Notes to Chapter 4 of Zabarella’s Liber De Mente Agente


Note 1: Zabarella is here referring to a section in his own commentary where he discusses the opening lines of De Anima III.5 (430a10ff.) where Aristotle writes,

Ἐπεὶ δ’ [ὥσπερ] ἐν ἁπάσῃ τῇ φύσει ἐστὶ [τι] τὸ μὲν ὕλη (10) ἑκάστῳ γένει (τοῦτο δὲ ὃ πάντα δυνάμει ἐκεῖνα), ἕτερον δὲ τὸ αἴτιον καὶ ποιητικόν, τῷ ποιεῖν πάντα, οἷον ἡ τέχνη πρὸς τὴν ὕλην πέπονθεν, ἀνάγκη καὶ ἐν τῇ ψυχῇ ὑπάρχειν ταύτας τὰς διαφοράς·

I translate:

Since [just as] in the whole of nature there is [what is] on the one hand the material principle for each kind of thing (since this is what is potentially all of them), and on the other, there is what acts as a cause and is a productive principle whereby it makes all things, just as a craft is related to material undergoing change, so it must also follow that these differences should pertain to the soul.

In Zabarella’s commentary on the passage, he begins the main part of his explication by writing that Aristotle’s intent was to show the need for an agent intellect  since the existence of a passive element implies the existence of an agent (see col. 1 p. 58). He structures Aristotle’s text in the form of an argument (syllogismus) as follows:

[1] Quoniam in omni natura reperitur aliquid, quod est tanquam materia, & patiens, & [2] aliquid, quod se habet, ut agens respectu illius patientis, [3] necesse est in anima intellectiva reperti has duas differentias, videlicet intellectum patientem & intellectum agentem. (col. 1, p. 58: the numbers here are mine and correspond to the “major,” and “minor” premises and to the “conclusion” respectively)

I translate:

Since in the whole of nature there is found [1] something that is like matter and a passive element and something [2] that is so disposed that it is like an agent with respect to that patient, [3] it is necessary that in the these differentiations should be found in the intellective soul, which is to say, the enactive and passive intellects.

He goes on to write that a controversy has emerged concerning the word “nature” (as well as specifically regarding the minor premise ([2] above) that applies directly to this chapter and finally to the central debate over the nature of the intellect: whether it should be considered as something human or as something separate from the human soul and perhaps even divine. The discussion has relevance to Aristotle’s conception of the relationship between the mind and body, as well as to the immorality of the soul.

Aquinas

According to Aquinas, the use of the term “nature” (species naturae: see quotation below) is applicable to the intellect as a whole and thus provides the evidence for thinking that Aristotle intended the passive intellect to be something that, like all other natural things, requires an agent of some kind in order to fully actualize its potential (again, see the reference to the “minor premise” [2] above). The term “natural” refers to the natural world, and Aristotle clearly applies the term to the intellect in some way, but Aquinas took Aristotle’s words to apply to an intellect that is part of our soul and is in us just as the “sensitive” and “vegetative” aspects of the soul are a part of the total human person. In fact, it might be noticed that Aquinas makes both aspects of the intellect part part of our mind, though he states that neither operates with a bodily organ and concludes that both intellects together are immortal and eternal when separate from the body (see Aquinas’ Commentary on the De Anima, Lectio X, of book III and especially sections 728, where he comments on the relationship between the intellect and nature and 734 where he discusses separability as it applies to the agent intellect). Here is Zabarella’s summation :

Aliqui nanque ut Sanctus Thomas talem minorem sumit, in qua anima accipiatur, vt species naturae, ut minor sit haec, at anima intellectiva est quaedam natura, ideo Sanctus Thomas per naturam, intelligit omne id, quod est imperfectum, & ad perfectione tendit, & exit de potentia ad actum, nam etiam intellectus
est huiusmodi.

I translate:

For, in fact, some, such as Saint Thomas, have taken up the aforesaid minor, in which the soul is understood as a natural form, as though the minor read as follows, “but the intellectual soul is a kind of nature….” Thus, Saint Thomas understands by “nature” everything that is imperfect and that tends to perfection, and that proceeds from potency to actuality: for even the intellect is this sort of thing. (Italics placed for emphasis)

Aristotle and Aquinas on the Agent-Patient relationship in the Natural World

The tie to “nature” that Aquinas has in view here can be linked explicitly to Aristotle’s Physics II.1 where he distinguishes two senses of nature, two factors involved in motion and rest one corresponding to a thing’s matter; the other to its form. In the first case matter by itself provides an impetus, as fire tends upward and stones downward.  In the case of thing’s “form,” it acts as a “goal” for its natural growth or corresponds to the end point of any natural process that all substances at any stage of coming-to-be have naturally within them, inasmuch as they are “natural” substances.

