(My Sociology, 2021)
You can also download and read or share a .pdf of the complete text of this essay by scrolling down to the bottom of this post and clicking on the Download tab.
What Are Concepts?
1. Introduction
All discursive human thinking proceeds by means of concepts.[i] But what are concepts? They’re essentially general mental representations that can describe anything under the sun, provided that it’s not gibberish or nonsense, including actual or possible objects, other people, oneself, non-human animals and other organisms, super-human beings, processes, qualities, relations, properties of all sorts, and states-of-affairs of all sorts, including impossible ones—even including logical contradictions and paradoxes—and so-on and so forth, ad infinitum: roughly speaking, you name it, and then some concept can describe it. And you can also form concepts of things that don’t have names yet.
But so far, this is all very informal and rough; hence in what follows, I want to present the basic outlines of what I think is an intelligible and defensible theory of the nature of concepts.
2. A Working Definition of the Concept “Concepts”
Here’s a working definition of the concept “concepts”: concepts are the inherently descriptive, general, contextless, veridical or non-veridical meanings of either (i) various abstract nouns (including gerunds) in natural language (for example, “altruism,” “analysis,” “analyzing,” “bachelorhood,” “capital-B Beauty” “coercion,” “cognition,” “cognizing,” “communication,” “conceptualization,” “conceptualizing,” “dignity,” “ethics,” “etiquette,” “freedom,” “friendship,” “capital-G Goodness,” “hatred,” “honor,” “humility,” “hypocrisy,” “intending,” “intentionality,” “logic,” “logicism,” “love,” “loving,” “morality,” “moralism,” “mysticism,” “phenomenology,” “philosophizing,” “philosophy,” “psychology,” “psychologism,” “the social institution of marriage,” “the social institution of science,” “scientism,” “skepticism,” “slavery,” “capital-T Truth,” and so-on), or (ii) predicative or many-place relational terms in natural language (for example, “cat(s),” “mat(s),” “moon(s),” “x is sitting on y,” etc.), that also inherently belong to two different kinds of larger meaning-complexes: (iia) “propositions” or “thoughts,” whether simple or compound, built up out of concepts and various kinds of logical operators (for example, the “is” of predication, the “is” of identity, the “is” of assertion, “the/one and only one,” “all/every,“ “some/at least one,” the “not” of propositional/ thought negation, the “non-” of predicate negation, “and,” “either … or,” “if … then,” etc.), that either correspond to actual facts in the world and are true (for example, “Some cat is sitting on a mat”) or fail to do so and are false “Some cat is sitting on the moon”), and (iib) arguments or inferences, which are chains or sequences of thoughts/propositions governed by laws of logical validity and soundness (for example, “Every cat is are sitting on some mat or another. Therefore some cat is sitting on some mat”). This, I think, is a more-or-less classical definition of “concepts.”
Nevertheless, a striking and also philosophically troubling feature of recent and contemporary post-classical Analytic philosophy and so-called “Continental philosophy” alike, is the lack of any generally-accepted, prima facie plausible theory of the nature of concepts;[ii] indeed, some contemporary post-classical Analytic philosophers of mind are outright skeptical about the very idea of a concept, or else they offer debunking alternatives that are still called “concepts,” but only in a Pickwickian sense (see, e.g., Machery, 2009; and Queloz, 2021). In order to face up to that problem, as I mentioned, I’m going to present the basic outlines of what I think is an intelligible and defensible theory of the nature of concepts, and also assert that concepts in this sense really do exist.
3. On the Nature of Concepts[iii]
Mental content in general is the individuating, normatively guiding, cognitive or practical information about objects, locations, events, processes, actions or performances, other minded animals, or oneself, that is contained in a mental representation—aka an intentional act, state, or process—insofar as that mental representation is an intersubjectively shareable type that is also tokened in and directly cognitively accessible to individual minded animals on particular occasions and in particular contexts.
So, among the various different kinds of mental contents and mental representation types, what, more specifically, are concepts? In a nutshell, my answer is that a concept is an essentially descriptive, essentially general, categorizing mental content with inherent linguistic and logical form. This nutshell can then be opened up into four basic proposals about the nature of concepts.
