Abstract
According to the semantic view of computation, computations cannot be individuated without invoking semantic properties. A traditional argument for the semantic view is what we shall refer to as the argument from the cognitive science practice. In its general form, this argument rests on the idea that, since cognitive scientists describe computations (in explanations and theories) in semantic terms, computations are individuated semantically. Although commonly invoked in the computational literature, the argument from the cognitive science practice has never been discussed in detail. In this paper, we shall provide a critical reconstruction of this argument and an extensive analysis of its prospects, taking into account some ways of defending it that have never been explored so far. We shall argue that explanatory considerations support at best a weak version of the argument from the cognitive science practice, according to which semantic properties concur with formal syntactic properties in individuating computations in cognitive science, but not a strong version, according to which computation individuation in cognitive science is semantic as opposed to formal syntactic.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Note that Rescorla’s two conditions for formal syntactic processes (non-semantic and not tied to a particular neural realization) can also be easily satisfied by non-classical and not language-like computational systems such as connectionist or Bayesian models. Therefore, the term “formal syntactic” in this broad sense is supposed to be applicable also to such systems (e.g., Rescorla, 2017b; see Sect. 4). As observed by an anonymous reviewer, however, one might object that this broad application is unjustified since there is no clear sense in which connectionist or Bayesian models have a syntax (although these systems can indeed be formal). Other terms (e.g., “abstract causal”, see Sect. 4) might be preferable in this context. Although we agree with this observation, we prefer to keep the “formal syntactic” terminology in order to avoid confusion and preserve consistency with previous literature (especially with Rescorla’s writings, which represent the main polemical target of the last sections of the present articles; see Sects. 4–5).
As noted by Coelho Mollo, who takes up a point made by Ramsey (2007), «talk of representation in the cognitive sciences, though widespread, may be largely empty, and we must examine in each case whether it is playing its proper explanatory role» (2020, p. 107). The defenders of the semantic view of computation might object that, without a strong reason to the contrary, we should not assume that scientists are confused or otherwise misguided. Rather, we should take the actual practice in cognitive science as a guide of explanatory success (see Rescorla, 2012). However, it is clear that premise (P) in its descriptive reading does not provide strong support for conclusion (C).
It is important to stress that these considerations are not meant to challenge the premise (P) in its normative force but rather the passage from (P) to the conclusion (C). As we have seen, Egan is explicit in arguing that semantic content is a critical and perhaps necessary component of computational explanations in psychology. She denies, however, the implication that goes from explanatory to metaphysical/individuative considerations. She claims, for instance: «[b]ecause the ascription of distal contents is necessary to explain how a computational process constitutes the exercise of a cognitive capacity in a particular context, I shall call the interpretation that enables the assignment of such distal contents the cognitive interpretation. The cognitive interpretation is to be sharply distinguished from the mathematical interpretation […]. Only the latter plays an individuative role» (2012, p. 266). Similarly, she claims that «[r]epresentational content does not play an indivituative role in cognitive theories; but it does play an important explanatory role» (2010, p. 255; see also Egan 1992, 1995, 2014). Of course, this does not mean that, according to Egan, only semantic properties have explanatory relevance in cognitive science. In her view, also formal syntactic properties have a critical explanatory role in addition to their individuative role (e.g.,1992, 1995, 2014).
Note that these modal considerations seem to hold also in the opposite direction. Consider again the Laplacian example. Assume that another visual system performs edge-detection not by computing the Laplacian (and zero-crossings), but by computing the extreme points of first derivatives (differentiation). In the two cases, the inputs and outputs would receive the same semantic interpretation, whilst the computation (as mathematically described) would be different. Again, this suggests that semantic content is not among the essential properties of computations (we thank an anonymous reviewer for raising this point).
The neat independence between explanatory and metaphysical/individuative considerations seems to be professed not only by defenders of the formal syntactic view of computation, such as Egan, but even by some defenders of the semantic view. For instance, Shagrir (2020) has explicitly argued that the semantic view of computation individuation, being a matter of essential properties, is independent of whether and how semantic properties are used in explanation and theories (e.g., Shagrir 2020, p. 4088).