Finally, these strands may now be tied together. As Physics II.1 says, there are material and formal principles in all natural things.  But furthermore, it has been seen that for Aristotle, these material and formal principles relate to the potentiality and actuality of a thing in such a way that the material principle relates to its natural potential, whereas the realization of its form, or its arrival at the end of a process might be said to be the manifestation of that natural potential.  In a manner of speaking, one might say that the natural impetus to become an apple tree is in the material nature of an apple seed prior to becoming an apple tree; however, in another way, the flowering apple tree represents a complete notion of what it means to be an apple tree, and perhaps helps us understand the nature of apple trees better than anything else. Each case represents a sense in which the term “nature” might be used.

These considerations relate directly to Aquinas’ argument here: that the agent-patient relationship is one that Aristotle holds to applicable to all natural things. Therefore, it makes sense to think that Aristotle might be  taking the passive intellect to be the material element with the potential to understand (perhaps due to its inner impetus to understand, as Aristotle writes in Metaphysics I.1) whose end is to understand particular things in the proximate sense, and even all things in a more ultimate and final sense.

The agent intellect, on the other hand, ought to represent actuality itself and so does not become, being incapable of being acted upon (the sun in Aristotle’s analogy is perfectly apt as a reference point).  Being in act, it represents being-at-its-end, being fulfilled and complete with respect to its own potentiality so that the two are united in a state of pure actualization.  Yet, unlike the case of a fulfillment of a process, the agent intellect is, in its fulfillment also pure activity–as if its natural potency were to be active–so that being always active, it subsists in state of pure actualized potency. The sense, finally, that the two intellects interact is then applicable to an agent-patient relationship that mirrors the craft-material relationship.

Averroes

On the other hand, a view opposite to this was put forward by Averroes’, Aquinas’ clear opponent on this subject. According to Zabarella’s account, all the Greek commentators shared his opinion that Aristotle did not intent that the passive intellect should be understood as something properly speaking “natural” (sub natura tanquam speciem)(See the bottom of the column 1 on page 58). However, these interpreters understood Aristotle’s reference to the natural realm to be only an analogical one. A primary text in favor of the Averroist view, and one that Zabarella cites as a passage used by Averroists to support their view (see pdf p. 532, col. 2 of the De Mente Agente, ch. 12) is De Partibus Animalium, I, 1 641a32-b10, where Aristotle appears to say that the natural science should not include the study of the intellect, with the implication that it is not something natural. His argument includes the point that the intellect is not itself the source of motion in animals, and as may be seen in the quotations from Physics II.1, the “natural” pertains properly to what has within itself a source of motion or rest.  Such attributes as being a source of growth apply to what might be called the “vegetable” aspect of the soul, while attributes such as sensation or locomotion apply to the “animal” aspects of the soul and these pertain to natural science as subjects of discussion, as can be seen in the content of his zoological treatises-but neither of these apply to the intellect.

[To be continued…]

 

Zabarella’s De Mente Agente, Chapter 3


Main topics covered: the activity of the agent intellect, the role of the passive intellect in relation to the agent intellect, Simplicius’ account of the role of the agent intellect in relation to the passive intellect, refutation of Simplicius’ view, refutation of Jean of Jandun’s view, the analogy between sight and understanding, the Aristotelian theory of vision, light as a perfecting form, agency vs.  perfecting form, refutation of Aquinas and John Baconthorpe on the role of the agent intellect

Chapter 3 : The Refutation of the Aforesaid Opinions

Dictarum  opinionum Confutatio Cap. III

Nos vero aliorum errores patefaciendo simul ipsam rei veritatem iuxta Aristot. mentem declarabimus.

 

In primis a veritate prorsus aberrant illi, qui dicunt intellectum agentem agere in phantasmata, quoniam, ut modo alios considerasse dicebamus, Aristot. intellectum agentem invenit propter intellectum patibilem, & ut in eum ageret tanquam patientem.

 

 

Sed obstare nobis videntur verba Aristot. in contextu 18 quando intellectum agentem cum lumine comparat, & significat ipsum agere in phantasmata: iis tamen verbis  bene intellectis tolletur omnis difficultas, & ipsa rei veritas manifesta fiet. Sciendum igitur est, quando dicimus aliquid facere de tali potestate tale actu verbum illud (facere) ambiguum esse, & posse intelligi duobus modis, potest enim facere ut forma, potest etiam facere ut agens, forma enim humana adveniens materiae, facit de homine potestate hominem in actu,

 

 

neque ob id est agens, quia facit hominem ad modum formae, non ad modum agentis, homo vero generans alterum hominem dicitur facere hominem tanquam agens: differentia igitur est in hoc constituta, quod ignis generat alterum ignem & facit de igne potestate ignem actu, producendo in illa materia alteram  formam ignis sibi similem; illa vero altera forma producta facit de igne potestate ignem actu, non tamen  producendo aliam formam, sed ut forma ipsa quia non id facit ut agens: iungitur enim ipsi rei; at agens generans externum est, neque cum materia coniungitur.

