 at 22:51
  • 1
    Despite the vagueness of "consciousness" and "life", "if conscious then alive" seems pretty close to analytic truth on common meaning of those words. Panpsychists attribute some sort of proto-consciousness to all objects, animate or inanimate. Then the question becomes whether one is also a panvitalist, i.e. attributes some sort of proto-life to all of them as well, which seems natural.
    – Conifold
    Oct 23, 2023 at 23:09
  • What is consciousness? I think that is the open question right now. If you choose the right definition then your phone could probably be considered conscious, but is that what you want your definition to include. And what is life? If you require self replicating then we could just choose a sterile person and you have your answer Oct 24, 2023 at 7:58

10 Answers 10

12

I think the jury is still out on this one -- if you could answer it definitively that would be a huge discovery.

The challenge is that we don't have a good theory of consciousness yet. There are numerous competing ideas at the moment ranging from Eliminative Materialism (consciousness doesn't exist, we just think it does) to different flavors of dualism.

Even worse than no theory is our inability to detect that a particular assemblage of matter has a subjective experience associated with it.

One of the most promising general theories of consciousness I've run across is Integrated Information Theory (IIT) by Tononi et al. They provide a method for quantifying the degree of consciousness possessed by an object based on an assessment of several of its physical and operational properties. It does suggest that relatively simple things like thermostats can possess a type of subjective experience while very complex things like ChatGPT and computer-simulated brains will not actually be consciousness (but will emulate it very well).

While this theory has a strong intuition behind it and is derived from axioms about the basic character of consciousness, it still does not explain why these particular qualities get to have an experience and other don't. This is not unique to IIT, but to any physicalist/functionalist theory of mind. We are left with Chalmers' "Hard Problem" of why subjective experience exists vs just its functional/physical aspects.

2
  • 2
    The paper "The rise of machine consciousness: Studying consciousness with computational models" sciencedirect.com/science/article/abs/pii/… is a survey about the assets and drawbacks of 5 current models, including the Tononi approach. On request I can send it to you.
    – Jo Wehler
    Oct 23, 2023 at 19:00
  • 1
    @JoWehler -- nice! This part of the abstract is quite telling... "...no existing approach to artificial consciousness has presented a compelling demonstration of phenomenal machine consciousness, or even clear evidence that artificial phenomenal consciousness will eventually be possible. "
    – Annika
    Oct 23, 2023 at 19:07
6

We know from neuroscience that there are many unconscious processes operating in the human brain. It is even possible that a stimulus is registered in certain parts of our brain, but we do not have a consciuosness perception. Neverless, the stimulus can influence our action. See for an introduction to this field.

The necessary requirement that mental processes become conscious: The excitation has to pass the thalamus, more precisely: These processes have to activate the thalamo-cortical system.

  • Hence there is no reason to speculate that non-living natural systems have consciousness. They do not even have neurons for information processing.
  • Concerning artificial systems like networks of computers one cannot exclude that one day they evolve to the level of consciousness.

If one likes to apply the term “emergent” to any system which shows more properties than the sum of its components, one may call consciousness an emergent phenomenon. Because it presupposes the forward and backward activation between the subsystems of the thalamo-cortical system.

For the whole question see also a previous similar question.

Added: Presently an adversarial experimental study is on the way to distinguish between the predictions made by two competing theories to explain the relationship between conscious experience and brain activity. The study employs as working definition for consciousness:

“Phenomenologically, consciousness has been defined as subjective experience (or what it is like to perceive, feel, act or think from a first-person perspective).”

10
  • Maybe AI research should focus on emulating the actions of the thalamus and cortex?
    – Scott Rowe
    Oct 24, 2023 at 2:11
  • @ScottRowe No, because most AI research is not about producing consciousness.
    – kutschkem
    Oct 24, 2023 at 7:54
  • @kutschkem perhaps it should be.
    – Scott Rowe
    Oct 24, 2023 at 10:34
  • 1
    Note also that the definition used by the study does not require that a conscious entity have a thalamus.
    – Dcleve
    Oct 24, 2023 at 19:04
  • 1
    This answer seems to be nothing but guesses. What you think of as unconscious processes in the brain may be conscious processes with no ability to communicate that fact to the outside world. There's also no reason to believe that consciousness requires "neurons for information processing"; I'm happy to agree that our brain processes information, but you're assuming your conclusion if you assume that's a necessary element of consciousness.
    – benrg
    Oct 24, 2023 at 21:05
2

If consciousness is self-awareness, of oneself as separate from one's environment, I would think that inanimate matter (rocks, even made objects) do not have consciousness. However if consciousness consists of separating ourselves from the world around us, is sounds more like a disease with interesting results - it allows us to react to changes in the environment, and even to act on it.