One could argue that conclusion (C) in this form cannot be considered a proper formulation of the semantic view of computation. For it leaves open the possibility that computations in other domains (e.g., artificial computing) are non-semantically individuated. Shagrir (2020) calls “neither semantic nor non-semantic (NSNNS)” the view according to which «computational individuation takes into account semantic properties in some cases but not in others» (p. 4086). Some scholars might dispute this claim. Rescorla, for instance, although initially in line with this characterization (2013), has recently explicitly argued that «[c]ontent-involving computationalists need not say that all computational description is intentional […] They claim only that some important computational descriptions are content-involving» (2020). For the sake of argument, we are disposed to assume that it is possible to relativize considerations about computation individuation to specific classes of computational systems (e.g., brains) and to specific scientific fields (e.g., cognitive science). In the context of cognitive science, for instance, the semantic view is the thesis that cognitive computations are (or should be) individuated in contentful terms. As we will see, we are interested in ruling out an extreme version of this view, i.e., the idea that semantic as opposed to formal syntactic properties should be invoked by computation individuation in cognitive science.
For instance, as we have seen, Burge explicitly argues that no explanatory work is given to formal syntactic properties in actual cognitive science (2010; see Sect. 1). In a recent review article, Rescorla describes his own position in a similarly strong form: «Formal syntactic computationalism and content-involving computationalism are compatible […] However, many content-involving computationalists reject formal syntactic computationalism. Tyler Burge and Michael Rescorla question whether formal syntactic description adds any explanatory value to content-involving description. They question whether we can “hive off” a mental state’s representational properties to obtain a psychologically significant formal syntactic bearer of those properties» (2014b, p. 69). Similarly strong formulations can be found in several other Rescorla’s articles (e.g., 2015a, 2015b, 2016, 2017a, 2017b, 2020). Admittedly, Rescorla sometimes recognizes that there may be some areas in cognitive science in which the formal syntactic scheme is used (e.g., 2017b, p. 17; see Sect. 1). Nevertheless, by and large, «current cognitive science does not support any such formal syntactic taxonomic scheme […]. The proposed scheme plays no role within current scientific theories of perception, motor control, mammalian navigation, or numerous other core mental processes. Researchers describe these processes in representational terms» (2021, p. 172).
Note that, in such a view, «there is no rigid demarcation between computational and intentional description. In particular, certain scientifically valuable descriptions of mental activity are both computational and intentional» (Rescorla, 2020). Indeed, this does not mean that computationalism adds nothing to representationalism. Instead, it means that computational description (in explanations and theories) of cognitive phenomena are always semantically characterized; in cognitive science, there is no such a thing as semantically indeterminate, multiple realizable computational description (see later).
Granted, Rescorla does not argue that computational individuation in artificial computing is purely formal syntactic. According to him, semantic notions have an important explicatory and individuative role in many areas of artificial computing, such as computer science and probabilistic robotics (2017b, pp. 8–9). As we will see, Rescorla’s claim is that, differently from cognitive science, artificial computing does assign an important role to formal syntactic description alongside semantic description.
References
Carandini, M. (2012). From circuits to behavior: A bridge too far? Nature, 15(4), 507–509.
Carandini, M., & Heeger, D. J. (2012). Normalization as a canonical neural computation. Nature Reviews Neuroscience, 13, 51–62.
Chalmers, D. (2012). The varieties of computation: A reply. The Journal of Cognitive Science, 13, 213–248.
Chirimuuta, M. (2014). Minimal models and canonical neural computations: The distinctness of computational explanation in neuroscience. Synthese, 191, 127–153.
Chirimuuta, M. (2017). Explanation in neuroscience: causal and non-causal. British Journal of Philosophy of Science, 9, 849–880.
Chirimuuta, M. (2018). Vision. In Sprevak Colombo (eds.), The Routledge Handbook of The Computational Mind. London: Routledge.
Coelho Mollo, D. (2020). Content pragmatism defended. Topoi, 39, 103–113.
Colombo, M., & Hartmann, S. (2015). Bayesian cognitive science, unification, and explanation. The British Journal for the Philosophy of Science. https://doi.org/10.1093/bjps/axv036
Burge, T. (1986). Individualism and Psychology. The Philosophical Review, 95, 3–45.
Burge, T. (2010). Origins of Objectivity. Oxford: Oxford University Press.
Davies, M. (1991). Individualism and perceptual content. Mind, 100, 461–484.
Dennett, D. (1981). Three kinds of intentional psychology. In R. Healey (Ed.), Reduction, time and reality. Cambridge: Cambridge University Press.
Dennett, D. (1994). Cognitive science as reverse engineering: Several meanings of ‘top-down’ and ‘bottomup.’ In D. Prawitz, B. Skyrms, & D. Westerstahl (Eds.), Logic, methodology & philosophy of science IX. Amsterdam: Elsevier Science.
Dewhurst, J. (2018). Individuation without representation. The British Journal for the Philosophy of Science, 69, 103–116.
Dietrich, E. (1989). Semantics and the computational paradigm in cognitive psychology. Synthese, 79, 119–141.
Egan, F. (1991). Must psychology be individualistic? The Philosophical Review, 100, 179–203.