According to my first basic proposal, necessarily, X is a concept—or what is the same thing, X is a conceptual content—if and only if X is a mental content such that (i) X is either a material concept or a formal concept (the conceptual dualism condition). In turn, (ii) X is a material concept if and only if (iia) X provides for the definite or indefinite categorization, classification, discrimination, identification, and cognitively significant presentation of some actual or possible individual things (in a maximally broad sense that includes events, processes, locations, regions, etc.) in the manifestly real natural world, or unordered or ordered n-tuples of individual things in the manifestly real natural world (which allows for monadic concepts like BACHELOR and also for relational concepts like TALLER THAN), and X is thereby inherently descriptive of those individual manifestly real natural things, which in turn “fall under” X (the first-order descriptivity condition), (iib) X is such that a conscious cognizer need not necessarily be directly acquainted with or confronted by whatever is represented by X right then and there in order to understand X, provided that those things, as represented by X, have already been encountered essentially non-conceptually in sense perception, and that the memory of that earlier essentially non-conceptual perceptual acquaintance is cognitively accessible (the non-acquaintance condition), (iic) X is such that within its complex descriptive intensional structure there is at least one concept Y (possibly identical to X), such that Y is basic and Y requires an essentially non-conceptual perceptual acquaintance with at least one of the things represented by X (the acquaintance condition), (iid) X fullysupports the truth of some analytic propositions that are necessarily true in virtue of intensional containment (the containment analyticity condition), and finally (iie) the self-conscious cognition of X fully supports some sufficiently justified analytically necessarily true beliefs, i.e., a priori analytic knowledge (the analytic a priori knowledge condition). On the other hand, (iii) X is a formal concept if and only if (iiia) X provides for the definite or indefinite categorization, classification, discrimination, identification, and cognitively significant presentation of some material concepts, and X is thereby inherently descriptiveof those material concepts, which in turn are inherently descriptive of the individual manifestly real natural things that fall under them (the higher-order descriptivity condition), (iiib) X is such thata conscious cognizer need not necessarily bedirectly perceptually acquainted with or confronted by the individual manifestly real natural things, or unordered or ordered n-tuples of individual manifest natural things, that fall under any of the material concepts to which X applies (the higher-order non-acquaintance condition), (iiic) X partially or wholly provides for the logical consequence relation, logical constants, logical laws and/or logical inference rules of classical truth-functional logic, or classical first-order predicate logic plus identity (aka “elementary logic”), or some conservative or deviant extension of elementary logic (the logical notions condition), (iiid) X fully supports the truth of analytic propositions that are necessarily true in virtue of logic, i.e., logical truths (the logical truth condition), and finally (iiie) the self-conscious cognition of X supports some sufficiently justified analytically necessarily true logical beliefs, i.e., a priori logical knowledge (the logical a priori knowledge condition).
According to my second basic proposal about the nature of concepts, necessarily, all concepts, whether material or formal, are inherently related to natural language. More precisely, necessarily, if X is a concept, then, (iv) X is intersubjectively cognitively shareable and communicable by means of some or another natural language L, precisely because X is a linguistically- and logically-structured mental representation type that can be variously tokened in the minds of competent, rational speakers of L when they correctly use abstract- noun expressions like “bachelorhood” or “the social institution of marriage,”or n-place predicative expressions like “__ is a bachelor” and “__ is married to __,” sentential modifiers like negation, and sentential connectives like conjunction) of L that have X as their linguistic meaning, by virtue of the innate a priori cognitive capacities that all competent, rational speakers of L possess for generating linguistic and logical understanding (the linguistic cognitivism condition) (see Hanna, 2006: chs. 4-6).
