327 OPC on “Précis of The Philosophy of Affordances” Reclaiming Meaning, Reclaiming Normativity Laura Mojica Okinawa Institute of Science and Technology, Japan, and Universidad Autónoma Metropolitana, Mexico • laura.mojica460/at/gmail.com Abstract: I dispute Heras-Escribano’s reasons for denying that affordances are normative by offering an alternative reading of Wittgenstein’s considerations in which there is room for nonsocial but public normativity, and by defending that organisms’ affordance perception and engagement cannot be completely described in causal non-normative terms. Anthony Chemero (2009), Erik Rietveld (2008) and Rietveld and Julian Kiverstein (2014) have defended the view that affordances are normative because the abilities to perceive and engage with aspects of the environment as opportunities for action have correctness conditions. For example, one can be right or wrong in perceiving that one could pass through an aperture, or one can fail or succeed at climbing a mountain. Thus, affordances seem to imply that the organism establishes a way of relating with an aspect of the environment that can be right or wrong. Defending the normative character of affordances requires explaining how correctness conditions are established. There are at least two ways in which this explanation has been attempted: Proponents of autopoietic enactivism such as Ezequiel Di Paolo, Thomas Buhrmann and Xabier Barandiaran (2017) argue that affordances are normative because agents establish and follow their own intrinsic rule when acting, i.e., to preserve their form of life; Anthony Chemero’s (2009) preferred position is that affordances are normative because they fulfill a certain function, that is, they allow the agent to act. Note that the notion of normativity is oriented towards determining the right course of action rather than towards preserving truth or representing the environment accurately. Therefore, claiming that affordances are normative in this context roughly means that they determine right and wrong courses of action and are subjected to correctness assessments. In his 2019 book upon which this target article is based, Manuel Heras-Escribano (2019) gives various arguments against these two approaches to affordance normativity. In this commentary I will discuss two. The first one appeals to the external and inherently social character of normativity that he derives from Ludwig Wittgenstein’s rule-following considerations and private language argument, and the second one is based on the distinction between the nomological and the normative aspects in explaining behavior. His position has the virtue of making a clear distinction between biological error and cognitive error, and of highlighting the constitutive role of sociality in the mind. Despite these merits, I will dispute both of his arguments. Disentangling normativity Heras-Escribano’s first argument is based on the idea that normativity cannot be disentangled from sociality (§26). If this is the case, he argues, a phenomenon that can be explained in purely functional terms without appealing to a social dimension is not normative and, he claims, affordance perception and engagement are such phenomena. Moreover, if normativity is intrinsically social, the enactive account of agents as establishing and following their own intrinsic rules when perceiving and engaging with affordances are also invalidated. The reason why normativity is social is drawn from Wittgenstein’s rule-following considerations and private language argument. Wittgenstein’s rule-following considerations show that rule interpretation cannot determine whether a course of action is correct or not (Wittgenstein 2009: §201). The reason is that a rule can always be interpreted in such a way that the same action can be considered both wrong and right with respect to the same rule, and so it does not allow agents to distinguish whether they are right from the mere impression that they are. Thus, rule interpretation cannot be the way in which the correctness of an action or lack thereof is determined, which means that appealing to a rule cannot explain what makes an action meaningful or correct (ibid: §§185-201). What is required then? Normativity, as Heras-Escribano notes, must be open to the possibility of error, and it must provide external criteria to distinguish between acting right from the mere impression of acting right. Wittgenstein’s private language argument shows that whatever serves as a correctness criterion must be public and cannot be a reference to private phenomena (ibid: §270). This means that correctness criteria must be available to all the speakers and, in the particular case of language meaning, only social practices of language use are what is common to all – neither abstract rules nor inner thoughts and sensations (ibid: §23, §43, §260). Note that social practices are established as the sources of normativity for language because they are perfectly public and the only thing potentially available to all the speakers. This does not mean that sociality is the only possible source of normativity in general. Although normativity for both linguistic and nonlinguistic behavior is intrinsically public, it is not intrinsically social – why would any organism require social criteria to determine its right course of action? There are indeed public but not social ways of determining whether a course of action is correct or not: passing through an opening, hunting a prey, eating nonpoisonous berries, etc. have perfectly public correctness conditions. By their public character, I mean that being right or wrong, successful or not, and so on, have practical consequences that not only make a difference for the course of action of the agent, but that would be evident to an observer. Note that stipulating a potential observer for an action is too weak a criterion for calling it social, yet it is publicly available. This notion of publicness includes both socially constituted behavior, e.g., language use, and biological not socially constituted behavior that could go without being observed by others, e.g., an octopus hunting for food. The latter kind of behavior is nevertheless normative, not only because it has a purpose with respect to which the octopus can fail but also because its success or failure makes a clear practical difference for its survival. In that sense, perceiving and engaging with the environment for a practical purpose are perfectly public ways of relating between an agent’s bodily doings and aspects of the environment. In the unfolding of an action, agents engage with a particular aspect of the environment, i.e., a section of all the possibilities available to them, thereby making their purposes present in the environment. Their actions have practical consequences that reach beyond their intentions and make their success or failure evident to agents themselves even if they are never observed by others, e.g., having food to eat, being able to pass through an opening to find shelter. This means that affordance perception and engagement are perfectly public ways of relating that an agent establishes with the environment. Two possible objections could be raised here. First, it