The Problem of Consciousness: Easy, Hard or Tricky? (original) (raw)

Abstract

Phenomenal consciousness presents a distinctive explanatory problem. Some regard this problem as ‘hard’, which has troubling implications for the science and metaphysics of consciousness. Some regard it as ‘easy’, which ignores the special explanatory difficulties that consciousness offers. Others are unable to decide between these two uncomfortable positions. All three camps assume that the problem of consciousness is either easy or hard. I argue against this disjunction and suggest that the problem may be ‘tricky’—that is, partly easy and partly hard. This possibility emerges when we recognise that consciousness raises two explanatory questions. The Consciousness Question concerns why a subject is conscious rather than unconscious. The Character Question concerns why a conscious subject’s experience has the phenomenology it has rather than some other. I explore the possibility of one or other of these explanatory challenges being hard and the other easy, and consider the dialectical ramifications this has for all sides of the debate.

Access this article

Log in via an institution

Subscribe and save

Buy Now

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

Notes

  1. At times Chalmers builds non-functionality into his definition of the hard problem (e.g. 1997, p. 4). The need to diverge from his way of framing things will emerge over the course of the paper and will be addressed directly in Sect. 5.2.
  2. The phrase ‘explanatory gap_’_ was introduced by Levine (2002) and the phrase ‘epistemic gap’ (Chalmers 2002) is used in much the same way. The majority of philosophers believe that there is an explanatory gap and are thus Pessimists (although there is disagreement within this group about the metaphysical consequences of this gap). Those who deny that there is an explanatory gap (i.e. Optimists) are in the minority but still form a substantial contingent. This is reflected in Bourget and Chalmers (2013) survey of what philosophers believe. The main group in this survey consisted of 931 faculty philosophers. One of the 30 questions asked concerned the conceivability of zombies, which is a litmus test for whether one believes that there is an explanatory gap. 58.9 % regard zombies as conceivable (though less than half of those take them to be metaphysically possible) and 16.0 % take zombies to be inconceivable. 25.1 % answered ‘other’, and we can speculate that some of this group could accurately be described as ‘undecided’. In Sect. 3.3 I consider more closely the role zombie intuitions play in the debate.
  3. Block captures this kind of worry with his (2002) notion of ‘the harder problem of consciousness’.
  4. Epiphenomenalism runs into trouble with the role of consciousness in the production of behaviour, the function of consciousness, the evolutionary origin of consciousness and the possibility of knowledge of conscious states (see especially Kim 1989, 2002). Panpsychism suffers from a dramatic lack of parsimony and leaves human conscious experience unexplained (Stoljar 2006, p. 120). Type-B positions suffer from more technical difficulties concerning the credibility of necessitation relations that are epistemically opaque even to an ideal subject (Chalmers 1997, pp. 11–16).
  5. Chrisley’s suggestion that the constitutive form gets the most attention is a comment on Artificial Intelligence. It is not clear that the same holds for metaphysics.
  6. Explicit articulations of this distinction can be found in Levine (2003, p. 104), Kriegel (2009, p. 5) and Rosenthal (2011, p. 435).
  7. This kind of objection is pushed by Churchland (1996b) and Stoljar (2006).
  8. Subjectivity is a wildly ambiguous term. In line with Levine (2003) and others, I am using the term in the specific sense just outlined.
  9. For Levine, Kriegel and Chalmers the inexplicability of subjectivity is in turn diagnosed in terms of its non-functionality, though this further claim is optional.
  10. For more on the intrinsic/extrinsic distinction and its place in the problem of consciousness see Pereboom (2011).
  11. Chalmers and others hold that the purely structural nature of the physical is responsible for the apparent openness of not just the Character Question, but of the Consciousness Question too. I explore this possibility further in Sect. 5.2.
  12. On the distinction between creature-based and state-based research paradigms see Bayne (2007).
  13. Of course, if we had some explanation that entailed that S is having a reddish experience, we would thereby have an explanation that entailed that S is having an experience at all, thus answering the Consciousness Question. This is why it is important that the Character Question is framed conditionally—we are asking what phenomenology S enjoys if S is conscious, without committing to S actually being conscious.
  14. This limited explanation of precipitation would presumably depend on our being ignorant of some of the relevant facts. After all, if we had all the facts about atmospheric conditions we could surely deduce that it will precipitate. In contrast, our explanation of consciousness would be limited even when all the physical facts are in (or so advocates of the Hard/Easy View say). This disanalogy reflects the fact that there is no hard problem of precipitation! Nevertheless, cases like this partial explanation of precipitation should help us make sense of what a conditional explanation of qualitative character would look like. Thanks to Donnchadh O'Connail for pushing this point.
  15. This ties in closely with Chrisley’s notion of ‘lagom AC’ which is a middle ground between ‘strong’ and ‘weak’ AC. On the strong approach, the goal of AC is to create artificial systems that are sufficient for consciousness. On the weak approach, the aim is to create artificial systems that model consciousness—much like a computer simulation models a hurricane—but which need not themselves instantiate properties of explanatory relevance to consciousness. Both positions face serious objections that lagom AC promises to avoid. On lagom AC, the aim is to create artificial systems that instantiate properties of explanatory relevance to consciousness but which need not be sufficient for consciousness. One way of implementing this strategy is to pursue artificial systems that have the properties that differentiate experiences from one another, but which do not necessarily have the properties required for consciousness.
  16. Interestingly, Chalmers (forthcoming) shows sympathy for the idea that it is inexplicable how we get subjects out of non-subjects (i.e. out of purely objective constituents). He regards this as an objection specifically to panprotopsychism. But if the objective/subjective divide is a problem, surely it is a problem for any theory that attempts to account for consciousness in objective terms including any theory offered by cognitive scientists? As such, the possibility should be taken seriously that this is the explanatory divide that’s driving our intuition of openness.

References

Download references

Acknowledgments

This paper was completed with the support of ERC grant 313552 - "The Architecture of Consciousness". I am grateful for the detailed comments of Tim Bayne, Donnchadh O'Connail and two anonymous referees. Thanks also to the audience at the inaugural iCog conference at the University of Sheffield for comments on an early presentation of this material.

Author information

Authors and Affiliations

  1. University of Manchester, Manchester, UK
    Tom McClelland
  2. Arthur Lewis Building, University of Manchester, Oxford Road, Manchester, M13 9PL, UK
    Tom McClelland

Authors

  1. Tom McClelland
    You can also search for this author inPubMed Google Scholar

Corresponding author

Correspondence toTom McClelland.

Rights and permissions

About this article

Cite this article

McClelland, T. The Problem of Consciousness: Easy, Hard or Tricky?.Topoi 36, 17–30 (2017). https://doi.org/10.1007/s11245-014-9257-4

Download citation

Keywords