I'd be more interested to explore what kind of consciousness would exist for living organisms that don't have brains. A tree can react to its environment and act on it (walnuts poisoning the land their root ball resides in, for example). Is it self-aware? Maybe not in the sense we are biased to understand, since we have a brain, and have a privileged view of it.

Another example is tunicates (https://www.ck12.org/c/biology/tunicates/lesson/Tunicates-Advanced-BIO-ADV/): as larvae they need to move around, so they have a nervous system. As adult they become sessile, and lose the ability to move; when they do, they absorb and lose their nervous system. Do they lose any consciousness they might have had?

This being said, I enjoy the fact that there is no answer to the question. We can have a lot more interesting discussion when the question is all there is.

1
  • Yes, I think my idea is that the system has some awareness of itself as manifested in its behaviour. Examples are chemical equilibrium and quantum entanglement. It is also interesting that quantum theory seems to require consciousness.
    – Meanach
    Oct 26, 2023 at 11:35
2

There are multiple competing theories about consciousness (SEP), and they give different answers to your question.

Many of the physicalist theories about consciousness assume an identity theory (SEP) between either neurology, or a function and consciousness. The neurology-based identity theories would hold that non-living things cannot have consciousness, as they don't have neurology. The functionalist theories (SEP), would hold that non-living things can do the functions of a brain, and therefore they could have consciousness.

The problem that identity theories tend to have, is that we know we do not have consciousness in many cases when their identities are satisfied -- IE recursive neural nets do not always exhibit consciousness, contrary to Paul Churchland's identity claim https://www.amazon.com/gp/customer-reviews/R5048CH7VMV78?ref=pf_ov_at_pdctrvw_srp. Without 100% identity, these theories run into Chalmer's Hard Problem of Consciousness -- if there is no necessary coupling, ascribing consciousness on some occasions to a substrate is -- ad hoc and unpredicted.

Both of the above are examples of REDUCTIONIST physicalism, where mind just IS some feature of the substrate. Most physicalist theories of consciousness today are emergent (SEP) physicalism, where consciousness is something new, or added, and is not ONLY the reduced properties of the system. Emergent physicalism struggles to provide a clear concept of emergence, and therefore of testability for these theories. Jaegwon Kim, in Physicalism or Something Near Enough (GB), actually claims that there can be no emergent physicalism, because if physicalism's assumption of causal closure of the physical is true, then all causal consciousness reduces.

William James and Karl Popper proposed the evolutionary test case for any theory of consciousness that does not hold that consciousness is causal: if consciousness is only occasionally and not necessarily associated with an "identity" substrate, then evolutionary variance would decouple any apparent coupling. This would lead to an incoherent relationship between consciousness and the behavior of a living thing. The coherent coupling we see requires that consciousness be causal, independently of the substrate. Popper argued therefore that consciousness is emergent from the physical, but is
causal. This is interactive emergent dualism (wiley.com). Under interactive emergent dualism, whatever caused consciousness to emerge within living things, could in principle also occur in non-living things, but this is unlikely. Other posters may want to comment, but I am not aware of any failed test cases for Popperian emergent dualism, other than the past life and near death studies.

Aside from these emergent theories of consciousness arising from matter, there are also idealist views among some scientists, in which matter emerges from consciousness. In idealist models (SEP), non-living things could very well be conscious. Idealism struggles to explain the coherence and stubbornness of matter.

Interactive spiritual dualism holds that matter and spirit interact. In spiritual dualism, a spirit could in principle ensoul all sorts f non-living things, as well as living things, so under interactive spiritual dualism yes, non-living things could be conscious. The major challenging test cases for spiritual dualism are brain damage studies that show reduced consciousness.

11
  • +1 This answer addresses the question with a broad metaphysical flexibility.
    – J D
    Oct 24, 2023 at 19:38
  • Is it true that a neurological-centric theory wouldn't presume that the functional equivalency is the important property, and not the possession of neurons proper? The ANN of Pitts and McCulloch presumes that the model in some way captures the essence.
    – J D
    Oct 24, 2023 at 20:39
  • @JD I would rate the concept that artificial neural nets are conscious as a mechanical identity theory, however neural nets are currently primarily emulated, making their use in modern AI an algorithmic identity theory. Could you point to where Pitts and McCulloch make any identity theory claim though? cs.cmu.edu/~./epxing/Class/10715/reading/…
    – Dcleve
    Oct 24, 2023 at 22:01
  • lol It's been a while. I do have some annotations in my copy a la Boden's AI philosophy compendium, but I'll review to sharpen the focus. No one may address the question directly if neural functional equivalence crosses the claims that neurons are necessary for consciousness and that there are functional equivalences that are circumscribed with the notion of "neuron". Maybe Searle's work on philosophy and neurology (or was it a dialog edited between Searle and Dennett.)
    – J D
    Oct 24, 2023 at 22:25
  • "recursive neural nets do not always exhibit consciousness" - When have NNs exhibited consciousness? We haven't made a full artificial replica of the brain, and artificial NNs haven't achieved consciousness, so calling such a hypothetical future classification ad hoc at this stage would make about as much sense as saying "yeah, well, rocks aren't conscious, therefore calling humans conscious is ad hoc and unpredicted". You're ignoring the possibility of there being some non-ad-hoc differentiating criteria. Also, if consciousness is a spectrum, your argument fails much harder.
    – NotThatGuy
    Oct 25, 2023 at 12:08
1

Problem here is "living" and "consciousness" may be defined in several, often mutually incompatible, ways.