could be argued that affordance perception is substantially different from affordance engagement with respect to their public and normative characters, because while engaging with an affordance implies an overt exercise of skill, perceiving an affordance does not. However, drawing this distinction between action and perception so as to allow the former and not the latter to be normative goes against the core commitments of ecological psychology (Lobo, Heras-Escribano & Travieso 2018). Perception cannot be understood as separated from action because perception implies a sensorimotor loop in which what the agent does and feels are in a constant and inextricable feedback. Although this loop can be described causally, it still has normative conditions, i.e., that it should guide the agent in her action and flow of experience. Recall that the normative criteria at play here are not based on correspondence or truth preservation; they are instead pragmatic criteria the rightness, etc. of which are determined as actions unfold. As Claire Michaels and Claudia Carello (cited in Footnote 3) put it, perception should reveal useful aspects of the environment, but unlike them, I argue that usefulness implies a normative dimension. This brings us to the second objection: the correctness conditions at play in affordance perception and engagement are functional. As such, they can be completely accounted for in causal terms and so there is no need to appeal to normativity to characterize affordances. This is precisely Heras-Escribano’s second argument. Purposes and the nomological-normative distinction Unlike normative behavior, which has a social dimension and which cannot be reduced to fully causal terms, Heras-Escribano argues, we can make full sense of organisms’ behaviors in fully causal terms, i.e., by appealing to lawful regularities or to evolutionary functions (§23). Therefore, there is no need to introduce normative vocabulary to explain affordance engagement and perception (§29). As we saw above, the notion of normativity at play refers to action, more specifically, to what one ought to do but could err at doing. A source of normativity is what determines a course of action that ought to be followed and allows to differentiate it from other non-adequate behaviors; a behavior is normative if it is subjected to evaluation so it can be deemed right or wrong, successful or unsuccessful, etc. Note that, in the context of meaning, when the idea of normativity as truth or correspondence between a mental state and the environment is abandoned, the many distinctions under which a behavior can be evaluated point to the many dimensions of how one can make sense of an action. In line with Heras-Escribano, there are at least two requirements of normativity: its source should provide external criteria and a normative behavior should be open to the possibility of error (§31). However, unlike Heras-Escribano, I argue that organisms’ behaviors cannot be fully explained in causal terms. Whether a given behavior succeeds or fails is a normative assessment that depends on some purpose or value. However, purposes and values cannot be explained in purely causal terms: A description of bacteria swimming up towards an increasing sucrose gradient will be incomplete and meaningless if one fails to recognize that the sucrose is valuable as food for the bacteria. In general, any description of a living organism’s behavior is meaningless and incomplete if it fails to capture that there is purpose and value in the action of the organism, i.e., to preserve its form of life; otherwise it would not be motivated to select and engage with one of the multiple affordances available for it. In other words, a description in terms of causes and effects needs to be subsumed under a description of means and ends when the behavior of a living creature is at stake. The purposefulness of organisms serves as a clear criterion to evaluate their affordance perception and engagement. On the one hand, such an evaluation criterion warrants an external criterion: as I argued in the previous section, affordances are public relationships between agent and environment. The agent’s purposes are made present in the environment by the unfolding of her actions and by their practical consequences, thus, their success or failure is not only observable to others but also experienced by the agent despite its original intentions. On the other hand, organisms can fail in their affordance perception and engagement with respect to their own purposes, thus granting the possibility of error. Although Heras-Escribano is sympathetic to the recognition of biological and cognitive agency in nonhuman organisms (Footnote 17), he refrains from acknowledging their normative character because he wants to draw a clear distinction between the social component and the biophysiological component in meaningful behavior. However, denying normativity at the basic level of affordances comes with too high a cost: it obscures the continuity between human and other forms of life, thereby obscuring how nonsocial organisms could be taken seriously as agents with points of view in scientific explanations. Being a cognitive agent requires being able to interact with the environment according to the agent’s own purposes. If normativity is ruled out from the most basic form of cognition, i.e., affordance perception and engagement, then the relationship that nonsocial organisms establish with the environment cannot be explained in terms of their purposes. However, this implies that agency is ruled out for these forms of life. Thus, if Heras-Escribano wants to preserve nonhuman agency, he would either need to accept a basic nonsocial form of normativity, a proto-normativity, or else need to insist that a causal explanation is the full explanation of nonsocial living behavior and thereby commit to an eliminativist position. References Chemero A. (2009) Radical embodied cognitive science. MIT Press, Cambridge MA. Di Paolo E., Buhrmann T. & Barandiaran X. E. (2017) Sensorimotor life: An enactive proposal. Oxford University Press, Oxford. Heras-Escribano M. (2019) The philosophy of affordances. Palgrave Macmillan, Cham. Lobo L., Heras-Escribano M. & Travieso D. (2018) The history and philosophy of ecological psychology. Frontiers in Psychology 9: 2228. https://doi.org/10.3389/ fpsyg.2018.02228 Rietveld E. (2008) Situated normativity: The normative aspect of embodied cognition in unreflective action. Mind 117(468): 973–1001. https://cepa.info/4908 Wittgenstein L. (2009) Philosophical investigations. Blackwell, London. Originally published in 1953. The author Laura Mojica is currently a PhD student at the Universidad Autónoma Metropolitana, Mexico and a Special Research Student at the Okinawa Institute of Science and Technology, Japan. She has an MSc in logic from the University of Amsterdam, The Netherlands, and a BA in philosophy from the Universidad de los Andes, Colombia. During her MSc she worked on the philosophy of the later Wittgenstein. Now her research is focused on meaning normativity in 4E approaches to cognition and the limits of the operationalization of cognition. Received: 15 June 2020 Accepted: 17 June 2020