Quando igitur Arist. dicit intellectum agentem facere phantasmata actu intelligibilia de intelligibilibus potestate, non ob id declarat rationem agentis, quia non id praestat ut agens, sed ut forma; iungitur enim phantasmatibus lumen intellectus agentis tanquam forma, qua redduntur motiva, & actu intelligibilia, sicuti lumen iungitur colori tanquam forma, & perfectio, qua redditur actu visibilis, & actu motivus visus,

neque dicitur lumen esse agens respectu colorum; quia nihil in eis producit ut agens, sed eis iungitur ut forma, per quam totum hoc coniunctum, color illuminatus, constituitur in esse obiectivo, & fit actu motivum visus:

 

hoc significavit Arist. in eo ipso loco, dum dixit intellectum agente facere omnia tanquam habitum quendam; habitus enim formam denotat, non causam effecticem, quia efficiens est a patiente disiunctum, forma vero iungitur materiae recipienti, & habitus rei habenti habitum; sic lumen coloribus haeret ut forma, & perfectio neque ad eos se habet ut agens ad patiens.

 

 

 

 

Non est igitur verum id, quod prima secta dicit, intellectum agentem agere in phantasmata, & argumentum eorum nullius roboris est. Decepti etiam sunt illi qui putarunt intellectum agentem agere in intellectum patientem tanquam agens distiunctum a phantasmatibus, quod extra phantasmata agendo perficiat intellectum patientem, & ipsi tribuat intellectionem; hoc enim dato, sequeretur intellectum patientem  posse etiam sine phantasmatibus intelligere, nempe sumendo congnitionem immediate a solo agente;

 

id tamen Aristot. adversatur, qui in context. 30 & 39 lib. 3 de Anima aperte dicit, fieri nunquam posse ut intellectus intelligat, nisi phantasma aliquod speculando: quare secundum Arist. omnis nostra intellectio fit ex motione facta a phantasmatibus. Immo non solum falsum est id, quod dicunt, intellectum agentem tribuere patienti suam cognitionem, sed neque ea ratione, qua est agens, est intelligens; quamvis enim necessario consequatur ut sit mens aliqua, & actu intelligens, tamen ea ratione, qua est agens, nihil formaliter intelligit, sed solum effective, quatenus in homine intellectionem producit:

 

 

quomodo autem id faciat, postea declarabimus, & ostendemus intellectum agentem esse quidem semper intelligentem, agere tamen non ut intelligentem, sed ut intelligibilem:

 

 

falsum id quoque est, quod Gendavensis dicit, intellectum agentem in intellectu patiente producere actum intelligendi; postea enim ostendimus intellectum patibilem sufficientem sibi esse ad promendam intellectionem sine ope intellectus agentis. Quocunque igitur modo dicatur intellectum patibilem pati ab intellectu agente, tanquam ad agente distinctio a phantasmatibus, & tanquam a cognoscente, falsum est & ab Arist. alienum. Quum autem utraque haec secta erraverit, erraverit etiam tertia, quae unum cum altero errorem coniunxit, ut considerantibus manifestum est.

 

 

But we, by bringing to light the errors of others, shall at the same time make clear the very truth of the matter according to the mind of Aristotle.
Against Aquinas and John Baconthorpe
First of all, they wander entirely astray from the truth who say that the actualizing intellect acts upon images, since, in the way we have said others considered it, Aristotle devised the agent intellect in relation to the passive intellect so that it might, in turn, act upon it as upon something playing a passive role.
The Solution of the Argument

But Aristotle’s words in contextu 18 seem to disagree with to ours when he compares the agent intellect to a light and indicates that it acts upon images. However, all difficulty is removed and the real truth of the matter becomes evident when these words are properly understood. It should be recognized, therefore, that when we say that something makes of whatever is such that it is in a state of potentiality, to be such that it is in a state of actuality, that word “makes” is ambiguous and can be understood in two ways: for it can be understood (1) as form and can also be understood (2) as an agent, since the human form, arriving in matter, “makes” a potential human an actual human and not because it is an agent (since it “makes” a human being according to the mode of form, and not according to the mode of agency), whereas a human, creating another human is said to “make” a human as an agent.  The difference consists in this: that fire generates another fire and makes of the potential for fire an actual fire, by producing in the matter another form of fire similar to it; however, that other form which was produced, brings about an actual fire from a potential fire, not by producing another form, but as form itself, since it does not bring it about as an agent: for it is brought together with the aforesaid matter in question-but a generating agent is external and does not become conjoined with matter.