Furthermore, according to my third basic proposal about the nature of concepts, necessarily, concepts are possessible. More precisely, necessarily, if X is a concept, then (v) X is possessible, which entails that (va) X is deployable and usable, which is to say that X makes it really possible for human cognitive subjects to recognize X-type things when they perceive them, and also to distinguish X-type things from other types of things, (vb) it’s really possible for rational human cognitive subjects to be self-consciously aware of at least some of the intrinsic descriptive intensional elements of X, and (vc) it’s really possible for rational human cognitive subjects to make analytically necessary and a priori logical inferences that pick out at least some of the intrinsic descriptive intensional elements of X, but also (vd) it’s really possible for (va) to be satisfied by some human but non-rational cognitive subjects (for example, normal human toddlers and other young children) without their also satisfying either (vb) or (vc), and it’s really possible for (vb) and (vc) to be satisfied by other human but rational cognitive subjects, without their also satisfying (va) (for example, someone who knows that bachelors are unmarried adult males, but can’t actually recognize any bachelors on the ground or distinguish them from non-bachelors) and in all such cases there is no real possibility of concept-possession, and thus no conceptual contents in the strict sense, although some inherently concept-like contents, aka proto-concepts, are still present in the mental acts or states of those cognitive subjects (the concept-possession conditions).
Finally, according to my fourth basic proposal about the nature of concepts, necessarily, concepts relate to the manifestly real world via rational human cognition. More precisely, (vi) necessarily, if X is a material concept, then some actual or possible rational human cognizer (via) actually or really possibly uses X to detect some essential or accidental in rebus manifestly real properties or relations of actual individual manifestly real natural objects, which are also their mereological structures (Koslicki, 2008) (the world-detection condition), and also (vib) accurately mirrors and records this information in the descriptive intensional microstructure of the content of X when the rational animal cognizer cognitively generates it (the world-mirroring condition), nevertheless (vic) this is not to say that no concepts pick out either ante rem properties/relations or uninstantiated manifestly real properties/relations. Indeed and precisely on the contrary, all the formal concepts pick out ante rem properties or relations; and every consistent set of material concepts picks out a manifestly real property or relation, whether or not it is actually instantiated. The fact remains, however, that every material concept picks out at least one in rebus manifestly real property or relation. So all conceptual content is firmly anchored in the actual manifestly real natural world (the world-anchoring condition).
Just to give this four-proposal, six-part theory of concepts a convenient label, and also because it directly reflects the theory of “Logical Cognitivism” that I defended in Rationality and Logic (Hanna, 2006), I’ll call it The Logical Cognitivist Theory of Concepts, aka LCTC.
There are four things that follow directly from LCTC, and are most certainly worth noticing right from the get-go.
First, it follows specifically from conditions (iib) and (iic) on being a material concept—the non-acquaintance condition and the acquaintance condition—that all material concepts are complex descriptive intensional items that are ultimately grounded, via their basic conceptual parts, by essentially non-conceptual perceptual content (see, e.g., Hanna, 2005, 2008, 2015: ch. 2, 2020, 2021, 2024) and are cognitively constructed on that basis alone. All material concepts thus metaphysically and cognitively bottom out in essentially non-conceptual perceptual content, but without their being in any way either reducible to these essentially non-conceptual contents or atomistically composed out of them.
Second, it follows specifically from the five conditions on being a material concept—i.e., conditions (iia) through (iie)—together with the world-detection condition, the world-mirroring condition, and the world-anchoring condition, that one of the specific semantic roles of an analytic truth is to express essential connections between various manifestly real properties/relations and structural proper parts of macroscopic material things, via essential connections between the corresponding material concepts/ conceptual networks and structural proper parts of material concepts that intensionally mirror and record precisely those manifestly real essential connections in the world.
Now, this general point about material concepts and manifestly real properties/relations needs to be specially re-emphasized—that there is a one-to-one correlation between material concepts and manifestly real properties/relations. The only basic differences between material concepts and manifestly real properties/relations are (a) that all material concepts, as mental representation types, are either tokened in some actual rational human mind or else tokenable in some really possible rational human mind, but when a material concept is tokened in an actual rational human mind, it’s not necessarily the case—indeed, it’s normally not the case—that the corresponding manifestly real property/relation is also instantiated in that mind, and (b) when a manifestly real property/relation is instantiated in the manifestly real world, it’s not necessarily the case—indeed, it’s very often not the case— that its corresponding material concept is also tokened in an actual rational human mind.
