Further Thoughts on Conceptualism vs. Empiricism as a Paradigm for Knowledge

Can a notion of conceptualism better explain the ideas we have of knowledge, belief and truth discernment than the empiricism that has come down to us from David Hume and has for so long dominated the Anglo philosophical tradition -- and informed later epistemological theories like logical atomism, logical positivism (also known as logical empiricism), linguistic analysis and ordinary language philosophy?

We tend to look at the world and see it manifested before us in sensory experiences and so we think, aha, there's the bottom line, that's what the world is and all it is, and so all of it, everything about it, is reducible to sensed information, to what is observable through our sensing faculties. The rest of the stuff that we think of as part of our world is made up stuff, made up by us or by others and we are the all too gullible victims of the many stories told, whether by others or which we have learned to tell ourselves.

What cannot be explanatorily reduced to accounts of observable phenomena, on this empiricist model, lacks reality and is, at best sometimes convenient fictions.

Or was Kant on the right track when he posited categories inherent in human beings which shape the raw inputs of sensory experience into the knowledge of the world we take ourselves to have?

1st Interlocutor

(1) All human knowledge, and all concepts of what knowledge can be, come from human experience, specifically the experience of one’s own consciousness. This experience comprises all the contents of consciousness, which includes what we might call the “sense” domain, the “thought” domain, the “conceptual” domain, and any other domains such as the “emotional” domain.

In short, everything (except one thing) reduces to the phenomenal. The phenomenal is (except for that one thing) the sole ontological entity. Every thought or intuition we have about the phenomenal is part of the phenomenal.

(2) The one other ontological entity is reality. We are aware of reality only through its impact on the phenomenal. This is Plato’s cave, and also speaks to Kant’s insights. The fire is reality and the phenomenal is the play (over time) of shadows. Like you can’t see the wind, you can only see how it blows the leaves around.

(3) To be human is to be chained in the cave forever. We cannot and will not ever see the sun. However, we can change the play of the shadows, through development of our conceptualizations. We can blow on the leaves too, but we cannot ever out-blow the wind.

(4) Since our conceptualizations also make the shadows dance, do our conceptualizations have any reality? This is an unanswerable question. Our conceptualizations are phenomenal. We can’t “see” outside them, as I understand Kant to have said in different words. Because we cannot “see” the fire, we can only speculate about how the fire produces the play of shadows. Because we cannot “see” ourselves, we can only speculate about how we influence the play of shadows.

We look at this glass as half empty. Seen half full, we see glorious achievements in science, philosophy, the arts, civilization and socio-ethics. Without our achievements, we would see only what brutes see – the play of shadows absent our achievements.

(5) In this context, consciousness is as “unseen” as reality. Both shape the play of shadows, and our only access to understanding both mechanisms is by observing their impacts on the shadows.

This “syncs” with the fundamental Cartesian “phenomenology”—the cogito—which branched into the conceptual-empirical dichotomy. And it “syncs” with 20th century philosophy, in which conceptualists and “meta-empiricists” (Wittgenstein, Quine, Kripke, Dummett) all came to the same conclusion, just phrased in different language.

2nd interlocutor

The problem you raised cannot be set in terms of a DICHOTOMY conceptual vs empirical. But, clearly, the Kantian point of view, IN ESSENCE, must be considered here.

Let us condense Kant’s philosophy in the following two points: (1) Transcendental philosophy concerns those “transzendentale Bedingungen der Möglichkeit der Erfahrung” (Kant’s Kritik der reinen Vernunft) [conditions of the possibility of experience; i.e., of the empirical knowledge], making possible the a priori knowledge: “reine Anschauungen (Formen der Sinnlichkeit)” [pure intuitions (the forms of sensibility)] and “reine Verstandesbegriffe” [pure concepts of intellect]; (2) since what we have here is a synthesis a priori of both, the transcendental knowledge contains the claim of an ultimate foundation (Letztbegründung). Let us ask: is the Kantian view relevant to contemporary understanding of the topic? Yes, it is!

As concerns (1), the concept “transzendental” has made a long career in the post-Kantian philosophy; it can be found, for example, in the following forms: „logische Form der Sprache (Welt)”, „Sprachspiel” (L. Wittgenstein), „Intentionalität” (E.Husserl), „ursprüngliche Wahrheit” (M.Heidegger), „sprachliche Rahmen” (R. Carnap), „conceptual frames” (A. Koyre), „paradigm” (Th. Kuhn), „ideal of natural order” (S.Toulmin), „relativized a priori” (H.Reichenbach, M.Friedman), „Systeme der reinen Begriffe” (A.Einstein), „Denkstruktur”(W.Heisenberg).

As concerns (2), the problem of „Letztbegründung” and therefore the analysis of the relation conceptual – empirical (more generally, the relation Subject – World) considered in the inseparability of the respective relata, must be given in terms of the following quotations:

„...phenomenology studies subjectivity as the condition of possibility for the emergence of sense and reality, not as the adequate cause and sufficient reason for what is constituted” (R.Sokolowski, 1964, 192-3).

„ Es ist sinnlos, über die Gestalthaftigkeit der Gegebenheit von Bewusstsein/Welt, Ich/Nicht-Ich hinauszugehen. Weder kann man den ‚Hintergrund’ Bewusstsein verabsolutieren, noch der ‚Vordergrund’ Welt (oder umgekehrt). Deshalb ist es auch ein sinnloses Unternehmen, sagen zu wollen, was der ‚letzte Seinsgrund’ ist. Beschäftigen wir uns lieber mit der Analyse der Verflochtenheit von Bewusstsein und Welt bzw. Ich-Andere-Welt. Darüberhinaus gibt es nichts Sinnvolles zu sagen” (Merleau-Ponty, 1966).

1st Interlocutor

I don't see conceptualism and empiricism in conflict -- they both seem to reject Plato (which is a good thing imho). I'd say that someone (like me) reasonably well read in the Anglo philosophical tradition realizes that Locke, Berkeley and Hume wrote what they did based on the science of their day. Were they to write again today, given the scientific knowledge we have now about the human mind, they would reframe their ideas and express their ideas in ways that would not be as limited as they might seem today. Logical atomism, logical positivism, linguistic analysis and ordinary language philosophy all have their flaws and limitations, but they are valid extensions of classic empiricism in good ways -- they do a good job of building on Hume that make his empiricism more nuanced, that have more sophistication and depth.

More importantly, "empiricist" philosophers like Wilfrid Sellars, Saul Kripke and W.V.O. Quine demolished the whole sense-data foundation of the Anglo philosophical tradition imho. I think you could say they espoused and advocated for conceptualism, although I'm not aware that they ever used the term. With more time to think about it, I could say more to explain this point. In any event, I think Quine and others moved empiricism past the naive empiricism of all their predecessors, so I would say "yes, leaving aside the name conceptualism for a moment, "Quinean empiricism" does a good job of explaining knowledge, etc., the way you are asking about, at least its foundations.

I see Kant's categories as a huge insight and step forward in western thought. That said, research in science about the human mind in the last few decades has yielded huge revelations about how the mind "categorizes" or processes both perceptions and concepts in ways that are essentially "pre-wired" in the brain, essentially independent of experience. I'm not sure how accurate Kant's categories are, in terms of being pre-wired or learned. Nonetheless the thought that such things as "categories" exist has been richly validated in this research. . . .

My Response

I was thinking about the empiricist paradigm for knowledge. Since the idea that knowledge is driven by observation, by evidence seen in the world took hold on the English philosophical tradition with Locke, Hume and, indeed, Berkeley, this paradigm has exerted a powerful hold on thinkers in this tradition, even though, we have seen it become increasingly sophisticated as is surely warranted. In the end we end up with a notion that for anything to be real, true, worthy of belief, it must somehow be reducible to an explanation that rests wholly on what can be observable.

But if THAT is true a whole host.of epistemological problems are kicked up. What about minds? Are they real or just a way of talking about observed behaviors? What about subjective experience, where we have a mental life of feelings, perceptions, thoughts, beliefs? Just behaviors? But if so what if no overt behaviors are ever manifested? If I have a secret and never tell anyone about it, did I not have a secret?

And what about values and moral claims? What underlies them and why should anyone's moral views have any logical traction with others if they don't happen to share my particular moral sensibilities? Yet can we really dispense with moral judgements in our lives or conclude they are just a kind of fiction or illusion we fall into and fool ourselves and others with?

The empiricist paradigm of knowledge leaves us in a place where a whole slew of things we take for granted in our lives, our interactions with others, seem to lose their reality if we suppose that all things that are real are wholly reducible to statements about observables.

My question was to wonder whether there isn't a better way of explaining ontology than the epistemic strategy of the empiricist model. This is not to dismiss the importance of observation to knowledge but to suggest that there is a better strategy, which includes respect for observation without insisting on reducing everything to observables.

I think McDowell has the right idea with his version of ontology that rests on concept formation and deployment which he calls "conceptualism." It looks to me like a viable alternative to the still dominant empiricist paradigm that works so well in the sciences but not quite so well in daily life.

2nd Interlocutor

. . . you are right, the conceptualism and empiricism are not in conflict! And to place the thematization of knowledge in this dichotomy conceptual – empirical is a wrong choice. Rather, as you mentioned, we have to adopt a Kantian view. I agree! Let us condense Kant’s philosophy in the following two points: (1) Transcendental philosophy concerns those “transzendentale Bedingungen der Möglichkeit der Erfahrung” (Kant’s Kritik der reinen Vernunft) [conditions of the possibility of experience; i.e., of the empirical knowledge], making possible the a priori knowledge: “reine Anschauungen (Formen der Sinnlichkeit)” [pure intuitions (the forms of sensibility)] and “reine Verstandesbegriffe” [pure concepts of intellect]; (2) since what we have here is a synthesis a priori of both, the transcendental knowledge contains the claim of an ultimate foundation (Letztbegründung). Let us ask: is the Kantian view relevant to contemporary understanding of the topic? Yes, it is!

As concerns (1), the concept “transzendental” has made a long career in the post-Kantian philosophy; it can be found, for example, in the following forms: „logische Form der Sprache (Welt)”, „Sprachspiel” (L. Wittgenstein), „Intentionalität” (E.Husserl), „ursprüngliche Wahrheit” (M.Heidegger), „sprachliche Rahmen” (R. Carnap), „conceptual frames” (A. Koyre), „paradigm” (Th. Kuhn), „ideal of natural order” (S.Toulmin), „relativized a priori” (H.Reichenbach, M.Friedman), „Systeme der reinen Begriffe” (A.Einstein), „Denkstruktur”(W.Heisenberg).

As concerns (2), the problem of „Letztbegründung” and therefore the analysis of the relation conceptual – empirical (more generally, the relation Subject – World) considered in the inseparability of the respective relata, must be given in terms of the following quotations:

„...phenomenology studies subjectivity as the condition of possibility for the emergence of sense and reality, not as the adequate cause and sufficient reason for what is constituted” (R.Sokolowski, 1964, 192-3).

„ Es ist sinnlos, über die Gestalthaftigkeit der Gegebenheit von Bewusstsein/Welt, Ich/Nicht-Ich hinauszugehen. Weder kann man den ‚Hintergrund’ Bewusstsein verabsolutieren, noch der ‚Vordergrund’ Welt (oder umgekehrt). Deshalb ist es auch ein sinnloses Unternehmen, sagen zu wollen, was der ‚letzte Seinsgrund’ ist. Beschäftigen wir uns lieber mit der Analyse der Verflochtenheit von Bewusstsein und Welt bzw. Ich-Andere-Welt. Darüberhinaus gibt es nichts Sinnvolles zu sagen” (Merleau-Ponty, 1966). (subl.V.D.).

Response

Of course, turning to a conceptualist rather than an empiricist strategy brings with it its own problems. If it's concepts all the way down, so to speak, how do we touch the world we think is out there beyond ourselves? How can observation, de rigeuer for science, connect us to a reality outside our own minds? Idealism become a kind of explanatory trap then, doesn't it? Kant's solution doesn't work either because, if the noumenon is outside our sensory capacities, we can't touch it in any sense that implies knowledge of it. It's just a posited blank spot for us. We can't even say HOW it can affect us let alone what it is. Yet for science to do what we expect of it, it must be grounded on epistemically valid observations. Otherwise there's an iron curtain of ignorance between us and whatever it is that constitutes the world. So for science to deliver genuine knowledge, we cannot suppose it to be locked out of reality. That way leads to mysticism and confusion.

There IS a way around this though, I suggest, and that is to reconceive the empiricist strategy. Instead of supposing empirical knowledge to be a function of what is built up from observations, we must explain observation as an explanatory strategy used for some kinds of referent but not all. Of course, the referents that we take to constitute our external physical world are significant because they are an overwhelmingly large part of our experience. But they cannot be the bottom line of knowledge (taken as what can be known in a justifiable way, what is knowable because it fits within a justificatory structure of implications). To describe, denote, explain, etc., is to know things in this latter way, as part of a larger network of things affecting other thigs. This, connectivity after all, IS the paradigm for knowing that things are this or that, and it is these things that constitute our observed world, what we call "physical."

And they are "known" because we can frame them through concepts, the necessary step in describing, explaining, denoting, etc. The condition of having this kind of knowledge, of observed things, is having concepts geared to sensory inputs. Thus a whole class of our concepts are tied to what our senses provide for us, observation capability. But not all our concepts fit THIS paradigm, are tied so tightly to observation. This is where the question of a defined bottom line for ontology, according ontic status to elements of our experience, kicks in. Thus what is real is conceptual at bottom but not untouched by a realm outside ourselves, i.e., what we take to be THERE through our ability to think about that realm, the world we experience and think of AS a world.

Idealism isn't the opposite of a realist view of the world, where that view is understood empirically as whatever can be observed. It is merely a confusion we fall into when we think of it as the opposite of the empirical. There is no wall between the real and the ideal, the world and the mind. These are just categories we apply for explanatory purposes in certain contexts. And concepts, as in having and using frameworks to sort and shape our sensory inputs, are the realm in which we generate the various domains that make up our world, what we deem inner vs. outer, observable vs. what isn't. But on this view, not being wholly reducible to observable elements is NOT ontologically disqualifying.

Here observability is NOT the measure of what's real, what we take to be real, but only a very substantial part of the world we recognize as real. In this way we discover room for many other kinds of referents in our reality than just what can be observed through our senses, those conceptions we typically rely on to get about in the world and which are tied most tightly to the sensory inputs that make up so much our experience.

3rd Interlocutor

Perhaps we may need to jettison inherited paradigms that view 'Knowledge' as Plato's 'Justified True Beliefs'; and treat 'Knowledge' as only that proper subset of such beliefs which can be categorically communicated, and which Gualtiero Piccinini terms as 'Factually Grounded Beliefs'.

This might, then, enable us to view seemingly conflicting, essentially inherited, perspectives as complementary.

In other words, we might need to explicitly recognise the complementary roles each scholar plays—or, ideally, ought and be taught to play—as a natural scientist, as a philosopher, and as a mathematician, when wearing at different moments . . .

Response

We're stuck with paradigms but that doesn't mean we can't pick and choose or adjust them. In the present case I think we tend towards reliance on observation (and so recognizing and dealing with observables) since visual observation is a dominant mode of observing for us. Of course other sense faculties also play a part so "observation" cannot be limited to the visual. And insofar as observation in this broader sense is our mode of contact with what we think of as a world, it stands to reason that it will be the favored paradigm for knowledge. But I don't think we should suppose that it applies across all possible knowable things. After all, there are so many things we speak of and recognize as "there" in our lives that we cannot suppose our world consists only of what can be observed (in fact or in principle).

The mistake I want to address is the one where we suppose that observability is the ultimate test of ontic status, of a thing's being taken to be real. We don't actually conduct our lives like that nor, I would argue, could we. We have to recognize that minds and thoughts and the private experience of subjectivity are also real but, while they may be explained as caused by physical (observable) phenomena, are not themselves observable phenomena. The same is true for features we impute to the acts we take, understood on a valuational grid that includes but is not limited to moral valuation. It seems to me that these clearly non-observable elements in or aspects of our world can and must be accommodated in our description of our world but cannot be if we suppose all knowable things must be reducible to physical phenomena (the physical being how we characterize things with observable (in a sensory way) status.

Thus a better paradigm than that which empiricism provides (all known things must be reduced to observable phenomena) when explaining how we know things in the world, is a conceptualist one, i.e., a model that explains things in the world as referents (objects of reference where reference is what we do with language) rather than objects of observation. This raises the problem of disconnectedness with the observable world, of course, so there must be some connection between concepts and the world or else nothing is ever really real and claiming other wise is illusion. I suggest that the way out of this seeming conundrum is to view the connection between world and belief as holistic, i.e., as occurring on a system scale rather than a one to one relation between words and their objects or statements and what they are statements about. The way to get there is through semantic holism, to recognize that words and statements don't stand in isolation from one another nor do the relations they express. The content of our statements and the words they're made up of is a function of the relations between all the words and statements within the particular languages in which they occur. The connection to the feedback which that which we call "the world" provides and our utterances happens across the broad range of our linguistic practices and is not limited to single words or statements: semantic holism.

This does not eliminate that which we recognize as outside ourselves, which imposes itself, its reality, on us. It merely puts it in perspective while saving the ontic status of those elements of our world which are not on the spectrum of observable phenomena.

Then, in response to a further thought by interlocutor #3 in which he advocated for a Kantian-based approach to knowledge consistent with Husserl's Phenomenology qua philosophy, I added this:

Isn't this what confuses us though, the idea that the only reality, the only "real" thing, is consciousness itself? Of course that cannot be the case because "reality" is a characterizing term we apply in various contexts to various referents of very different types, a function of linguistic practices. We don't want to deny reality to consciousness, minds, mindedness, subjectness, etc., but these are not the objects of reference we typically concern ourselves with when referring. Supposing we can or should reduce all referents to consciousness is no better solution than supposing we can or should reduce all referents to physical (observable) phenomena and, if we can't, we must presume they are fictions and don't really exist. These are all conceptual categories, whether delineated as observed phenomena or as the condition of observing, itself.

As such, they make up our world but that does not mean we should treat them as if they are referenceable, that is, discussable, in the same way -- or that recognizing them as existents hinges on their being answerable to the same criteria.

To study consciousness is to study its presence and how it works in creatures or even other entities (AI's some day?), ourselves included, in a scientific way. But phenomenology, as such, to explore the various forms and aspects of consciousness in ourselves, is not science nor can it be anything like science, yielding information to be applied elsewhere in the world. Is it philosophy then or the proper subject matter of philosophical inquiry? In a limited sense, I suppose. But self reflection that turns inward, abandoning the world as such, seems rather pointless. Epistemology as the study of how we know things, yes. But phenomenology as the study of being, existence, or itself? I am skeptical of the value of that.

The issue is epistemological as in the forms knowledge takes (about things and how to do things and the relation between these different knowledge types). What can we know and how can we, and do we, know it? That we live in an envelope of air as fish live in water is metaphor, a useful realization but what is it useful for? To remind us that objects of knowledge are that because they are objects of reference, that it is this capacity to refer that makes this kind of knowledge possible and that it is this kind of knowledge, that allows us to "see" our world in such a multifaceted way, that turns the environments in which we find ourselves, consisting of an ongoing wash of sensory inputs, into a world constituted by innumerable things on many different levels of existence.

What we characterize as physical is grounded in a certain status we accord the observable elements presented to us by our sense organs. Physical referents are that because our assent to their occurrence is geared to observation in the broad sense of that term already mentioned. But not all real things are ascertained to exist in THAT way. That's why conceptualism is a better knowledge paradigm than the empirical which presumes all real things consist only of what can be sensorily observed. That's not the world we actually recognize as ours. But conceptualism doesn't eliminate the empirical (knowledge by observation). It just relegates it to a single domain or aspect of what we mean when we speak of having knowledge. It puts the empirical in a certain perspective while replacing it as a paradigm for our knowledge claims.

It is the ability to have and use concepts, achieved through language, that makes knowing, as in knowing about anything, possible. And it is this possibility that distinguishes us from those other sentient creatures we share this planet with which do not have the ability to form and deploy concepts.

Conclusion:

Whether this will mark the end of the discussion is still to be seen but I think it has been helpful in allowing me, at least, to lay out the approach to epistemology I am undertaking here when I advocate for replacing the idea of empiricism as the basic model for knowledge, for what it means to know anything at all, with conceptualism, a model that rests on the supposition that knowledge that anything is the case is, at bottom, a function of our conceptualizing capacities. These, I think, are best explained by equating them with the capabilities language use provides its users with. It makes it possible for speakers to apply more basic animal signaling capacities, which we have, as creatures, to another distinct practice, referencing.

To refer, we must be able to delineate and distinguish and it is this ability that language enables. Indeed, concept development and deployment is the core capability that distinguishes language from the signaling mechanisms and capacities that are its origin. Concept formation and use are more than mere signaling to others that are receptive of the particular kinds of signals we can produce. Language enables signalers to become referrers through the capacity to denote, describe and report. And all of that enables explanation as such, i.e., our ability to explain and form and apply theories of how things work and what they are, debating and testing these as we go along.

Together, these functions rest on the generation and use of concepts, on the ability of the language-enabled thinker to link thoughts to other thoughts and so to represent them in ever more complex ways, sharing them through our utterances (and other signing mechanisms, e.g., gestures, written symbols, of various types) with others capable of doing the same. Thus formulating global pictures of the otherwise inchoate sensory inputs in which we bathe everyday of our waking lives.

Thus it is language, as such, a particular capability creatures like ourselves possess, that makes having a world rather than a mere environment of ever present stimuli possible and so enables creatures like ourselves(cooperating on a grander scale than our non-language enabled fellow creatures) to build knowledge about a world that extends beyond the immediate environment in which we find ourselves, by recognizing the possibility of being present or absent, being here or there, being current, past or in the future (as a possibility). The world we know is a world that exists through concepts which place everything in such relations. Only conceptualizing does that.

But that doesn't mean the world we have isn't real but only that to know it as a world is to be able to make distinctions between amalgams of phenomena qua stimuli, and between what's real and what isn't. For this we must have concepts.

How correct are our concepts? How good are they, that is, as in terms of how much we can rely on them when acting? Well we test them everyday and as long as they keep passing our tests, we keep and use them and even expand them. If and when they start failing our tests we have to adjust or even replace them.

That's what knowledge is about when we come down to it. Here even the empirical (the empiricist paradigm of knowledge) is seen to rest on our conceptualizing capabilities which we have thanks to our language capacity.