Skip to main content
Log in

The transcendental deduction of Integrated Information Theory: connecting the axioms, postulates, and identity through categories

  • Original Research
  • Published:
Synthese Aims and scope Submit manuscript

Abstract

This paper deals with a foundational aspect of Integrated Information Theory (IIT) of consciousness: the nature of the relation between the axioms of phenomenology and the postulates of cause-effect power. There has been a lack of clarity in the literature regarding this crucial issue, for which IIT has received much criticism of its axiomatic method and basic tenets. The present contribution elucidates the problem by means of a categorial analysis of the theory’s foundations. Its main results are that: (i) IIT has a set of nine fundamental concepts of reason, called categories, which constitute its categorial lexicon and through which it formulates a system of principles incorporating the axioms, the postulates, and the central identity; and (ii) the connection between the axioms and postulates is grounded by their common root in this categorial lexicon, the categories of which find their justification by means of a phenomenological and transcendental deduction. Some further results are the unique origin of axioms and postulates in the categories; the distinction between conceptual and formalized postulates; a clarification of the uniqueness problem of categorial lexica in general; and an IIT account of objectivity by explicating how the physical is (re)defined by means of categories. All of this is put to use against various criticism targeting IIT’s theoretical core. If successful, the proposed interpretation illuminates a central issue in the contemporary study of consciousness and contributes to an environment of mutual understanding between defenders and critics of the theory.

This is a preview of subscription content, log in via an institution to check access.

Access this article

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

Instant access to the full article PDF.

Similar content being viewed by others

Data availability

Not applicable.

Code availability

Not applicable.

Notes

  1. For instance, concepts as phenomenal invariances are general, yet they are still specific contents of consciousness.

  2. Nothing is introduced through the back door, “atom” is just a name for the smallest unit that can be assessed operationally.

  3. This principle amounts to a methodological reductionism which should not be conflated with ontological reductionism (see Tononi, 2017b; Grasso et al., 2021).

  4. Nothing is surreptitiously added by talking of “causal distinctions” and “causal relations”. These are already contained in the category of structuredness, as the notion of a structure implies the notion of a part of the structure and of a relation between the parts. “Distinction” is just a name for a part of the cause-effect structure which is in “relation” to other parts.

  5. To emphasize the existence category, the identity could be expressed as “a particular experience exists as identical with…” without any impact on the theory.

  6. See Tononi (2017b) for a more extended discussion of IIT’s ontological taxonomy.

  7. See Bieri et al. (1979), Stern (1999, 2000, 2019), and Bitbol et al. (2009) for general discussion and debate on the scope, merits, weaknesses, and prospects of transcendental arguments, in both science and philosophy. There is also much recent work on transcendental arguments in epistemology (e.g., Schafer, 2021) and in science (Hoffman 2019; Colombo & Wright 2021).

  8. To be clear, concerning its origin, the identity principle itself can only be formulated and grounded by means of the categories (and the law of identity used extra-logically), even though it is employed afterward to explain the connection between the phenomenal and the causal domains. Not only an identity, rather than, e.g., emergence or co-variance, ensures parsimony of categories and justifications, but it also gives sufficiency of explanation in the long run, by opening a space of empirical possibilities for confirming the theory (see also Chis-Ciure 2022c). I discuss below the issue of whether a categorial lexicon can be proved necessary and sufficient apriorily.

  9. In personal communication, Giulio Tononi claimed that, in the case of IIT’s categories, uniqueness stems from the (conditional) completeness of the axioms and their translation into postulates which, coupled with the identity, can be in-principle experimentally probed. However, one still needs the test of experience to determine the explanatory completeness of IIT’s principles beyond conditional status. Thus, an a priori argument could be made that IIT’s categorial lexicon is unique if the axioms and the postulates are complete for the task, but this would be a weak, conditional uniqueness, the merits of which are not currently evident. Yet I believe strong, unconditional uniqueness of categories in a scientific context is an a posteriori affair.

  10. A historical parallel: equating thinking with judging, Kant ([1781/1786]1998) believed he discovered the twelve immutable pure concepts of understanding. Ironically, perhaps with the exception of the thing-in-itself, no other part of the Critique was more virulently attacked afterward than his categories table and their deduction.

  11. See Ellia et al. (2021, ch. 4) for a different strategy in answering the unfolding argument, and Chis-Ciure & Ellia (2021) for a defense of IIT against Chalmers’ conceivability argument that employs some overlapping resources.

  12. To clarify, the question of whether a categorial framework is unique for a theoretical task is separate from whether the postulates can be uniquely translated from the axioms. My answer to the first is no, at least when it comes to an a priori proof; my answer to the second is yes, by means of an a priori proof.

  13. Logical relations and laws being taken for granted, e.g., the identity principle.

References

Download references

Acknowledgements

I am very grateful to Giulio Tononi for his patience in giving me feedback on an earlier draft, as well as his generosity in filling me in on IIT aspects that were never published before. He was a necessary condition for the possibility of this paper in many ways. Moreover, I thank Larissa Albantakis for greatly improving the paper and my thinking on key matters. Garrett Mindt was instrumental in getting me to write about but not like German Idealists—I’m in his debt for the willingness to read and comment on multiple drafts of this paper. Precious feedback on a first draft is one of the many things I have to thank Matteo Grasso for. I also appreciate Andrew Haun’s openness to discussing obscure philosophical issues, quite removed from his interests. In addition, I benefitted greatly from Jon Mallatt’s suggestions, which are always thorough and to the point. Francis Fallon was so kind as to help me not put reviewers in a position to have to meet me halfway. I am also grateful to Mircea Dumitru for his generous support in all academic matters. Finally, I want to thank two anonymous reviewers who corrected missteps and made the argument punchier. All remaining faults of the paper are solely due to the author.

Funding

This work was supported by the Romanian-U.S. Fulbright Commission under a Fulbright Student Fellowship.

Author information

Authors and Affiliations

Authors

Contributions

The author is the sole contributor to this manuscript.

Corresponding author

Correspondence to Robert Chis-Ciure.

Ethics declarations

Conflict of interest

The author has no conflicts of interest to declare that are relevant to the content of this article.

Consent to participate

Not applicable.

Consent for publication

Not applicable.

Ethical approval

Not applicable.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Chis-Ciure, R. The transcendental deduction of Integrated Information Theory: connecting the axioms, postulates, and identity through categories. Synthese 200, 236 (2022). https://doi.org/10.1007/s11229-022-03704-z

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/s11229-022-03704-z

Keywords

Navigation