Skip to main content
Log in

How Knowing-That and Knowing-How Interface in Action: The Intelligence of Motor Representations

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

Abstract

What mental states are required for an agent to know-how to perform an action? This question fuels one of the hottest debates in the current literature on philosophy of action. Answering this question means facing what we call here The Challenge of Format Dualism, which consists in establishing which is the format of the mental representations involved in practical knowledge and, in case they are given in more than one format, explaining how these different formats can interlock. This challenge has generated two parallel debates: the debate between Intellectualism and Anti-Intellectualism on the one hand, and the debate on the Interface Problem on the other. While the former is about whether practical knowledge can be considered a species of propositional knowledge, the latter investigates how motoric and propositional states can be related. Here we offer a unified account capable of explicitly analyzing those two problems within the same philosophical framework. Our account suggests a new way for solving the Interface Problem that paves the way for addressing the debate between Intellectualism and Anti-Intellectualism.

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.

Institutional subscriptions

Fig. 1

Similar content being viewed by others

Notes

  1. We thank an anonymous Reviewer for focusing our attention to the importance of this point.

  2. Ryle’s crucial objection to the intellectualistic view is that it suffers from a regress argument. According to Ryle: “The consideration of propositions is itself an operation the execution of which can be more or less intelligent, less or more stupid. But if, for any operation to be intelligently executed, a prior theoretical operation had first to be performed and performed intelligently, it would be a logical impossibility for anyone ever to break into the circle” (Ryle,1949, p. 30). In other words, if the intellectualistic explanation of the fact that ‘A knows-how to perform Φ’ involves engaging S with the propositional state ‘that P’, we need to postulate the further action Φ 1 consisting of engaging S with P. Then, since the fact that A also knows how to perform Φ 1 requires an explanation, we need to postulate a further action Φ 2 consisting of engaging S with the propositional state that P1, and this inevitably leads to infinite regress. For different attempts of this sort, see Noë, 2004, 2005; Young, 2009, 2019; Jung & Newen, 2010, 2011; Wallis, 2008; Levy, 2015; Carter and Pritchard, 2013; Carter & Czarnecki, 2016; Fridland, 2012a, b, Ferretti, 2020

  3. For different middle-positions between these two, see (Bengson et al., 2009). For different aspects of the debate, see (Tsai, 2011; Poston, 2009; Stanley, 2011a; Williams, 2008; Schroeder, 2012; Nanay, 2009; Hutto, 2005; O’Regan & Noë, 2001; Noë, 2004, 2005; Jung & Newen, 2010, 2011; Fantl, 2016; Ferretti, 2020).

  4. We thank an anonymous Reviewer for suggesting this distinction.

  5. A further option would be that both intentions and MRs are non-propositionally structured representations (Campbell, 2018). This option is compatible with nAIT, since it makes practical knowledge entirely based on non-propositional representations. However, for this reason, it is not compatible with any versions of IT. Though it is a viable option, the present paper does not consider it.

  6. Although this argument has cogency, it does not exclude that knowing-how to perform Φ in a certain way w involves representational formats that are not present in the knowledge-that w is a way to Φ (Dickie, 2012; Ferretti, 2020; Jung & Newen, 2010; Noë, 2005; Rosefeldt, 2004). Indeed, the overt linguistic constructions by means of which people talk about skills ascriptions may not correspond to the way action skills are represented in the cognitive systems of the agents having the skills. Though common speakers are disposed to assign practical skills to an agent, imbuing her with PMCs of the form described above, the actual execution of a skilled action might not require the exclusive role of propositional states. However, supporters of nIT might address this argument by contending that, although the way we talk about knowing-how does not necessarily mirror the internal information underling action execution, the attribution of PMCs to an agent is still the best way to account for the rationality of action execution. Indeed, not only is the propositional format of a representation all it takes to give an account of our action, but there are no alternatives that can account for the rationality of action planning and execution better than the propositional format of representation. It should be noted that, although this view is not the only way to conceive intellectualism, it is nevertheless compatible with the definition of intellectualism as provided in the debate, and it has also been endorsed by some of the leading supporters of this view (Bengson et al., 2009; Stanley & Krakauer, 2013; Stanley, 2011b).

  7. Stanley and Williamson (2001) use the same example to argue that Ryle’s anti-intellectualistic view is ‘demonstrably false’, since the case of the trainer shows that knowing-how does not necessarily involve having any ability. However, this example can be used to show that in order to be able to execute an action, having propositional information is not enough.

  8. A way to understand the ‘practical mode of presentation’ would be that of conceiving it as a way of representing a practical skill by means of a non-propositional format, such as a motor representation. This allows one to save the intenSionality of knowing-how (Pavese, 2019), without facing the issues affecting nIT. Our view (§§ 3, 4, 5) can be read in this sense as well.

  9. It should be noted that these arguments do not exhaust all the issues that have been raised against the hypothesis that only propositional representations ground practical knowledge. For example, Bermudez (2007) & Pacherie (2011) have addressed what can be called the ‘argument of priority’, according to which knowing-that w is a way to perform Φ is dependent on the way Φ is performed, and not the other way around. Furthermore, Dickie (2012) has argued for the ‘manifestation without guidance argument’, according to which the manifestation of a practical ability does not necessarily reveal that such an ability is guided by propositional knowledge (for a recent review and support of this powerful argument, see Ferretti, 2020). We decided to treat here only some of these relevant arguments for reasons of space.

