Abstract
The mechanistic account of computation offers one promising and influential theory of computational implementation. The aim of this paper is to shore up its conceptual foundations by responding to several recent challenges. After outlining and responding to a recent proposal from Kuokkanen (Synthese 200:247, 2022a), I suggest that computational description should be conceptualised as a form of idealisation (selectively attending to modified subsets of model features) rather than abstraction (selectively attending to subsets of features within a target system). I argue that this conceptualisation not only offers the best way of making sense of computational implementation, but also a way of resolving each of the outstanding challenges facing the mechanistic account. The idealisation view allows the mechanistic account to make sense of the omission process found in computational descriptions without leaving the relationship between physical and computational properties mysterious.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Complementing the question of computational implementational is the question of computational individuation, i.e., the task of explaining under which conditions it is true or false to say of a physical system that it implements a particular computational model rather than another, e.g., AND versus OR (Sprevak, 2018). While important, the current paper focuses only on computational implementation for matters of scope. It leaves discussion open as to which account of computational individuation best fits with the mechanistic account, though it is worth noting that authors diverge on the issue (see, e.g., Fresco, 2010, 2021; Dewhurst, 2018; Coehlo Mollo, 2018; Fresco & Milkowski, 2021; Fresco, 2021).
For Piccinini (2020), a teleological function is a stable contribution towards a goal of an organism. Goals can be either biological or nonbiological. Biological goals include survival, development, reproduction or helping, whereas nonbiological goals are any other goals pursued by an organism, such as using a Tobacco pipe to hold tobacco.
Relevant here is the distinction between normative teleological functions and perspectival functions. For Piccinini (2015, 2020), teleological functions are objective in the sense that they are stable causal contributions to the goals of an organism. The functions can be individuated objectively in virtue of some combination of the organism and its environment. Perspectivalism, on the other hand, which says that functional attributions are relative to observer-interests (and hence subjective), is ill-suited to account of the functions of objects (artefactual or biological) because it does not do justice to our scientific practices – some traits have multiple functions, but not in virtue of multiple perspectives, e.g., the function of the heart pumping blood (see Piccinini (2015, p.103) for details. For further discussion of the perspectivalist position, see Schweizer (2019) or Lee (2021).
While I cannot defend the particular take of the mechanistic account on computational explanation here, for a defense, see Piccinini (2015, p.142).
There are other desiderata that are sometimes offered, such as miscomputation and non-circularity, but these three broadly capture some of the features regularly offered by theorists in favour of the mechanistic account (see, e.g., Miłkowski, 2013; Fresco, 2014; Piccinini, 2015; Sprevak, 2012, 2018).
For discussion of further problems that have been raised, such as the decomposition problem, see Shagrir (2022, Ch. 6).
Haimovici (2013) was the first to point out the tension, albeit in a slightly different form. The focus there was on the relation between functional and structural properties, but the conclusion is similar.
The original distinction is owed to Mäki (1992).
Kuokkanen (2022a) suggests that one can maintain the vertical-horizontal distinction, even if computational descriptions are arrived at by horizontal descriptive abstraction. The distinction is a more general feature of modelling practice.
For an alternative solution to the abstraction problem, see Kersten (2020).
To be clear, Kuokkanen (2022a, b) is more interested in analysing Piccinini’s version of MAC (e.g., how to interprets Piccinini’s talk about abstraction) and offering some theoretical clarifications than arguing for or defend the mechanistic account. Here I am simply extending the analysis to mechanistic computation more generally insofar as it offers a solution to the three challenges.
Schweizer (2019) articulates a somewhat related idea. However, while his “computational perspectivalism” also emphasises computational descriptions as a form of idealisation, it does not do so from within a mechanistic view, nor does it develop a conception of computation as idealisation specifically with an eye to explicating physical computation. The view is also explicitly non-objective and observer-dependent.
For further detail and defence of the view, see Portides (2018).
This characterisation of idealisation is similar to ‘Galilean’ and ‘minimalist’ forms, in that it also can involve distortions and false assumptions under the guidance of a representational ideal, but it does not require that its modifications, strictly speaking, have to be false (see Weisberg, 2007).
To be clear, while it may be the case that idealisation is involved in computational explanation more generally, the current proposal is only meant to apply to the mechanistic account. The question of the nature of computational explanation is a complex and vexed one. I am here simply working under the assumption that the mechanistic account offers one plausible account of computational explanation.
It is worth mentioning that Shagrir (2020, 2022) marshals this point, in his master argument, to argue for more than one possibility open for mapping. However, the argument, while important, applies more directly to account of individuation rather than implementation, which is the present concern, and so it worth setting it to one side for the moment.
For a similar worry based on interpretative models in cognitive neuroscience, see Chirimuuta (2014).
