Concepts and predication from perception to cognition

Abstract

One popular doctrine in 20th-century philosophy was conceptualism about perception. The core idea was that perceptual awareness is structured by concepts possessed by the perceiver. A primary motivation for conceptualism was epistemological: perception provides justification for belief, and this justificatory relation is only intelligible if perception, like belief, is conceptually structured (Brewer, 1999; McDowell, 1994; Sellars, 1956). We perceive that a is F, and thereby grasp perceptual evidence that justifies the belief that a is F and inferentially integrates with premises like If a is F then a is G to produce the belief that a is G. Conceptualism is less popular today (cf. Bengson, Grube, & Korman, 2011; Mandelbaum, 2018; Mandik, 2012). The a priori justification for conceptualism has crashed face-first into a wall of empirical evidence. For instance, children and non-human animals possess perceptual capacities despite lacking many hallmarks of conceptual cognition (Bermudez, 1998; Burge, 2010a; Block ms). Meanwhile, in adults, mental imagery and related phenomena implicate iconic rather than conceptual/propositional formats (Carey, 2009; Fodor, 2007; Quilty-Dunn, 2019a). A growing contingent of theorists thus regard perception as a natural kind marked by its proprietary nonconceptual representations (Burge, 2014; Burnston, 2017a; Carey, 2009; Kulvicki, 2015a; Toribio, 2011; Block, ms; see also Evans (1982); Hopp (2011); Peacocke (2001) for other nonconceptualist arguments). Though opinion has shifted strongly in favor of nonconceptualism, it may be time for the pendulum to swing back. Putting the traditional normative motivations for conceptualism aside, it makes sense even from a purely descriptive, naturalistic perspective that at least some of the vehicles of perception should be conceptual. Many cognitive operations make use of concepts; thus many cognitive responses to perception would be facilitated if some outputs of perception came prepackaged in a conceptualized format. This point fits with modularity-based accounts of perception, and was fittingly made by Fodor in his discussion of input modules as “subsidiary systems” that must “provide the central machine with information about the world; information expressed by mental symbols in whatever format cognitive processes demand of the representations that they apply to” (Fodor, 1983, p. 40). Similarly, Mandelbaum argues that the outputs of modular perceptual systems ought to be conceptualized in order to “actually guide action by entering into other cognitive processes” (2018, p. 271). It is an underemphasized explanatory virtue of modularity that it allows for a system to be distinctly perceptual (in virtue of its modularity) while outputting representations that are immediately consumable by cognition (in virtue of their format). Modularity-based versions of conceptualism thereby avoid full-fledged versions of the “interface problem” in interactions between perception, cognition, and action (Burnston, 2017b; Butterfill & Sinagaglia, 2014; Mylopoulos & Pacherie, 2017; Shepherd, 2018; 2019). It is fully compatible with this modularity-based conceptualism that some perceptual processes output representations in nonconceptual (e.g., iconic) formats. Instead of insisting on conceptual structure as a transcendental epistemological requirement, modularity-based conceptualists can be pluralists about perceptual representation (Quilty-Dunn, 2019b). As long as some significant component of perception is conceptual and feeds immediately into cognition, there is room for other perceptual representations to have other formats with other functional advantages. For example, perhaps iconic representations allow for richer, messier content to be encoded in perception, while sparse conceptual representations provide neatly packaged categorizations to central cognition. However, perception is older than cognition. One might object that our perceptual systems evolved from creatures who lacked cognition, therefore there was no evolutionary pressure for concepts to figure in perception. In what follows, I'll sketch a version of conceptualism that posits concepts in perception independently of stimulus-independent cognitive abilities. In particular, I'll argue not only that adult humans have conceptually structured perceptual representations, but also that these conceptual outputs of perception constitute a natural representational kind found in children and animals alike. Perceptual object representations function to segment out particulars, track them, and predicate features of them, including conceptual categories. These object representations constitute an evolutionarily ancient and developmentally early source of predicate-argument propositional structure that is useful for (1) tracking individuals, (2) subsuming them under categories, and (3) distinguishing reference-guiding elements from pure attributions. These structures can function as evidential inputs to inferential processes in creatures that have the requisite inferential abilities. I will first argue against stimulus-independence as a constitutive condition on conceptuality (Prinz, 2002, p. 197; Beck, 2018; Burge, 2010b; Camp, 2009) in favor of a Cartesian view that concepts are simply representations of a certain sort that, in principle, require no particular mental abilities for their instantiation in human and animal minds (Fodor, 2004). I'll then use empirical evidence to argue that, in fact, perceptual object representations are conceptualized propositional structures that develop (and likely evolved) prior to creatures’ abilities to use them in inference. The resulting picture preserves much of the letter—if not exactly the spirit—of traditional conceptualism. It is not entirely clear how we ought to understand stimulus-dependence. Lots of mental activity might happen to be prompted by a pattern of stimulation and happen to end when stimulation ends, but this wide net might capture a messy variety of mentation rather than a natural kind. One could reasonably demand well-defined, testable characterizations of stimulation and dependence thereupon, and difficulties will surely arise in trying to provide them (see Beck (2018) for careful discussion of the details). I'll discuss two forms of