  10. As we shall see more in details in (§4.1), the main difference between an executable and a non-executable action concepts is that the former involves the suitable representation of how to execute the related action, while the latter does not (Pacherie, 2011, Ferretti & Zipoli Caiani, 2018).

  11. It should be noted that the fact that executable action concepts, qua prescriptive representations, are not suitable bearers of truth values does not depend on their direction of fit, but on their function. Indeed, executable action concepts prescribe how to perform an action by denoting the relevant motor acts, rather than describing such motor acts. The point is that prescriptions are not the type of representations suitable to be true or false (for a recent and specific discussion of this point see Ferretti, 2020; for the general discussion on these features of propositions see, McGrath and Devin, 2018).

  12. There is evidence of the fact that a long training in executing a skilled action changes the cortical organization of the motor cortex. For example, Monda et al. (2017) have recently shown significant differences in the motor cortex excitability between trained athletes and non-athletes, supporting the hypothesis that training determines a specific organization of the motor cortex. Further evidence of the existence of differences in the functional organization of the motor cortex between trained and non-trained agents can be revealed by experiments on ‘motor imagery’ (e.g., Lacourse, Orr, Cramer, & Cohen, 2005; Wei & Luo, 2010). There is, indeed, a special link between motor imagery and MRs (see Jeannerod, 2006; Ferretti, 2019).

  13. This confirms that, even if the subject using the appropriate MR might manifest propositional knowledge in Dickie’s sense (2012), her mental states related to appropriate motor processing are not guided by propositional knowledge alone (Dickie, 2012, pp. 737–738; see Ferretti, 2020)

  14. The reader should note that evidence shows that the activity of the motor system is functionally involved in the processing of executable action concepts, but it does not show that the functioning of the motor system is necessary for the processing of (non-executable) action concepts in general: the evidence is silent about whether an agent could normally use and understand an action concept even if she is not able to execute the related action. Accordingly, for the sake of our argument, it is not relevant whether an agent may have a (non-motoric) grasp of some non-executable action concept. What is relevant for our claim is that lesions of the motor system have an impact on the use and understanding of an executable action concept. Indeed, only executable action concepts are relevant in the processing of the intentions leading to action execution. Thus, what is important here is that lesions at the motor cortex that impair the execution of an action impact also on the processing of the related executable action concept.

  15. The Same Format Thesis is partially in line with two recent accounts aiming to solve the interface problem (Burnston, 2017; Shepherd, 2017). For more details about this relation, see (Ferretti & Zipoli Caiani, 2018, 2021; Ferretti, 2020; §6).

  16. For a discussion of this point, in line with the idea that knowing-how cannot be seen as a species of knowing that, as “sensorimotor representations underlying practical capacities and (…) propositional representations mark different representational formats”, see (Jung & Newen, 2010, p. 128). For a recent and meticulous discussion, framed within the debate on intellectualism, of how action concepts are non-propositional building blocks of a proposition, see (Ferretti, 2020).

  17. An important specification is the following. Our argument suggests also that, while MRs are biased by semantic competence, an agent does not need any propositional knowledge in order to generate the appropriate MR. However, this is not in conflict with the idea that, in skilled Φ-ing, the agent also relies on mixed representational formats, as she usually has a propositional intention to Φ and she also propositionally knows that w is a way to Φ. We explained how these states can properly interlock

  18. This article was initially conceived, deeply discussed and finally written by means of a shared work of both authors, so much so that it is impossible to attribute a particular merit to one or the other author, given the deep interplay at the basis of such a shared effort. However, having to recognize responsibilities in case of administrative reasons, we can say that Silvano Zipoli Caiani wrote the first part (§§1–3), which was meticolously reviewed, amended and finally approved by Gabriele Ferretti, while Gabriele Ferretti is responsible for the second part (§§ 4–6), which was meticulously reviewed, amended and finally approved by Silvano Zipoli Caiani. We would like to warmly thank six anonymous reviewers and the editor for addressing important comments, both on the empirical and on the philosophical side, which led us to improve the first version of this article. We also thank those scholars who have always been very enthusiastic in discussing with us the topics of this paper during the last years. Gabriele Ferretti would like to thank the audience of the 16th Annual Conference of the British Philosophy of Sport Association, held at the Trinity College, Oxford, for offering very interesting comments on this topic. Gabriele Ferretti has also some specific acknowledgments to do. Since the finalization of the work for the publication of the present article has been done during the transition from one research position to another, Gabriele Ferretti would like to acknowledge support from both the corresponding fellowships. This work was supported by a NOMIS Fellowship, awarded by the Eikones—Center for the Theory and History of the Image at the University of Basel, Switzerland. This work was also supported by a Humboldt Fellowship, hosted by Professor Albert Newen at the Institute for Philosophy II, Ruhr-University Bochum, Germany.

References

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Gabriele Ferretti.

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

Ferretti, G., Zipoli Caiani, S. How Knowing-That and Knowing-How Interface in Action: The Intelligence of Motor Representations. Erkenn 88, 1103–1133 (2023). https://doi.org/10.1007/s10670-021-00395-9

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10670-021-00395-9

Navigation