Chirimuuta (2014) argues that it is efficient coding which delineates the set of counterfactual dependencies between input to the system (e.g. sensory information) and/or system requirements (e.g. task for which information is needed) and the computational properties of the system. For replies showing how efficient coding might be integrated with the mechanistic account, see Wajnerman Paz (2017) or Fuentes (2023).
For an externalist reading of this point, see Kersten (2017).
For further discussion of the relation, see Piccinini (2018).
References
Bechtel, W., & Shagrir, O. (2011). The non-redundant contributions of Marr’s three levels of analysis for explaining information processing mechanisms. Topics in Cognitive Science, 7(2), 312–322. https://doi.org/10.1111/tops.12141.
Boone, W., & Piccinini, G. (2016). Mechanistic abstraction. Philosophy of Science, 83, 686–697. https://doi.org/10.1086/687855.
Chalmers, D. (1994). On implementing a computation. Minds and Machines, 4(4), 391–402. https://doi.org/10.1007/BF00974166.
Chalmers, D. (2011). A computational foundation for the study of cognition. Journal of Cognitive Science, 12(1), 323–357.
Chirimuuta, M. (2014). Minimal models and canonical neural computations: The distinctness of computational explanation in neuroscience. Synthese, 191(2), 127–153. https://doi.org/10.1007/s11229-013-0369-y.
Chrisley, R. (1995). Why everything doesn’t realize every computation. Minds and Machines, 4, 403–430.
Coelho Mollo, D. (2018). Functional individuation, mechanistic implementation: The proper way of seeing the mechanistic view of concrete computation. Synthese, 195, 3477–3497. https://doi.org/10.1007/s11229-017-1380.
Craver, C. (2007). Constitutive explanatory relevance. Journal of Philosophical Research, 32, 1–20. https://doi.org/10.5840/jpr_2007_4.
Craver, C. (2013). Functions and mechanisms: A perspectivalist account. In P. Huneman (Ed.), Functions: Selections and Mechanisms (pp. 133– 158). Dordrecht: Springer.
Dewhurst, J. (2018). Individuation without representation. British Journal for the Philosophy of Science, 69(1), 103–116. https://doi.org/10.1093/bjps/axw018.
Elber-Dorozko, L., & Shagrir, O. (2018). Computation and levels in the cognitive and neural sciences. In M. Sprevak, & M. Colombo (Eds.), The Routledge Handbook of the computational mind (pp. 205–225). Routledge. https://doi.org/10.4324/9781315643670.
Elber-Dorozko, L., & Shagrir, O. (2019). Integrating computation into the mechanistic hierarchy in the cognitive and neural sciences. Synthese. https://doi.org/10.1007/s11229-019-02230-9.
Fodor, J. (1981). Representations: Philosophical essays on the foundations of Cognitive Science. MIT Press. A Bradford Book.
Fresco, N. (2010). Explaining computation without semantics: Keeping it simple. Minds and Machines, 20, 165–181. https://doi.org/10.1007/s11023-010-9199-6.
Fresco, N. (2014). Physical computation and Cognitive Science. Springer.
Fresco, N. (2021). Long-arm functional individuation of computation. Synthese, 199(6), 5. https://doi.org/10.1007/s11229-021-03407-x.
Fresco, N., & Miłkowski, M. (2021). Mechanistic computational individuation without biting the bullet. British Journal for the Philosophy of Science, 72(2), 431–438.
Fuentes, J. (2023). Efficient mechanisms. Philosophical Psychology. https://doi.org/10.1080/09515089.2023.2193216.
Garson, J. (2003). The introduction of information into Neurobiology. Philosophy of Science, 70(5), 926–936. https://doi.org/10.1086/377378.
Godfrey-Smith, P. (2009). Abstractions, idealizations and evolutionary biology. In A. Barberousse, M. Morange, & T. Pradue (Eds.), Mapping the future of biology: Evolving concepts and theories. Boston studies in the philosophy of science (pp. 47–55). Springer.
Haimovici, S. (2013). A problem for the mechanistic account of Computation. Journal of Cognitive Science, 14, 151–181.
Hutto, D., Myin, E., Peeters, A., & Zahnoun, F. (2019). The cognitive basis of computation: Putting Computation back in its place. In M. Colombo, & M. Sprevak (Eds.), The Routledge Handbook of the computational mind (pp. 265–281). Routledge.
Jones, M. (2005). Idealization and abstraction: A framework. In M. Jones, & N. Cartwright (Eds.), Idealization XII: Correcting the model, idealization and abstraction in the sciences (pp. 173–217). Rodopi.
Kersten, L. (2017). A mechanistic account of wide computationalism. Review of Philosophy and Psychology, 8(3), 501–517. https://doi.org/10.1007/s13164-016-0322-3.