stimulus-independent use: recombinability and logical inference. However, I propose to grant in general that there is some notion of stimulus (in)dependence that's coherent enough to figure in a candidate condition on concept possession. What matters for present purposes is the following claim: concepts are the sorts of mental phenomena that can only occur in creatures that have the ability to deploy them independently of what their transducers are doing at the moment. I'll also put aside a particularly strong form of the claim at issue. Beck (2018) argues that perception and cognition are distinguished by means of stimulus-independence: a state/process is perceptual iff it is stimulus-dependent, and cognitive iff stimulus-independent. This formulation runs into a counterexample: perception-based demonstrative thought, which is stimulus-dependent but cognitive (Beck, 2018, pp. 328–329). Beck's way of responding to this counterexample is to add the condition that every of a representation must be stimulus-dependent for the representation (and the that to be perceptual (2018, p. demonstrative have concepts as the concept is the in the that is concepts are demonstrative this However, this simply out the of concepts in perceptual systems by It like a empirical humans can deploy concepts & Mandelbaum, in the of this to avoid One for Beck's view would be to and concepts are stimulus-independent but they can be stimulus-dependent perceptual this perception-based demonstrative a Beck's version of the a avoid demonstrative as a counterexample but perceptual of concepts by allow for the perceptual of concepts but to the is not the that perception should be in of but the that concepts the ability to use them in a stimulus-independent that compatible with stimulus-independence that perception is by as proprietary representational formats even compatible with this that concepts could be stimulus-dependent As long as a with the concept can deploy that concept independently of entirely that the might have some stimulus-dependent means of it as Though stimulus-independence allows for concept in perception, it significant on the of creatures that can deploy In particular, creatures must possess the ability for stimulus-independent A of a that has evolved perceptual systems and them to guide action but central cognition have creatures are at the of their p. lacking the cognitive that conceptual for Camp, the representations they deploy in perception and must be For a even be a candidate for it must be p. concepts are representational abilities that are in an stimulus-independent p. for Burge, a representation is conceptual iff it function in pure p. which the of a p. For example, in is the predicate the of a while purely that perceptual have structures (e.g., therefore for pure concepts the ability to use them in pure of concepts the ability to use concepts perceptual For Burge, a of use is logical at view to a version of A of the of stimulus-independence is that it what the of The and of human that concepts are the sorts of mental that can be into the of many (e.g., The of recombinability has in a can be with the that a is F, then must have the conceptual for the that a is for every of G of which has a p. The is as a constitutive condition on concept (Beck, Camp, 2009; in versions argues that stimulus-independence the sense in which concepts are not but in a way that p. A point for is that concepts are mental abilities p. Burge, p. A primary of of stimulus-independence is to develop a of concepts the core of cognitive that we concepts to p. is thus a version of what Fodor p. and doctrine of philosophy of p. to concept is some sort of of stimulus-independence to a form of concept on which the ability to form independently of stimulation is constitutive of concept to concept is to which to the concept is to be to about as p. is a about what it is to possess a but it fits with a of concepts concepts are mental representations rather than and the is that mental particular the of which allows to as (Fodor, p. is thus the natural of representational of (Fodor, to which mental are representations, and mental processes are operations It is to that mental representations are than cognitive abilities. as the ability to are in of mental as and mental processes are in as operations mental representations, as the operations structures that & & Mandelbaum, If this is then it also out that mental representations are as cognitive on of If our cognitive abilities arise out of processes representations, then representations must be independently of abilities. provides an mental representations are vehicles with representational is by a to capture a of cognitive p. by the ability to them into the of concepts & 2004). For cognitive abilities can be to be by and some can even be of concepts than other sorts of representations, they be constitutive of concepts their A cognitive abilities in virtue of the sort of for the sort of concepts; the concepts the abilities rather than to and abilities to cognitive might be evidence to of concepts to some the are not made by the the abilities. This of concept to be a of However, be to be in the of a One only the that concepts are symbols p. pp. If concepts are then an inferential other cognitive abilities are constitutive of concept possession. first a not an ability to the in particular and thus a concept not require the ability to use it in stimulus-independent concept grant that cognitive abilities constitute on In his formulation of the Evans in a that there must be about the of the to the p. following Peacocke pp. that other pp. may These are not to conceptual abilities p. However, to the cognitive use of concepts in not arise As Peacocke can arise the of p. no might not happen to be to the of some they might be of mental functional that are in representational as between (e.g., pp. could in be of mental that concepts from certain One way to make sense of this is to a with they possess the concept that is than the on every time they to that concept into a like is a and the structure fully This the ability to this but they possess the concept the not in the concept but in a might that and concepts but for the ability to deploy them in certain to other A stimulus-independence A might possess a concept that it for perceptual However, its in the of input It thus