Egan, F. (1992). Individualism, computation, and perceptual content. Mind, 101, 443–459.
Egan, F. (1995). Computation and content. The Philosophical Review, 104, 181–203.
Egan, F. (1999). In defense of narrow mindedness. Mind and Language, 14, 177–194.
Egan, F. (2010). A modest role for content. Studies in History and Philosophy of Science, 41, 253–259.
Egan, F. (2012). Represenationalism. In E. Margolis, R. Samuels, & S. Stich (Eds.), The Oxford handbook of philosophy of cognitive science. Oxford: Oxford University Press.
Egan, F. (2014). How to Think About Mental Content. Philosophical Studies, 170, 115–135.
Egan, F. (2019). The nature and function of content in computational models. In M. Colombo & M. Sprevak (Eds.), The routledge handbook of the computational mind. London: Routledge.
Egan, F. (2020). A deflationary account of Mental Representations, in J. Smortchkova, K. Dołrega, T. Schlicht (2020). What are Mental Representations? Oxford: Oxford University Press.
Field, H. (2001). Truth and the Absence of Fact. Oxford: Oxford University Press.
Fine, K. (1994). Essence and modality: The second philosophical perspectives lecture. Philosophical Perspectives, 8, 1–16.
Fodor, J. (1975). The language of Thought. New York: Thomas J. Crowell.
Fodor, J. (1981). Methodological Solipsism Considered as a Research Strategy in Cognitive Psychology. In J. Fodor (Ed.), Representation. Cambridge: MIT Press.
Friston, K. (2009). The free-energy principle: A rough guide to the brain? Trends in Cognitive Sciences, 13, 293–301.
Friston, K. (2010). The free-energy principle: A unified brain theory? Nature Review Neuroscience, 11, 127–138.
Gallistel, C. R., & King, A. (2009). Memory and the Computational Brain. New York: Wiley-Blackwell, Malden.
Hohwy, J. (2013). The predictive mind. Oxford: Oxford University Press.
Haugeland, J. (1985). Artificial intelligence: The very idea. Cambridge: MIT Press.
Kaplan, D. (2018). Neural computation, multiple realizability, and the prospects for mechanistic explanation. In D. Kaplan (Ed.), Explanation and integration in mind and brain science. Oxford: Oxford University Press.
Keas, M. N. (2017). Systematizing the theoretical virtues. Synthese, 1–33.
Kitcher, P. (1989). Explanatory unification and the causal structure of the world. In P. Kitcher & W. Salmon (Eds.), Scientific explanation. Minneapolis: University of Minnesota Press.
Knill, D. C., & Pouget, A. (2004). The Bayesian brain: The role of uncertainty in neural coding and computation. Trends in Cognitive Sciences, 27(12), 712–719.
Kripke, S. (1980). Naming and Necessity. Cambridge: Harvard University Press.
Kolmogorov, A. (1950) [1933]. Foundations of the theory of probability. New York: Chelsea Publishing Company.
Lee, J. (2018). Mechanisms, Wide Functions, and Content: Towards a Computational Pluralism. The British Journal for the Philosophy of Science, 2018, axy061.
Lowe, E. J. (2003). Individuation. In M. J. Loux & D. W. Zimmerman (Eds.), The Oxford Handbook of Metaphysics. Oxford: Oxford University Press.
Ma, W. J., Beck, J. M., Latham, P. E., & Pouget, A. (2006). bayesian inference with probabilistic population codes. Nature Neuroscience, 9, 1432–1438.
Marr, D. (1982). Vision. San Francisco: W.H. Freeman.
Millikan, R. (1995). Pushmi-pullyu representations. Philosophical Perspectives, 9, 185–200.
Myrvold, W. C. (2003). A Bayesian account of the virtue of unification. Philosophy of Science, 70, 399–423.
O’Brien, G. (2012). Defending the semantic conception of computation in cognitive science. Journal of Cognitive Science, 12(4), 381–399.
Peacocke, C. (1994). Content, computation, and externalism. Mind and Language, 9, 303–335.
Peacocke, C. (1999). Computation as involving content: A response to Egan. Mind and Language, 14, 195–202.
Pemberton, J. (2018). Individuating processes. In O. Bueno, M. B. Fagan, & R. Chen (Eds.), Individuation across experimental and theoretical sciences. Oxford: Oxford University Press.
Piccinini, G. (2008). Computation without representation. Philosophical Studies, 137, 205–241.
Piccinini, G. (2015). Physical computation: A mechanistic account. Oxford: Oxford University Press.
Potochnik, A. (2010). Levels of explanation reconceived. Philosophy of Science, 77(1), 59–72.
