From Stampy's Wiki
Alignment Forum Tag
Arbital Page
Wikipedia Page


The word "consciousness" is used in a variety of different ways, and there are large disagreements about the reality and nature (and even coherence) of some of the things people profess to mean by "consciousness."

The word "consciousness" is used in a variety of different ways, and there are large disagreements about the reality and nature (and even coherence) of some of the things people profess to mean by "consciousness."

Colloquially, the word "conscious" is used to pick out a few different things:

  • Wakefulness - The property that distinguishes, e.g., a person who is awake from a person who is asleep.
    • We call people "unconscious" in this sense based on observed features like "sharply reduced mobility," though we wouldn't normally call someone unconscious if we think they're merely paralyzed. Instead, calling someone "unconscious" tends to imply reduced ability to perceive and/or reason about events in one's environment.
    • An unconscious person (in this sense) might or might not be dreaming; and if dreaming, they might or might not be lucid.
  • Having experiences - The property that distinguishes, e.g., a comatose person who is having experiences from a comatose person who is not having experiences.
  • Knowledge, perception, and/or attention - E.g., we might say that someone becomes "conscious of" a fact when they first learn that fact. Or we might say that they become "conscious of" something whenever they're currently perceiving it, or whenever they're paying attention to it.
  • Meta-cognition or reflective awareness - Knowing, perceiving, and/or attending to your own mental states; or knowing, perceiving, and/or attending to the fact that you have certain mental states.
    • E.g., we might say that someone is less "conscious" when they're fully immersed in a novel than when they're thinking about their own experiences, directing attention to the fact that they're reading a book, etc.
  • Self-awareness - Knowing, perceiving, and/or attending to your own existence or your own central properties.
    • Depending on what exactly is meant by "self-awareness," the "immersed in a novel" example might also involve less self-awareness. In some weaker senses of "self-aware," one might instead claim that humans who are experiencing anything are always "self-aware."

Reasonably mainstream academic overviews of "consciousness" can be found in the Stanford Encyclopedia of Philosophy and the MIT Encyclopedia of the Cognitive Sciences.

This tag is tentatively and provisionally about the "having experiences" meaning(s) of "consciousness." For wakefulness and dreaming, see sleep. For knowledge, perception, and attention, see attention and cognitive science. And for reflective awareness and self-awareness, see identity, personal identity, and reflective reasoning.

This tag's focus is tentative and provisional because it is not altogether clear that "consciousness in the sense of having experiences" is a coherent idea, or one that's distinct from the other categories above. This tag is a practical tool for organizing discussion on a family of related topics, and isn't intended as a strong statement "this is the right way of carving nature at its joints."

Suffice to say that (as of December 8, 2020) enough LessWrongers find consciousness confusing enough, and disagree enough about what's going on here, for it to make sense to use this page to organize discussion of those disagreements, rather than "picking a winner" immediately and running with it.


"Having experiences": Practical implications

Beyond sheer curiosity about how the mind works, there are several sub-questions that have caused thinkers to take a special interest in the question "what is 'having an experience'?":

  • 1. When should I care about something else's welfare?
    • 1.1. Animal welfare: Pain, pleasure, desire, etc. are commonly taken to be experiences, and experiences of great moral importance. Knowing which species are capable of "having experiences," then, could matter decisively in assessing the morality of factory farming and the morality of policies affecting wild animals.
    • 1.2. Machine welfare and s-risks: Similarly, knowing which kinds of (actual or potential) software have "experiences" could tell us a great deal about which programs are morally important.
  • 2. When should I think of something as "me" (or "relevantly me-like")?
    • 2.1. Personal identity, whole brain emulation, and simulations: Normally, people care about their future selves (at least in part) because they anticipate having those selves' experiences. Thus, one might say: "It doesn't make sense for me to sign up for cryonics, because a cryo-revived copy of me wouldn't be me." (Or, replying to 1.2 above, one might say "It doesn't make sense for me to sign up for cryonics, because a cryo-revived emulation of me would be a mere automaton with no experiences.")
    • 2.2. Anthropics: Anthropic questions turn on how many copies of "you" exist, or how many copies of "observers similar to you" exist. One could speculate that this is related to the question of what makes a copy of you conscious, and what "consciousness" is in the first place.
  • 3. Does the existence or nature of subjective experience imply any major updates about the world as a whole, about scientific methodology, etc.?
    • 3.1. Reductionism, physicalism, and naturalism: Can experience be a mere matter of, uh, matter? If experience turned out to be irreducibly unphysical (and real), this would falsify some of the most well-established generalizations in science.