that a might have a concept and the ability to deploy it in to stimulation while also a mental that the concept independently of One way to see the of this is to in mental be to that into the following is if we to a into the of the and their they would be to the of the concept independently of stimulation and it with their other the representations that are to function as premises in but the to make However, it might be that they would the ability to make them we to the A might at the they is that stimulus-independent use in and require than the of a require the as if we are to grant on stimulus-independent then no clear we should that concept could when they arise from the rather than to only the ability to about as the present of that even view is Fodor surely that concepts are symbols and that their use in on features of mental In that about as that a be from and is a that can of are and that the information to be as the that the of an to a may therefore that some might be and not be for (e.g., have The concept would this it is that the could arise from of mental some evolved but time symbols their in the of that that the symbols in the minds of but has them as a of if concepts are then a concept not cognitive even the ability to use the concept in a concept is simply a certain in This it the idea that the is a system and that concepts are symbols and in that may many as However, it is the notion of cognitive A natural way of in the concepts (and in this thus is in of an ability that a can at a the a concept is possessed by that is into a about what the can at that In that sense of that no cognitive abilities are constitutive of concept possession. One might abilities to creatures who them in (e.g., of One might also abilities not to creatures but to concepts that recombinability to conceptual capacities p. p. these we might develop the following a concept that the a that has the ability to be in stimulus-independent and the including mental a could this However, a primary motivation for is that it concept to cognitive and thus with of conceptuality from a useful that the core of cognitive that we concepts to p. The view the between concept and the of cognitive abilities in a way that the motivation concept It also for present creatures that stimulus-independent cognitive abilities may possess concepts What other evidence could concept stimulus-independent For a what matters is what of is in the rather than the cognitive abilities possessed by the representations as they the sort of representational format. I that the representational format of perception provides an means of these about In particular, we can some perceptual representations have a predicate-argument structure that is for logical the representations might be conceptual even if the creatures that possess them conceptual cognitive abilities. For and Camp, the for stimulus-independence on a of can into as when and in In particular, concepts into propositional like this is a What the propositional structures is their predicate-argument A picture of a might an object and even it as a but it by means of a predicate-argument to as an of content of a and as an of format. One might argue that a picture in that its content a of some but predicate-argument structure is not in the structure of the its format). predicate-argument structure that sort of functional relation some sort of logical relation the of p. In a like is a relation of function a relation of the to the to the and the relation between them to the instantiation of by the This sort of structure is a of predicate-argument structure as predicate and as In a picture of a this structure is are not two for an and for the of a the of the picture that the also its In this iconic representations are & Quilty-Dunn, 2017; Quilty-Dunn, 2019a). not in sense of a of elements as 2018, p. have a of (e.g., and their at For example, a of a picture might as as features at the as and A is by means of of the that other information as including of the their and p. Kulvicki, p. ms). that perceptual object representations the representations we use to and track have a rather than format & Quilty-Dunn, 2017; Quilty-Dunn, 2019b). I'll these for and an that out and to track them even when information is & 2011; & The and of these tracking abilities posits for that are to information & which fits a than an iconic for the and of a by means of the of the can features but them independently of other in & 2011; & & 2011; & & & 2017; ms). The of features in that features are format. for evidence of in that they like independently of other features & 2014; & & & For example, the in a perceptual object an in the of a & and a of a an in the of a from that & The therefore in a way that is not to even (cf. Burge, These are and by the that are in rather than The of these is that have a representational format (Quilty-Dunn, 2019b). for present purposes is the particular kind of format are at least in virtue of symbols that function as Camp, 2007; Kulvicki, the format of predicate-argument structure as a even as some the content of is & 1999; Kulvicki, Camp, 2018; In the of are by and and The function of is to information about and features by them to the In that it that sort of functional relation some sort of logical relation the of p. The of a that are related to the that the object in a way that makes the iff the object the What about notion of As by Burge, pure to stimulus-independent abilities and is thus to apply to perception. However, perhaps pure can be For Burge, pure that an function as a predicate the of a F is a predicate and an that to guide of that in perception function and is to the perceptual function of p. that perception pure claim that perceptual guide is an empirical provide track the system must an object at a is a a object that is the & the information about an object to guide to We can therefore the claim that perceptual guide perceptual by the of in object has long to on information rather than other in & 2009; evidence that features like can figure in object which is with idea that a reference-guiding & For example, in features like can object a is for on & found that in object when are to are but the to when the to the One might object that perhaps encoded in the but found that that In other when like are encoded with the system them to guide when the object