Third, it follows specifically from condition (iv)—the linguistic cognitivism condition—that it must be possible to convey the content of a concept linguistically to someone else who’s not actually directly acquainted with or confronted with the individual thing or many things represented by that conceptual content right then and there, provided that they’ve been already been acquainted with them and that their episodic memory (Russell and Hanna, 2012) of that earlier acquaintance is cognitively accessible. For example, it must be possible to convey that conceptual content linguistically to someone else over the telephone, in the actual then-and-there absence of the individual thing or many things represented by that concept, provided that she has already been acquainted with them and that her memory of that earlier acquaintance is cognitively accessible (and also provided, of course, that she is sufficiently capable of hearing, and that the other obvious ceteris paribus conditions are all met).[iv]
Fourth, in view of the conjunction of the linguistic cognitivism condition and the concept-possession conditions, it must be strongly emphasized that LCTC does not entail that there cannot be non-linguistic concepts in any sense. On the contrary, as I indirectly indicated in passing in (vd) above, it’s plausible to hold that there are some non-linguistic yet also inherently concept-like mental contents, aka proto-concepts.
4. Concluding Postscript on Proto-Concepts
The claim I made at the end of the just-previous section—namely, that it’s plausible to hold that there are some non-linguistic yet also inherently concept-like mental contents, i.e., proto-concepts—is supported by the following four considerations.
(A) Many normal pre-linguistic human children (for example, many young children between six months and one year of age, and in the pre-toddler stage, and also many non-human minded animals) can effectively deploy and use inherently concept-like contents as object-categorizing, object-classifying, object-discriminating, and object-identifying devices for the purposes of cognition and intentional action (for example, in the case of pre-linguistic children, effectively recognizing their primary care-givers and telling them apart from other things and people) even if they cannot, strictly speaking, possess these inherently concept-like contents.
(B) These inherently concept-like contents can also be effectively deployed and used in the absence of the objects represented by them—for example, in the case of pre-linguistic children, insofar as they demand the constant presence of their primary care givers, food, warmth, etc.
(C) These inherently concept-like contents are intersubjectively shareable by other human minded animals and non-human minded animals alike.
(D) These inherently concept-like contents are present in their mental lives causally, phenomenologically, and semantically, but not by means of linguistic vehicles.
So according to LCTC, there are indeed some non-linguistic yet also inherently concept-like contents, namely proto-concepts. This is shown by the fact that that the conscious, cognitive or intentional, affective, or practical acts, states, or processes of some pre-linguistic humans or non-human minded animals do actually contain psychologically real inherently concept-like contents, as tokens of their corresponding mental representation types, that are effectively deployed and used by those animals in cognition, knowledge, and intentional action. But these proto-concepts inherently lack linguistic vehicles, and thus proto-concepts really cannot be possessed by those creatures.
The proto-concept corollary of LCTC, in turn, allows me to offer a tentative answer to the following very hard question: whether sentient or sensible non-human minded animals are also in any cognitively high-powered sense also sapient or discursive minded animals—that is, are they also believers, conceptualizers, judgers, or thinkers? My answer, as you’ve probably already anticipated, rejects the idea that there’s a simple unqualified positive or unqualified negative answer to that question. More precisely, what I’m saying is: “yes” in one sense, and “no” in another sense. My answer is “yes,” in the sense that a great many non-human minded animals (for example, cats, dogs, and horses), can deploy and use proto-concepts. To that extent, they are proto-believers, proto-conceptualizers, proto-judgers, and proto-thinkers. But my answer is also “no,” in the sense that proto-concepts are not concepts in the fullest, cognitively high-powered sense of that notion that’s provided by LCTC. Hence those same non-human minded animals are also not believers, conceptualizers, judgers, or thinkers in the fullest, cognitively high-powered sense of those notions. The possibility of proto-concepts obviously doesn’t in any way rule out the possibility of the sort of relatively weak but still quite substantive necessary connection between concepts and language[v] that LCTC provides. LCTC entails—again via its condition (iv), the linguistic cognitivism condition—that no material concept is such that it cannot be conveyed by means of some possible natural language to someone else who is not actually directly acquainted with or confronted by the individual thing or things represented by that concept right then and there, provided that they have already been acquainted with them and that their episodic memory of that earlier acquaintance is cognitively accessible. Every concept is thereby possessible by some rational minded animal or another, including of course rational human minded animals. Thus the possible natural linguistic expressibility of every concept suffices to guarantee the inherently intersubjectiveand non-solipsistic character of concepts in the strict sense, as well as the inherently intersubjectiveand non-solipsistic character of all inherently concept-like contents, hence of all proto-concepts, even for pre-linguistic human minde animals and non-linguistic non-human minded animals, whose proto-concept deployment and proto-concept use involve concept-like contents that lack linguistic vehicles, and this fact thereby prevents their possessing any conceptual contents, even though they do effectively deploy and use proto-concepts in cognition and intentional action. But at least some of the proto-concept-deploying and proto-concept-using minded animals whose mental representations lack linguistic vehicles, lack them only contingently, not necessarily. For obviously, many or even most normal, healthy pre-linguistic children actually grow up to become linguistic animals.