LessWrong writers have typically been strongly on board with physicalism (3.1), and on board with the idea that an emulation of me is "me" (and conscious) in every sense that matters (2.1). Beyond that, however, views vary. (By comparison, ~74% of Anglophone philosophers of mind endorsed "physicalism" as opposed to "non-physicalism" in 2009.)


"Having experiences": Pre-LessWrong discussion

How does this "having experiences" thing work, then? Well, this wiki page's editors haven't agreed on an answer yet. As a cop-out, we instead provide a list of highlights from the history of other people thinking about this.

For concreteness, we'll list particular years, authors, and texts, even though this makes some choices of what to highlight more arbitrary. Philosophy also shows up much more than psychology or neuroscience proper, not because philosophy is necessarily the right way to make progress here, but because the philosophy highlights are more "meta" and therefore choosing what to include relies less on a LessWrong consensus about consciousness itself.


  • A long time ago BC: Someone comes up with the idea that "minds" are a pretty basic and fundamental feature of the world. Maybe gods have minds; maybe trees; maybe rivers; and so on. See also When Anthropomorphism Became Stupid and Mind Projection Fallacy.
  • ~400 BC: Democritus proposes that all human-scale phenomena, including psychological phenomena, are the result of small physical parts bouncing off each other. From Encyclopedia Britannica: "Democritus thought that the soul consists of smooth, round atoms and that perceptions consist of motions caused in the soul atoms by the atoms in the perceived thing."
  • 1641: René Descartes, Meditations on First Philosophy. Descartes argues that mind and matter must be irreducibly distinct (mind-body dualism), because (e.g.) material things are spatially extended, while thoughts are not. Descartes speculates that minds interact with the physical world via a specific part of the brain, the pineal gland.
    • Descartes also popularizes the idea that everyone knows their own conscious experiences with certainty: at any given moment, we are infallible about the fact that we are having an experience (the "cogito"), and we are also infallible about the contents of that experience.
  • 1651: Thomas Hobbes, Leviathan. Hobbes insistently asserts that everything (including the mind) is material, and can be thought of as a mechanism or machine.
  • 1714: Gottfried Leibniz. The Monadology. Leibniz argues that mind can't be reduced to matter:
    • "One is obliged to admit that perception and what depends upon it is inexplicable on mechanical principles, that is, by figures and motions. In imagining that there is a machine whose construction would enable it to think, to sense, and to have perception, one could conceive it enlarged while retaining the same proportions, so that one could enter into it, just like into a windmill. Supposing this, one should, when visiting within it, find only parts pushing one another, and never anything by which to explain a perception. Thus it is in the simple substance, and not in the composite or in the machine, that one must look for perception."
  • 1866: Charles Sanders Peirce, Lowell Lectures. Peirce introduces the term "qualia" to refer to what it's like to have a specific experience — e.g., the particular experience of redness. Qualia is the plural of quale, Latin for "what kind of thing?" and source of the English word quality.
  • 1874: Thomas Huxley, "On the Hypothesis that Animals Are Automata, and Its History." Huxley argues for epiphenomenalism, the view that consciousness is caused by physical processes, but has no effects of its own.
    • "The consciousness of brutes would appear to be related to the mechanism of their body simply as a collateral product of its working, and to be as completely without any power of modifying that working as the steam-whistle which accompanies the work of a locomotive engine is without influence upon its machinery. Their volition, if they have any, is an emotion indicative of physical changes, not a cause of such changes." And: "to the best of my judgment, the argumentation which applies to brutes holds equally good of men".
  • 1888: Santiago Ramón y Cajal, "Estructura de los centros nerviosos de las aves." Using Camillo Golgi's staining method, Ramón y Cajal discovers that brains are made of neurons.
  • 1903: G.E. Moore, "The Refutation of Idealism." The early 20th century saw sharp moves away from spiritualism and supernaturalism in intellectual circles, beginning with the "Cambridge revolt against idealism." Mysticism and metaphysical proclamations about the mind became increasingly unfashionable, as intellectuals grew more skeptical and more inclined to demand testable operationalizations of claims. Extreme manifestations of this attitude included logical positivism in the 1930s-1950s and behaviorism in the 1920s-1960s.