is out and is the the but it no The system out that, even the object has this representational means it should not guide This a of the from the of the object to the predicate pure of the form object is at least some of the of like concepts that like as and and representations of like in that independently of and are to These features are of concepts like A is that the (e.g., simply with other features as related and In that like should like while are they not & the is to information that under the concept (e.g., the and of as would be if a of the These in favor of conceptual of hallmarks of conceptuality and a between and reference-guiding in other it that which guide on a particular is by which conceptual figure in is for object it a useful to object & and (e.g., and and object found a but only for have is of while can be object for but not is thus to as a not only to the particular object in the but also its for object categories. from a on which a of in a way that's between the object the and in general & and in particular & and as they the the found that, likely to see than for This provides evidence that concepts like can figure in and perceptual the is to rather than an between and for these that there was no between and that the for of that the is not to it that is by the conceptual of the stimulus and it that is useful in object thus not only figure in as they also the of in core perceptual and the of other like and as reference-guiding The version of conceptualism a concepts in a predicate-argument they must be to function in logical the cognitive if we stimulus-independence and concept conceptualized require abilities. conceptualism about makes some empirical we should be to with a predicate-argument format independently of but abilities are we should evidence that can function as these and non-human there is significant evidence from and abilities arise in the first of and that the are for in and (Carey, 2009; & & tracking has also in & 2011; & not much evidence traditional of in animals (cf. & it is to for evidence of their in cognition. are two we of in which the of a is and the which the of many and by rather than in with (Carey, 2009; & 2004). For instance, in can immediately see that A to how many but we can see that than A to its in and a of has long to on (Carey, 2009; Carey, & In a of this and a tracking while a of found that the that could be by for every the to track are by cognition to while the system a of format (Beck, not & these about in we can to the cognition to the that animals possess in humans are and like tracking and evidence of that in other animals provides but evidence for format in animal is present in many including & and & animals to abilities can For example, are to in of other and evidence they of but not that in to a they also have a system & & this ability may in to a some will the the of has while they are not to of they are to and cognition & have for and perceptual & that are and present in and However, the evidence for predicate-argument structure has thus from adult that a representational kind is for in and to this is while to evidence for of the representational kind a is to for & that are present evolutionary (Carey, p. The of use in tracking and thus the evidence should be as strong evidence for in and some evidence predicate-argument structure in are to track of the and of but to when a object from a to a & Carey, about 2004). this evidence may with from and could reasonably object that information is not to the representations at and to in which two with and at a the object an when the the object may have its found that at the when the object its a of they only when the object was the object For the the in and and this but a the object found that, while they in the This strongly that these have that track their and but that the information can be while the that an object representation that is & p. a in independently the tracking in that that have for and their as a of that form of evidence for conceptual conceptual of in a way that from in would provide evidence for predicate-argument structure in while also the of adult humans and creatures that and cognition. and the to of but of a and the a and a human If the object was with a object in the of to However, if the are conceptual This is the we would if we concepts like human could function as in in creatures to The evidence that with predicate-argument structure are present in and The that concept stimulus-independent as logical representations with predicate-argument structure are to figure as premises in logical in creatures of that have predicate-argument an is they independently of abilities and figure in when abilities it is to for (Burge, p. can for other inferential like in particular this pattern as a of inference. a a to a the first and no and immediately runs the One might that the 2004). However, other could this predicate-argument structure (Burge, 2010b; For example, a could use a mental to two and then A only could we creatures are this logical and a the two the two and two of an in of of this a of could form representations and A with the could only of which was and then that A was this the two The which inferentially integrates with to The simply if children logical they see A is they should immediately they should be as likely to of the other found that children the about which they for than to but have long this provides evidence that abilities. However, and that the between and in this with a in capacities p. One might that abilities in older children and on representations, not In a and found that a at the the abilities to not to on some allows to to pp. 2019). to on representations in which are & 2011; Quilty-Dunn, 2019b). Though that representations are propositional p. their on the that are If we that then the following a are representations with predicate-argument conceptualized propositional arise independently of abilities abilities can figure as premises in logical This is and with the evidence and a view of This has a sketch of form conceptualism might that is in and this our capacities to track and conceptualized propositional These structures to constitute a form of perceptual evidence on the of which we about the is not on stimulus-independent cognitive but rather provides a representational out of which abilities

Topics & Concepts