Scientifically, given that we are just pieces of universe (earth), then parts of the universe can experience phenomenal consciousness.

"Panpsychism" says that therefore phenomenality must be inherent to (or at least ubiquitous in) what the universe is made of at some fundamental level. This can bring to mind absurd anthropomorphic ideas of particles or fields experiencing complex things, but sometimes it's referred to as pan proto-psychism. Bertrand Russell apparently held some version of the above though it seems it fell out of favour, but has had a mild renewal of interest (Goff etc)? But has "combination problem" and "binding problem" etc etc.

Whereas functionalism only commits to certain functional arrangements of matter being phenomenally conscious. Often based on billions of years of evolution-by-natural-replication-and-selection having created new complexity. But can also include simple switches having some unit of consciousness, which I think the physicist Michio Kaku holds to.

But firstly it seems that the function of something can only be defined by reference to a wider context. Either the evolutionary past (which in an eternalist block universe also exists, though I don't know if that's relevant) or for example the outside of the skull as well as the inside. I don't mean to get into brains in vats etc. But just, if say a function is to react to light hitting the eye, to then guide behaviour, somehow the brain knows that's its function, because it has qualia of say a green object out there.

Even though that function is not inherent to that pattern of electrochemical neural activity, but only to the whole context (inter not just intra). See also philosopher of biology McShea on how at any moment we are guided by our environment just as much as we guide to our own ends. For example just as a torpedo locks onto a ship, the ship is guiding the torpedo. He gives a number of non-organic examples which make the same point, which also highlight that the apparent teleology of evolution can be found nonorganically too, because ultimately it's all the same physical stuff.

Therefore, how do "functions" not go down to the micro or quantum level like panpsychism?

[Edit by another person: Or, put another way, if teleological notions of biological functionality are so hard to paraphrase away, then does functionalism become panpsychism in the limit, or is there a way to block such a transformation?]

  • 1
    The title doesn’t really match the concluding question right now. That said for your last question, originally epiphenominalism was that consciousness is not functional; its inconsequential residue of biological functions. Therefore it’s not teleological and ubiquitous across all of spacetime. But even more generally, new functions are hardly teleological toward any comprehensible end in the far future. Unpredictable, uncountable. See Stuart Kauffman for this wholesale biological functional unpredictability. It echos Wittgenstein’s impossible to enumerate uses of language in a way.
    – J Kusin
    Commented Oct 7, 2023 at 22:24
  • 2
    Hinted by the German philosopher Heidegger's thesis there may be a way to block your contemplated teleological version of functionalism to slippery slope to the uncountably inaccessible and inexhaustible panpsychism which is nothing but Dasein with a critical sailing affective mood insofar as 'being-with' here or there in the 'thrown' context of its 'caring' facing with its own 'uncanny'. Such modified affection and ownmost uncanny contribute to block your potential infinite regress qua recursive functions themselves with countable layers of their supposed causal telos account upon layers... Commented Oct 7, 2023 at 23:24
  • 1
    @Yop I accept that you did not explicitly have teleological questions in view, but between your expression of uncertainty about your own question, and the Community bot's comment, modulo the references to functionalism and panpsychism in the title, I figured the OP could be rounded off/summed up with a nod towards biological teleology re: general intentionality. If this is not apropos of your point, I will freely revoke my edit. Commented Oct 8, 2023 at 1:02
  • 2
    Scientific American, Sept '23: Is Consciousness Part of the Fabric of the Universe? Commented Oct 8, 2023 at 12:49
  • 1
    Smolin's energetic causal set theory does literally mean on the basic level there're only (partial) views or visions of events and a unit of consciousness is not a single qualia, but the entire of a view of the universe as seen from one event. Inspired by Tononi's IIT, the way to block panpsychism for low level simple views of particles is to use some complexity measure based on difference of new view compared with its historical sky views, and since all low level views are near copies without major difference throughout the cosmos, thus they don't have consciousness in the sense of free will. Commented Oct 11, 2023 at 7:03

2 Answers 2


Short answer

There is no intrinsic aspect of functionalism that prevents it from going all the way down to elementary particle state changes, and for consciousness to be associated with every single function in our universe.

Longer answer

Functionalists realized this from the outset, and added an "emergence threshold" assumption into functionalism, where the nature and magnitude of this threshold was TBD. The initial assumption among functionalists was that computational thruput was the threshold, and our early electronic calculating machines were just too low in thruput to be conscious, while humans with our very complex brains, were above the threshold.

The drastic increase in computer thruput over the decades, with no apparent achievement of consciousness, has led to the abandonment of thruput as the hypothesized threshold property for consciousness to emerge.

There are multitudes of other alternative threshold criteria that have been hypothesized. Organizational complexity (a version of which IIT uses, but without the threshold), only self-referential Higher Order Processing, only functions that use a Global Workspace, only functions that operate on a neural net, etc.


In general, these functionalist approaches to consciousness have suffered from failing falsification tests. The thruput test case is the first example, but all of the others have suffered the same fate. The vast majority of our neural net processing is unconscious, contrary to that theory. Our unconscious processing seems to do Higher Order thinking just fine. Our unconscious processing is on the same network as the conscious processing, contrary to IIT. Etc, etc.

Functionalism also suffers from the problem of "what is a function". Functions are abstract objects. They are in the category of math, and logic. A function is a function whether it is being "implemented" or not. Functionalism needs to have "implemented" added to it, to prevent the paper text of a code from being as conscious as a computer implementing that code.

So now -- what is "implementing". That one traces a logic path thru a specific network of logic gates and memory locations, or synapses and axons, begs the question of why not trace this logic thru a different set of molecules, or cellular structures. Logically, one can "implement" a function by reassigning values as readily as changing material states. Hence Searle's critique of computationalism -- the relations between the molecules on his desk can and are implementing all possible functions, just by reassigning of variables moment to moment. So is his desk conscious? This is the same binding problem that panpsychism suffers from. WHY would consciousness anneal to humans?

The causal irrelevancy of consciousness in both worldviews also causes them both to be falsified by James and Popper's Evolutionary Test Case for consciousness (see this answer: What are the main arguments that emergence-based theories of consciousness can't solve the hard problem of consciousness?).


I do not know. But it is a good question. My understanding is that functionalism refers to minds. Panpsychism proposes consciousness at the universal and fundamental levels. Panpsychism is not concerned with minds per se. As I understand it, the combination problem and binding problem refer to minds. So they are irrelevant to panpsychism. I am not convinced that philosophical theories are amenable to compromise, and I doubt that this is the case here. On reflection, my answer to your question is no.

You must log in to answer this question.

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