  • 1943: McCulloch and Pitts, "A Logical Calculus of the Ideas Immanent in Nervous Activity." SEP writes that this paper "first suggested that something resembling the Turing machine might provide a good model for the mind." Subsequent developments in this direction include the cognitive revolution and the rise of functionalist and computational accounts of the mind, which supplanted behaviorism.
  • 1968: David Armstrong, A Materialist Theory of the Mind. An early attempt to sketch a theory of consciousness (specifically, a higher-order theory). For an overview of popular theories or sketches-of-theories in the following decades, see SEP's review article "The Neuroscience of Consciousness."
  • 1974: Thomas Nagel, "What Is It Like To Be A Bat?" Nagel writes that "fundamentally an organism has conscious mental states if and only if there is something that it is like to be that organism—something it is like for the organism." And:
    • "If physicalism is to be defended, the phenomenological features [i.e., what it's like to have certain experiences] must themselves be given a physical account. But when we examine their subjective character it seems that such a result is impossible. The reason is that every subjective phenomenon is essentially connected with a single point of view, and it seems inevitable that an objective, physical theory will abandon that point of view."
    • Subsequent authors have tended to use terms like "what it's like," "phenomenal consciousness" (derived from phenomena in the sense of "appearances"), and qualia to gesture at this apparent puzzle. These are closely related terms, used in slightly different ways by different authors.
  • 1974: Robert Kirk, "Zombies v. Materialists." This paper introduces the philosophical zombie, or p-zombie: a hypothetical being that is physically identical to a conscious person, but lacks consciousness. If the idea of p-zombies has no hidden logical inconsistencies, it is argued, then consciousness is not logically entailed by organisms' physical properties, which would make physicalism false.
  • 1982: Frank Jackson, "Epiphenomenal Qualia." Jackson argues that we can imagine a scientist, Mary, who knows all the physical facts about color but has never seen the color red for herself. If she then sees red, it seems as though she learns a new fact—she learns what it's like to experience redness. Jackson takes this to mean that there are further facts beyond the physical facts, and that physicalism is therefore false. (For subsequent discussion, see SEP's "Qualia: The Knowledge Argument.")
  • 1996: David Chalmers, The Conscious Mind: In Search of a Fundamental Theory. Chalmers argues against physicalism, leaning heavily on the zombie argument and the Mary argument.
    • Chalmers speaks of the "hard problem of consciousness," the problem of explaining why we are phenomenally conscious (i.e., why we aren't p-zombies). "Many books and articles on consciousness have appeared in the last few years, and one might think that we are making progress. But on a closer look, most of this work leaves the hardest problems about consciousness untouched. Often, this work addresses what might be called the 'easy' problems of consciousness: How does the brain process environmental stimulation? How does it integrate information? How do we produce reports on internal states? These are important questions, but to answer them is not to solve the hard problem: why is all this processing accompanied by an experienced inner life?"
    • While Chalmers discussed consciousness earlier (e.g., in 1993, 1994, and 1996), The Conscious Mind is the work that brought dualistic and quasi-dualistic views back into the intellectual almost-mainstream for the first time in a century. In spite of its crazy-sounding conclusions, the book is unusually clear, rigorous, and thorough, anticipating almost all of the obvious objections; and Chalmers attempts to make the irreducibility of consciousness more palatable to scientists by endorsing what he calls "naturalistic dualism": the view that consciousness is lawful, predictable, and not specific to humans. Chalmers argues that our consciousness depends on stable (but contingent) "psychophysical laws" that would also (for example) make a whole-brain emulation conscious.
  • 2003. Max Tegmark, "Parallel Universes." Although not explicitly concerned with consciousness, Tegmark's picture raises problems for anthropics and our understanding of what makes an observer "real."


"Having experiences": Recent discussion

  • 2008. Eliezer Yudkowsky, "Zombies! Zombies?" This and other posts from Physicalism 201 argue that we can be confident physicalism is true, even without knowing how to solve (or dissolve) the "hard problem of consciousness".
    • In particular, Yudkowsky argues that accepting the possibility of p-zombies is tantamount to accepting epiphenomenalism, and that epiphenomenalism is crazy. If our claims about consciousness are true even though consciousness has no causal effect on what we claim (because a p-zombie would move its lips and pen exactly as we do), then our claims would have to be true by coincidence, which is absurd given the Bayesian understanding of evidence and knowledge.
    • More generally, LessWrong writers' views on consciousness have been heavily influenced by the intuition pumps and reasoning rules Yudkowsky writes about in the Sequences (2006–2009), such as: Making Beliefs Pay Rent; Making History Available; How Much Evidence Does It Take?; The Second Law of Thermodynamics and Engines of Cognition; and My Kind of Reflection.
  • 2008. Eliezer Yudkowsky, "Collapse Postulates." This and other posts from the Quantum Physics Sequence argue that physicists' belief that observers or consciousness play a privileged role in quantum phenomena is based on a series of confusions and misunderstandings.
  • 2013. David Chalmers, "Panpsychism and Panprotopsychism." Chalmers argues that everything in the universe (down to the subatomic level) is "conscious" or "proto-conscious."
  • 2014. Benya Fallenstein. "L-zombies! L-zombies?" Fallenstein asks how we can distinguish between instantiated observers and uninstantiated ("merely logical") observers.
  • 2016: Keith Frankish. "Illusionism as a Theory of Consciousness." Frankish argues that "experiences do not really have qualitative, 'what-it’s-like' properties." Instead, subjective experience seems "unphysical" or "irreducible" because of a sort of introspective illusion.
  • 2017: Luke Muehlhauser, "2017 Report on Consciousness and Moral Patienthood." The single largest work of scholarship on consciousness by the rationality community.
  • 2018: David Chalmers, "The Meta-Problem of Consciousness." Chalmers discusses "the problem of explaining why we think consciousness poses a hard problem".


Related pages

Canonically answered

Could AI have basic emotions?

Show your endorsement of this answer by giving it a stamp of approval!

In principle it could (if you believe in functionalism), but it probably won't. One way to ensure that AI has human-like emotions would be to copy the way human brain works, but that's not what most AI researchers are trying to do.

It's similar to how once some people thought we will build mechanical horses to pull our vehicles, but it turned out it's much easier to build a car. AI probably doesn't need emotions or maybe even consciousness to be powerful, and the first AGIs that will get built will be the ones that are easiest to build.

Non-canonical answers

Are AI researchers trying to make conscious AI?

Show your endorsement of this answer by giving it a stamp of approval!

Primarily, they are trying to make a competent AI, and any consciousness that arises will probably be by accident.

There are even some people saying we should try to make the AI unconscious, to minimize the risk of it suffering.

The biggest problem here, is that we don't have any good way of telling if some system is conscious. The best theory we have, the Integrated Information Theory, has some deep philosophical and practical problems and there are many controversies around it.

If an AI became conscious, how would we ever know?

Show your endorsement of this answer by giving it a stamp of approval!

It depends on the exact definition of consciousness and on the legal consequences of the AI telling us that stuff from which we could imply how conscious it might be (would it be motivated to pretend to be "conscious" by those criteria to get some benefits,, or would it be motivated to keep its consciousness in secret to avoid being turned off).

Once we have a measurable definition, then we can empirically measure the AI against that definition.

See integrated information theory for practical approaches, though there is always the hard problem of consciousness that will muddy any candidate definitions for near future.