Pincock, C. (2012). Mathematics and scientific representation. Oxford: Oxford University Press.
Putnam, H. (1967). Psychophysical Predicates. In W. Capitan & D. Merrill (Eds.), Art, Mind, and Religion. Pittsburgh: University of Pittsburgh Press.
Ramsey, W. (2007). Representation reconsidered. Cambridge: Cambridge University Press.
Ramsey, W. (2020). Defending representational realism. In J. Smortchkova, K. Dołęga, & T. Schlicht (Eds.), What are mental representations? Oxford: Oxford University Press.
Rescorla, M. (2012). How to integrate representation into computational modeling, and why we should. The Journal of Cognitive Science, 13, 1–38.
Rescorla, M. (2013a). Against Structuralist theories of computational implementation. The British Journal for the Philosophy of Science, 64, 681–707.
Rescorla, M. (2013b). Millikan on honeybee navigation and communication. In D. Ryder, J. Kingsbury, & K. Williford (Eds.), Millikan and her critics. Wiley-Blackwell: Malden.
Rescorla, M. (2014a). The causal relevance of content to computation. Philosophy and Phenomenological Research, 88, 173–208.
Rescorla, M. (2014b). Computational modeling of the mind: What role for mental representation? Wiley Interdisciplinary Reviews: Cognitive Science, 6, 65–73.
Rescorla, M. (2015a). Bayesian perceptual psychology. In M. Matthen (Ed.), The Oxford handbook of the philosophy of perception. Oxford: Oxford University Press.
Rescorla, M. (2015b). The representational foundations of computation. Philosophia Mathematica, 23, 338–366.
Rescorla, M. (2016). Bayesian sensorimotor psychology. Mind and Language, 31, 3–36.
Rescorla, M. (2017a). From Ockham to Turing—and Back Again. In A. Bokulich & J. Floyd (Eds.), In Turing 100: Philosophical explorations of the legacy of Alan Turing. Cham: Springer.
Rescorla,. (2017b). Levels of Computational Explanation. In T. Powers (Ed.), Philosophy and Computing: Essays in Epistemology, Philosophy of Mind, Logic, and Ethics. Berlin: Springer.
Rescorla, M. (2017c). Maps in the head? In K. Andrews & J. Beck (Eds.), The Routledge handbook of philosophy of animal minds. Basingstoke: Taylor & Francis.
Rescorla, M. (2020). The computational theory of mind, in E. Zalta (Ed.), (2015, Fall) The Stanford encyclopedia of philosophy. https://plato.stanford.edu/cgibin/encyclopedia/archinfo.cgi?entryDcomputational-mind
Rescorla, M. (2021). Reifying representations. In J. Smorthchkova, T. Schlicht, & K. Dolega (Eds.), What are mental representations? (pp. 135–177). Oxford: Oxford University Press.
Ritchie, B. (2019). The content of Marr’s information-processing framework. Philosophical Psychology, 32(7), 1078–1099.
Schurz, G. (2017). Patterns of Abductive Inference. In L. Magnani & T. Bertolotti (Eds.), Springer handbook of model-based science. Berlin: Springer.
Shagrir, O. (1999). What is computer science about? The Monist, 82(1), 131–149.
Shagrir, O. (2001). Content, Computation, and Externalism. Mind, 110, 369–400.
Shagrir, O. (2010). Marr on Computational-Level Theories. Philosophy of Science, 77(4), 477–500.
Shagrir, O. (2020). In defense of the semantic view of computation. Synthese, 197, 4083–4108.
Shapiro, L. (1993). Content, Kinds, and Individualism in Marr’s Theory of Vision. The Philosophical Review, 102(4), 489–513.
Shapiro, L. (1997). A Clearer Vision. Philosophy of Science, 64, 131–153.
Sprevak, M. (2010). Computation, Individuation, and the Received View on Representation. Studies in History and Philosophy of Science, 41, 260–270.
Stich, S. (1983). From folk psychology to cognitive science. Cambridge: MIT Press.
Zednik, C., & Jäkel, F. (2018). Bayesian reverse-engineering considered as a research strategy for cognitive science. Synthese, 193, 3951–3985.
Acknowledgements
We thank Vincenzo Crupi, Diego Marconi, Agostino Pinna Pintor, Jan Sprenger, and two anonymous reviewers of Synthese for very useful comments on this paper and/or discussions on this issue.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Calzavarini, F., Paternoster, A. The semantic view of computation and the argument from the cognitive science practice. Synthese 200, 77 (2022). https://doi.org/10.1007/s11229-022-03542-z
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s11229-022-03542-z