And not only that. According to LCTC, the later Wittgenstein was mistaken when he famously remarked that “if a lion could talk, we could not understand him” (Wittgenstein, 1953: p. 223e). In opposition to this, what LCTC implies is that counterfactually, if a lion could talk, that is, if a lion were to possess the cognitive capacities jointly constitutive of linguistic competence, then we would indeed be able to understand that lion. Think, for example, of the fictional leonine Lord Aslan in C.S. Lewis’s The Lion, the Witch, and the Wardrobe. Therefore, in these actual and counterfactual senses, all pre-linguistic human minded animal proto-conceptualizers and all non-linguistic non-human minded animal proto-conceptualizers are also proto-linguisticcreatures. Or to play a riff on another famous Wittgensteinian remark, namely, that
[o]ur language can be seen as an ancient city: a maze of little streets and squares, of old and new houses, and of houses with additions from various periods; and this surrounded by a multitude of new suburbs (Vororte) with straight regular streets and uniform houses (Wittgenstein, 1953: p. 8e, §18),
according to LCTC, pre-linguistic human proto-conceptualizers and non-linguistic non-human minded animal proto-conceptualizers don’t live in the maze-like, cognitively complex center of The City of Language. But they do all live in its cognitively simpler and more spread-out suburbs, while the human proto-conceptualizers prepare themselves to commute into or live in the center of The City at a later stage of their cognitive and social development.
NOTES
[i] It’s important to note that not all human thinking is discursive, conceptual, and mediated by language; on the contrary, some human thinking is non-discursive, essentially non-conceptual, and non-linguistic (Hanna, 2024). Moreover, as I’ll argue in section 4 below, some human thinking is proto-discursive and proto-conceptual but also non-linguistic.
[ii] This isn’t to say that there haven’t been many interesting and/or important studies of the nature of concepts published by post-classical Analytic philosophers or by cognitive psychologists; on the contrary. See, e.g., (Bealer, 1982; Peacocke, 1992; Fodor, 1998; Margolis and Laurence, 1999, 2015, 2023). And on Deleuze’s theory of concepts, see (Smith, 2012).
[iii] This section and the fourth section of this essay are adapted, with some additions, elisions, and revisions, from (Hanna, 2015: section 2.3, pp. 60-66).
[iv] Many thanks to Jane Heal for suggesting to me this informal “over-the-telephone test” for conceptuality.
[v] See, e.g., (Carruthers, 1998). Like Carruthers, I hold that there’s a substantive connection between conceptual thought and language; but unlike Carruthers, who is a higher-order thought theorist about consciousness, I don’t think that the substantive connection between conceptual thought and language inherently constrains the nature of consciousness, which, on my view, has an essentially non-conceptual basis in sensorimotor subjectivity and affect or emotion, i.e., feelings, desires, and passions (Hanna and Maiese, 2009).
REFERENCES
(Bealer, 1982). Bealer, G. Quality and Concept. Oxford: Clarendon/Oxford Univ. Press.
(Carruthers, 1998). Carruthers, P. Language, Thought, and Consciousness. Cambridge: Cambridge Univ. Press.