Kersten, L. (2020). How to be concrete: Mechanistic computation and the abstraction problem. Philosophical Explorations, 23(3), 251–266. https://doi.org/10.1080/13869795.2020.1799664.
Kuokkanen, J. (2022a). Vertical-horizontal distinction in resolving the abstraction, generality and hierarchy problems. Synthese, 200, 247. https://doi.org/10.1007/s11229-022-03725-8.
Kuokkanen, J. (2022b). No computation without implementation? A potential problem for the single hierarchy view of physical computation. Synthese, 200(370). https://doi.org/10.1007/s11229-022-03696-w.
Kuokkanen, J., & Rusanen, A. (2018). Making too many enemies: Hutto and Myin’s attack on Computationalism. Philosophical Explorations, 21(2), 282–294. https://doi.org/10.1080/13869795.2018.1477980.
Lee, J. (2021). Mechanisms, wide functions, and content: Towards a Computationalism pluralism. British Journal for the Philosophy of Science, 72(1), 221–244.
Leigh, J., & Zee, D. (2006). The neurology of Eye Movements. Oxford University Press.
Maley, C. J. (forthcoming). Analogue computation and representation. The British Journal for the Philosophy of Science. https://doi.org/10.1086/715031.
Miłkowski, M. (2013). Explaining the computational mind. MIT Press.
Miłkowski, M. (2015). Computational mechanism and models of Cognition. Philosophia Scientiae, 18(3), 1–14.
Mäki, U. (1992). On the method of isolation in Economics. Poznan Studies in the Philosophy of the Sciences and the Humanities, 26(4), 317–351.
Morrison, M. (2015). Reconstructing reality. Oxford University Press.
Piccinini, G. (2007). Computing mechanisms. Philosophy of Science, 74, 501–526. https://doi.org/10.1086/522851.
Piccinini, G. (2015). Physical computation: A mechanistic account. Oxford University Press.
Piccinini, G. (2018). Computational mechanisms. In S. Glennan, & P. Illari (Eds.), The Routledge Handbook of mechanisms and Mechanical Philosophy (pp. 435–446). Routledge Taylor & Francis Group.
Piccinini, G. (2020). Neurocognitive mechanisms: Explaining Biological Cognition. Oxford University Press. https://doi.org/10.1093/oso/9780198866282.001.0001.
Piccinini, G., & Bahar, S. (2013). Neural computation and the computational theory of Cognition. Cognitive Science, 37(3), 453–488. https://doi.org/10.1111/cogs.2013.37.issue3.
Piccinini, G., & Craver, C. (2011). Integrating psychology and neuroscience: Functional analyses as mechanism sketches. Synthese, 183, 283–311.
Portides, D. (2018). Idealization and abstraction in scientific modeling. Synthese. https://doi.org/10.1007/s11229-018-01919-7.
Putnam, H. (1975). The mental life of some machines. Mind, language and reality, philosophical papers, volume 2, (pp. 408– 28). Cambridge University Press: Cambridge.
Schweizer, P. (2019). Computation in Physical systems: A normative mapping account. In D. Berkich, & M. Vincenzo d’Alfonso (Eds.), On the cognitive, ethical, and scientific dimensions of Artificial Intelligence (pp. 27–47). Philosophical Studies Series, Springer.
Shagrir, O. (2006). Why we view the brain as a computer. Synthese, 153, 393–416. https://doi.org/10.1007/s11229-006-9099-8.
Shagrir, O. (2020). In defense of the semantic view of computation. Synthese, 197, 4083–4108.
Shagrir, O. (2022). The nature of physical computation. Oxford University Press.
Sprevak, M. (2010). Computation, Individuation and the received view on representation. Studies in the History of Philosophy of Science Part A, 41, 260–270.
Sprevak, M. (2012). Three challenges to Chalmers on computational implementation. Journal of Cognitive Science, 13, 107–143.
Sprevak, M. (2018). Triviality Arguments About Computational Implementation. In The Routledge Handbook of the Computational Mind, edited by M. Sprevak, and M. Colombo, 175– 191. London: Routledge.
Wajnerman Paz, A. (2017). Pluralistic mechanism. Theoria: Revista De Teoría. Historia y Fundamentos de la Ciencia, 32(2), 161–175.
Weisberg, M. (2007). Three kinds of idealization. Journal of Philosophy, 104(12), 639–659.
Acknowledgements
I want to thank two anonymous reviewers from the journal for their helpful feedback and comments on earlier drafts of the paper. This research was generously supported by the Killam Trusts.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Competing Interests
The author declares no competing interests.
Additional information
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Kersten, L. An idealised account of mechanistic computation. Synthese 203, 99 (2024). https://doi.org/10.1007/s11229-024-04526-x
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s11229-024-04526-x