Answer to your question will change with definitions change, of course.

I know it's rather difficult to come up with a consistent definition of those terms, but you should remember the famous quote: "if you can’t clearly and simply define the words and terms you are using, you don’t really know what you’re talking about" (Feynman).

To the best of my knowledge "consciousness" is a subjective experience, which has several consequences:

  • It may or may not have physical existence, even subjectively.
  • It is impossible to ascertain if other people have the same experience, you just know they say so, but you don't know (and CAN'T know) if it resembles to your (or my, of course) own.
  • Extending it to other "entities", be them dogs, cats or A.I., is even more problematic.
  • It may be possible to find a sound behavioral definition... in which case I suspect "machines" would end up beating the "testing suite", more sooner than later, as they conquered Turing Test.
  • Beating the "Consciousness Test Suite" won't necessarily mean they are "self conscious", just they can fake it very well... as I (and you and everybody) can do.
1

It's far from meaningless. It's a live philosophical question that people have very different beliefs and intuitions about. It's particularly an issue for scientific material monism (arguably the currently dominant non-religious view of the universe) which has so far failed to find a widely convincing explanation of consciousness (the so called "hard problem"). The emergentism you mentioned is one of the major contenders.

Although this has been an active philosophical debate for centuries, it's has been given renewed vitality by the recent advances in AI. We appear to be approaching a convincing simulation of consciousness (some people have claimed we are already there, although that's an unpopular viewpoint). And according to the line of thought descending from Alan Turing, a convincing simulation of consciousness is de facto evidence of consciousness.

If we did reach that point, however, it wouldn't necessarily resolve the underlying debate. On the face, it's evidence that consciousness is material--but some thinkers propose instead that the material is built on the mental, another debate that has gained new life because of technological advances (in this case, towards convincing simulations). If it demonstrates that consciousness is emergent, it still isn't an explanation of how that emergence works. And even traditional dualism might argue, perhaps, that the sentient computer has somehow gained or grown a soul, rather than that it is a purely mechanical, soulless object.

0

Non-living systems can possess machine consciousness, that is , a consciousness conditioned by science. There are other types of consciousness like consciousness conditioned by vegetarian food or consciousness conditioned by business rules or consciousness conditioned by suffering etc.

There are infinite number of conditioned consciousness possible. Machine consciousness, which will eventually emerge from logic and algorithms, is just one instance of a type of consciousness out of infinite number of possibilities.

0

I can answer your question in the affirmative from introspection:

When the molecules in my brain re-obtain a state of equilibrium after a night of binge drinking, I usually regain consciousness.

Which is immediate proof: Properly arranged molecules can have consciousness. Introspection has the added benefit of relieving us of the need to ascertain consciousness: Cogito sciens esse, ergo sciens sum.

How many brain molecules are left after so many baaaad nights — which would establish an upper limit for the minimum needed to have consciousness — is anybody's guess but it's still more than I can count right now, sorry.—

Does anybody have a glass of water for me?

0

I am going to attempt a definition of consciousness in the context of my question. A conscious system exhibits some awareness of itself which manifests in its behaviour. This definition is universal, so it does not require a living system or any kind of organic structure or function. Self-consciousness in the human sense is unnecessary. Can an experiment be designed which tests the null hypothesis "non-living systems have no consciousness"? Quantum physics throws up situations which suggest consciousness, such as entanglement, which is an emergent property. Consciousness may also be an emergent property.

2
  • 1
    Your answer could be improved with additional supporting information. Please edit to add further details, such as citations or documentation, so that others can confirm that your answer is correct. You can find more information on how to write good answers in the help center.
    – Community Bot
    Oct 24, 2023 at 19:36
  • 1
    You need to define consciousness without using the word conscious. Sounds like you are actually referring to self consciousness + agency (vs passive "awareness")
    – Annika
    Oct 24, 2023 at 22:04
0

It does appear prima facie that it cannot be proved that non-living systems do not have consciousness as manifested in their behaviour. Quantum theory throws up the possibility of quantum entanglement - spooky action at a distance, as Einstein called it. If the quantum level exhibits consciousness then it may be a universal property.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .