Avicenna’s Empiricism Revisited

I.  Introduction

In this presentation, I discuss the new approach to Avicenna’s empiricism as outlined in my recent book, Analytic Philosophy and Avicenna (Routledge, 2020). Against recent trends in Avicenna scholarship, I argue that Avicenna is not a proto-Lockean empiricist in that for him empirical knowledge does not rest on a foundation of self-authenticating non-inferential knowledge of matter of fact. I will then make sense of the notion of a non-self-authenticating, non-inferential empirical knowledge by drawing on analytic pragmatists like Wilfrid Sellars and John McDowell (who, in turn, inherit from Kant). I will conclude that Avicenna is also in possession of such a rendition of empirical knowledge, in part because of his appropriation of Aristotle’s account of experience.

II. Meno’s Paradox

            In book one, chapter six of Demonstration, of al-Shifā’, Avicenna re-states Meno’s paradox and then illustrates it with an example. Avicenna states:

It was mentioned that Meno, who addressed Socrates regarding the nullification of teaching and learning, said to him, ‘The seeker of some kind of knowledge will either be seeking knowledge of what he [already knows, in which case his quest will be inconsequential, or else he would be seeking knowledge of what he does not know – [in this case,] how would he know it, once he attains it?”[1]

In the Platonic dialogue, Meno, Socrates’s interlocutor, asks him about the teachability of virtue, and Socrates immediately alerts him to the complexity of that question and its presupposition that the meaning of virtue is unproblematically available. Meno, who is an especially thickheaded conversational partner, confidently derides Socrates’s ironic response. Socrates, in turn, refutes several of Meno’s attempts to define virtue. Meno, frustrated in his confrontation with Socrates, bursts into a rhetorical tirade, accusing him intentionally perplexing him and undermining his confidence in the knowledge he has of virtue.[2] He then confronts Socrates with a paradox that has borne his name throughout the history of philosophy. This is a paraphrase of the paradox:

If I know something, then I do not need to inquire into it; and if I do not know something then I cannot inquire into it.

Either I know something or I do not know it.

Therefore, I do not need to nor can I inquire into anything. In other words, inquiry is either unnecessary or impossible. [3]

Socrates’ response is peculiar. He could have easily replied by escaping between the horns of the dilemma and stating that Meno is appealing to a false dichotomy. He could have said that we do know a bit of something and we might want to know more of it. For example, we know something about physical things and then we then study physics to expand and refine our knowledge. Dominic Scott, in his commentary: Plato’s Meno, labels this reading of the dilemma as the paradox of inquiry and argues persuasively that Socrates’s response to the dilemma (not its re-statement) addresses a different problem, which he labels the paradox of discovery.[4] Socrates reply to Meno is ostensibly different from the aforementioned solution to the paradox of inquiry. He says:

As the soul is immortal, has been born often and has seen all things here and in the underworld, there is nothing which it has not learned; so it is in no way surprising that it can recollect the things it knew before, both about virtue and other things. As the whole of nature is akin, and the soul has learned everything, nothing prevents a man, after recalling one thing only – a process men call learning – discovering everything else for himself, if he is brave and does not tire of the search, for searching and learning are, as a whole, recollection.[5]

This response, which is prefaced by an invocation of divine poets and religious authorities whose task is account for divine inspiration, grasps the second horn and can be understood as a rhetorical ploy employed by Socrates to affect Meno whose understanding of what it is to know something is to recall the words of inspired teachers.[6] Scott, however, is correct to emphasize that Socrates’s response also highlights a different philosophical problem – the paradox of discovery – based on the second horn of the dilemma. This paradox is closer to the way Meno formulates the problem originally: “[h]ow are you to search for virtue, if you do not know what it is?” [7] Socrates replies that learning is recollecting our fore-knowledge. More precisely, to have knowledge of the unknown presupposes a fore-knowledge of it; otherwise we are grounding knowledge in something that is not knowledge.

Socrates’s response contains an early formulation of Wilfrid Sellars’s famous rejection of the Myth of the Given. The epistemological Given, according to Sellars, aims to raise the edifice of knowledge on self-authenticating claims to knowledge, which are miraculously endowed with a rational standing and ground our other claims to knowledge. Sellars rejects the Myth of the Given, accusing its perpetrators of committing the naturalistic fallacy: reducing epistemic facts about knowers to non-epistemic facts. Instead, he argues that “in character­izing an episode or state as that of knowing, we are not giving an empirical description of that episode or state; we are placing it in the logical space of reasons, of justifying and being able to justify what one says.”[8] Empirical descriptions, according to Sellars, express non-epistemic facts, and therefore any characterization of knowing must reject non-epistemic groundings and respect its place in the autonomous, logical space of reasons. In the light of the Sellarsian view, Socrates’s response to Meno’s paradox is a rejection of the epistemological Given in that it spurns the possibility of accounting for knowledge based on considerations that are not cases of knowing, such as descriptions of our contingent empirical transactions with par­ticulars. Plato’s Socrates responds that we must already have prior knowledge and if we do not have knowledge now it is because we have forgotten our fore-knowledge.

II.  Aristotle’s Solution to Meno’s Paradox

In the beginning of the Posterior Analytic, Aristotle declares that “all intellectual teaching and learning come about from already existing knowledge (gnōsis).”[9] To me, this echoes the Sellarsian thesis that knowings are episodes or states in the space of reasons. Aristotle calls the foreknowledge gnōsis and that is, of course, not the Platonic epistēmē. “Sometimes the verb is used interchangeably with epistatsthai; but it ranges considerably more widely,” writes Gail Fine. “Indeed, Aristotle uses it in cases that he does not count as knowledge. For example, he says that aisthēsis, even of the sort had by animals, is gnōsis. But he emphasizes that this gnōsis does not amount to epistēmē.”[10] My thesis is that Aristotle’s reply to Meno’s paradox involves a pre-existent gnōsis, which is, in fact, human aisthēsis, i.e., sense perception. Avicenna, as we will see, specifies further that this knowledge is not propositional, so it is not epistēmē, but as an instance of knowledge, it is mind-involving (i.e., conceptual) nevertheless.

It is my contention that Aristotle’s handling of the paradox of discovery in Meno’s dilemma takes place in Posterior Analytics 2.19. Aristotle begins 2.19 with the question concerning the demonstrative archai: how do we know them? He rejects the possibility that we get to know that of which we previously are ignorant. He asks: “if we get them without having them earlier, how might we become familiar with them and learn them from no pre-existing knowledge?”[11] This echoes Plato’s rejection of the myth of epistemic foundations that acquire their authority non-epistemically.

Aristotle also considers the Platonic possibility that foundational states of knowledge are innate, we possess them from birth, but “they escape notice.” He then rejects this possibility: “[w]ell, if we have them, it is absurd; for it results that we have pieces of knowledge more precise than demonstration and yet this escapes notice.”[12] In other words, in this possibility, the first principles (archai), as innate epistemic states must be self-evident and more perfect than demonstrated knowledge because their status as knowledge is not conferred on them demonstrably. How could we have these states and not notice them? This is an attack on Plato’s recollection thesis.

Aristotle, as we have seen already, allows for pre-existent knowledge but modifies the Platonic account so that the knowledge is not more accurate than knowledge as a developed state (hexis). He writes: “[w]e must possess a capacity of some sort, but not such as to be higher in accuracy than those developed states (hexis).”[13] In other words, there must be a capacity for knowledge, which as capacity is lower in accuracy than actualized knowledge states. For Aristotle, the states lower-in-accuracy-than-developed-knowledge are cases of sense perception (aisthēsis).[14] But to defend against the Platonic rejection of sense perception as knowledge he says that these states already involve the mind (nous).[15] Michael Ferejohn is correct to point out that Aristotle’s “metaphysics allows him to analyze perception, as Plato cannot, as a confrontation not just with an individual substance, but also with the universals that substance instantiates, since they are for him actually present at the site of perception.”[16] The universals are present at the site of perception because the intellect (nous) is drawn upon by the physical interaction. It is not that there is a physical friction and then intellectual activity is added on later, rather they must be simultaneous. If perception was anchored on the physical contact, Aristotle would have to be trading Plato’s recollection for what Sellars calls the epistemological Given. A more charitable and proper way of reading Aristotle is to say that he adjusts perception so as to be immune to the charge of givenness. For this the concepts, whose application is the proper function of judging and reasoning, must also be involved in perception. Therefore, I interpret Aristotle’s sense perception as the pre-existent knowledge (gnōsis) that he mentions in the beginning of Posterior Analytics. It is not a case of epistēmē, which is perhaps available for the first time through experience, but it is conceptual, as it involves the activity of the intellect and is the site of an instantiated universal.

III.  On Avicenna’s Modified Aristotelianism

As I have suggested above, Avicenna’s treatment of Plato’s response to Meno’s paradox betrays an awareness of the latter’s concern with the paradox of discovery. Avicenna, however, is also aware of Aristotle’s response to the same paradox in the Posterior Analytics 2.19, and opts in favor of Aristotle. Recall that Aristotle, in his concern with the paradox of discovery in Posterior Analytics 2.19, argues that what is prior to knowledge (epistēmē) is fore-knowledge or immediate cognition (gnōsis), which is an outcome of the cooperation between sensation and intellect (nous). In Avicenna’s locution in the Demonstration 1.6 of the Book of Healing, gnōsis is rendered as ma‘rifa (immediate cognition) and epistēmē is translated as ‘ilm (scientific knowledge). Avicenna relates sense perception (al-ḥiss) to immediate cognition thus:

If then through the senses we attain immediate cognition (ma‘rifa) that he [Zayd] exists and that he is human, without this being sought after through a [demonstrative] syllogism (maṭlūban) or [this being] taught, and if this is connected with [universal] knowledge (‘ilm), which is realized for us, also without a [demonstrative] syllogism, in [the] manner by which the connection that in itself brings about a third knowledge [‘ilm], we would then know that Zayd is an animal  … Of the two, immediate cognition comes about through sensation, [universal] knowledge through the intellect.[17]

So Zayd is known to us through sense perception, but only potentially in terms of assent because what is sensed is prior to a judgment or assent about it. We don’t know about him as specified in actuality (makhṣūṣ bi-l-fi‘l), as that knowledge is the conclusion of a syllogistic demonstration (e.g., Zayd is an animal), with a universal premise (all humans are animals).[18] Furthermore, sense perception yields non-inferential knowledge: Avicenna says, “[i]f through sensation we observe (shāhadnā) some particulars without our seeking them, they would immediately fall within [the category] of being actual, within the category of first knowledge (‘ilm al-awwal).”[19] First knowledge includes the premises of first syllogisms: “self-evident premises or else acquired by induction (istiqrā’), experience (tajriba), or sensory perception without a syllogism.”[20]

Knowledge as assent, for Avicenna, is based fundamentally on sense perception, because experience and induction, as cases of “first” assent, are also based on sense perception (which is not an assent). By reading Avicenna along with Aristotle, it seems that sense perception should involve the participation of the intellect. I also submit that Avicenna further develops the Aristotelian account in Posterior Analytics 2.19 by attributing a pre-propositional structure to sensory intelligibility.

            That the intellect is involved in sense perception is already implied by the epistemic status that sense perception has in Avicenna’s account. Wolfson, in “The Terms Taṣawwur and Taṣdīq in Arabic Philosophy and Their Greek, Latin and Hebrew Equivalents,” argues that while the distinction between conceptualization (taṣawwur) and assertion (taṣdīq) is typically traced back to Alfarabi,[21] it is actually rooted in the Aristotelian distinction between intellection (noēsis) and judgment (logos apophantikos).[22] Wolfson, by drawing on Averroes’s commentaries on the Arabic translations of Aristotle, defends the claim that taṣawwur (conceptualization) is the Arabic rendition of the Aristotelian intellection (noēsis),[23] which as we have seen, is involved in sense perception. Wolfson, however, seem to overlook that taṣawwur is holistic and the primary concepts are primary, in the Aristotelian sense, in that they establish the unity of categories which gets implicated in sense perception. For Avicenna and Aristotle, the whole of the logical space of concepts is drawn on in sense perception, and the human discursive activity carves out the concepts and expresses them in inferentially related judgments. These judgments are ultimately answerable to the conceptualizations (taṣawwur) in sensory perception.

Drawing on The Healing, Wolfson points out what we have already seen: Avicenna refers to taṣawwur as the first knowledge, (al-’ilm al-awwal).[24] While Wolfson does not explicitly make the connection between sense perception and taṣawwur,[25] he does emphasize that the latter, as “knowledge,” is not propositional. In other words, taṣawwur is not an assertion but a conceptualization. To emphasize the latter, Wolfson also points out that, according to Avicenna, taṣawwur is neither true nor false. Hence, the content of sense perception’s taṣawwur is not propositional. Moreover, the conceptualization in sense perception is not representational. In other words, the epistemic status of sense perception is brought out in the non-propositional yet intellectual conceptualization in which the categorial unity is implicated. This in turn grounds all of our knowledge.

In my book, I clarify this point in relation to McDowell’s later criticisms of the Sellarsian view and his own position in Mind and World. In order to respect the time limitation on my presentation, I skip this discussion.

IV.  Reply to Gutas

Dimitri Gutas defends an empiricist reading of Avicenna and maintains that, for Avicenna, knowledge is accomplished within the limits of human intellect without any aid from an external source. On this reading, Gutas emphasizes what the emanatists call “preparatory processes,” for the mind’s reception of the intelligible forms from the Active Intellect.[26] For Gutas, these perceptual processes are not preparatory in the emanatist sense but in the abstractive sense and result in knowledge. He writes, “[w]hat has to be kept in mind is that for Avicenna the concept of the emanation of the intelligibles from the active intellect has its place in his cosmology and it serves to solve essentially an ontological problem, not an epistemological one, which is the location of the intelligibles.”[27] Gutas thinks that Avicenna’s references to the involvement of the Active Intellect in the process of knowing serve to allow for an ontological solution to the problem of intellectual memory. For if the forms were stored in the human intellect, then the human mind would constantly think them, and that is impossible.

As apparent from the above reading of the primacy of sense perception in Avicenna’s epistemology, I also defend an empiricist reading of Avicenna, but my reading diverges from that of Gutas, who interprets Avicenna as a proto-Lockean empiricist.[28] My position shares Gutas’s recoil from an emanatism that interprets our sensory experience as not knowledge but a mere preparation for it. In my view, emanation is already involved in sensory experience vis-à-vis the illuminations of the external Active Intellect which informs our sensory experience and provides a depository for acquired intelligible forms. My point that intellectual emanation is involved in sensory experience is not just exegetical; rather, a rigorous account of such involvement, I propose, is also philosophically attractive. To that end, I urge that we learn from Sellars and his successors and allow for a non-mythological, non-inferential knowledge of matter of fact. Sellars regards sensory experience as the seamless product of the cooperation of sensory receptivity and rational activity. His arguments and the contributions of his successors to them constitute a cutting-edge discourse in contemporary philosophy, and I draw on them to inform a more sophisticated interpretation of Avicenna’s epistemology.