(Fodor, 1998). Fodor, J. Concepts. Oxford: Clarendon/Oxford Univ. Press.
(Hanna, 2005). Hanna, R. “Kant and Nonconceptual Content.” European Journal of Philosophy 13: 247-290. Also available online in preview at URL = <https://www.academia.edu/861111/Kant_and_Nonconceptual_Content>.
(Hanna, 2006). Hanna, R. Rationality and Logic. Cambridge: MIT Press. Also available online in preview HERE.
(Hanna, 2008). Hanna, R. ) “Kantian Non-Conceptualism,” Philosophical Studies 137: 41-64. Also available online in preview at URL = <https://www.academia.edu/861129/Kantian_Non_Conceptualism>.
(Hanna, 2015). Hanna, R. Cognition, Content, and the A Priori: A Study in the Philosophy of Mind and Knowledge . THE RATIONAL HUMAN CONDITION, Vol. 5. Oxford: Oxford Univ. Press. Available online in preview HERE.
(Hanna, 2020). Hanna, R., “The Essential Non-Conceptuality of the Imagination.” Contemporary Studies in Kantian Philosophy 5: 53-72. Available online at URL = <https://www.cckp.space/single-post/2020/06/15/CSKP5-2020-The-Essential-Non-Conceptuality-of-the-Imagination>.
(Hanna, 2021). Hanna, R. “The Debate About Non-Conceptual Content Revisited: On Corijn van Mazijk’s Perception and Reality in Kant, Husserl, and McDowell.” International Journal of Philosophical Studies: 29: 90-115. Available online at URL = <https://www.tandfonline.com/doi/abs/10.1080/09672559.2021.1873543> and in preview HERE.
(Hanna, 2024). Hanna, R. “Is Language Necessary For Human Thinking?” Unpublished MS. Available online at URL = <https://www.academia.edu/121294950/Is_Language_Necessary_For_Human_Thinking_June_2024_version_>.
(Hanna and Maiese, 2009). Hanna, R. and Maiese, M., Embodied Minds in Action. Oxford: Oxford Univ. Press. Available online in preview HERE.
(Koslicki, 2008). Koslicki, K. The Structure of Objects. Oxford: Oxford Univ. Press.
(Machery, 2009). Machery, E. Doing without Concepts. Oxford: Oxford Univ. Press.
(Margolis and Laurence, 1999). Margolis, E. and Laurence, S. (eds.) Concepts: Core Readings. Cambridge MA: MIT Press.
(Margolis and Laurence, 2015). Margolis, E. and Laurence, S. The Conceptual Mind: New Directions in the Study of Concepts. Cambridge MA: MIT Press.
(Margolis and Laurence, 2023). Margolis, E. and Laurence, S. “Concepts.” In E.N. Zalta and U. Nodelman (eds.), The Stanford Encyclopedia of Philosophy. Fall Edition. Available online at URL = <https://plato.stanford.edu/archives/fall2023/entries/concepts/>.
(My Sociology, 2021). “The Nature of Concepts.” YouTube. 14 September. Available online at URL = <https://www.youtube.com/watch?v=KoiuHN1CrH0>.
(Peacocke, 1992). Peacocke, C. A Study of Concepts. Cambridge MA: MIT Press.
(Queloz, 2021). Queloz, M. The Practical Origins of Ideas: Genealogy as Conceptual Reverse-Engineering. Oxford: Oxford Univ. Press, 2021.
(Russell and Hanna, 2012). Russell, J. and Hanna, R. “A Minimalist Approach to the Development of Episodic Memory.” Mind and Language 27: 29-54. Available online at URL = <https://onlinelibrary.wiley.com/doi/10.1111/j.1468-0017.2011.01434.x>.
(Smith, 2021). Smith, D.W. “On the Nature of Concepts.” Parallax 18: 62-73.
(Wittgenstein, 1953). Wittgenstein, L. Philosophical Investigations. Trans. G.E.M. Anscombe. New York: Macmillan.
Against Professional Philosophy is a sub-project of the online mega-project Philosophy Without Borders, which is home-based on Patreon here.
Please consider becoming a patron!