Therefore, when Aristotle says that the agent intellect “makes” images actually intelligible from a state of potential intelligibility, he should not be taken as discussing the nature of an agent, because he does not present it as an agent, but as form: for the light of the agent intellect is joined to images as a form, whereby they are set in motion and are made actually intelligible, just as light is conjoined with color as a form and a perfection , whereby it is rendered actually visible, and vision is actually moved [See note 1].  Nor is light said to be an agent with respect to colors, since it produces nothing in them as an agent, but is joined to them as a form whereby the whole is conjoined. Therefore, when color is illuminated, a phantasm begins to exist as an object, and the sense of sight comes to be actually moved.

Aristotle indicated this in the same place where he said that the agent intellect makes all things in the way a “disposition” does [see note 2]: for a disposition indicates a form, not an efficient cause [see note 3], since being a motive cause of change is outside the role of what is essentially passive; however, a form is joined to the matter of the recipient and a disposition is held by what has it: thus, light adheres to colors as a form and a fulfillment (perfectio) and not to them in the way an agent is related to what is essentially passive.

Refutation of Simplicius and of Others

Therefore, it is not true that, as the first sect holds, the agent intellect acts upon images and their argument lacks any strength. They were also deceived who thought that the agent intellect acts upon the passive intellect such that the agent is not conjoined with images, since it brings to fulfillment and imparts intellectual activity to the intellect which is essentially passive, without acting upon images. By having said this, it follows that the essentially passive intellect can think the forms in images even without images, that is, by taking cognition immediately from the agent alone.

But this conflicts with Aristotle’s view, who in contexts 30 and 39 of the De Anima clearly says it can never happen  that the intellect might think without attending to an image. For that reason, according to Aristotle, our intellect becomes all things from the motion created by images.  Indeed, what they say is not only false, when they say that the agent intellect imparts its own cognitive activity to the the passive intellect, but neither, by that reasoning, considering the activity of the agent, is it able to think. For even though it necessarily follows that it is a mind and actually able to think, nevertheless, by that reasoning accoring to which it is an agent, it understands nothing formally, but only as an effect of something else, to the extent that it produces intellectual activity in humans.

But we will explain how it does so in what follows, and will show that the agent intellect is, in fact, always understanding, however not as  understanding, but as intelligible

.

[Reply to Jean of Jandun (Gandavensis)]

What Jean of Jandun claims is also false, that the agent intellect produces the act of understanding in the passive intellect: for we will later show the passive intellect to be sufficient unto itself to bring about understanding without the resources of the agent intellect. Therefore, in whatever way it might be said that the passive intellect is susceptible to the activity of the agent intellect, whether as to an agent distinct from images or as from thinking, it is false and foreign to Aristotle. And since each of these sects will have erred, they will also have erred in a third way as  one error entails another, so that it is made clear by the following considerations.

Notes to Chapter 3 of Zabarella’s De Mente Agente


Main topics covered: Aristotle’s theory of vision, intromission and extromission theories of vision, the relationship between perception and the intellect, the signification of the term “habitus”/hexis, hexis in relation to ethics, the relationship between the intellect as a hexis as a “maker” and as a “form,” efficient causality and the intellect, Aquinas and species as moving causes of intellection.

Note 1:  The term “motiva” might well seem to come in oddly here and it might even be wondered whether this is a reference to the extramission theory of vision that was widely accepted during the renaissance, since the analogy clearly pertains to the conditions necessary for sight (for two sources on the history of theories of vision, see below). However, Zabarella discusses the Aristotelian explanation of vision in terms of movement following Aristotle’s text, the main points of which are to be found in De Anima II.7 at 418b1:

Every color has in it the power to set in movement what is actually transparent;

at 419a9-10:

At present it is obvious that what is seen in light is always colour. That is why without the help of light colour remains invisible. Its being color at all means precisely its having in it the power to set in movement what is actually transparent, and the actuality of what is transparent is just light.

and again at 419a13-15:

Color sets in movement what is transparent, e.g. the air, and that, extending continuously from the object of the organ, sets the latter in movement. (1)

These three texts provide us with the basics of Aristotle’s theory of vision, according to which the color of an object acts upon an illuminated medium (i.e. the otherwise unoccupied translucent field of air between the eye and the visible object) to effect a change in the sense organs, which it accomplishes by means of a movement it produces within the medium.  Finding the eye to contain a watery substance, Aristotle thought that the water in the eye was susceptible of carrying the movement produced by the color of an object, so that it, in turn, produced sight within the eye (cf. De Sensu et Sensibilia, 438b where Aristotle makes explicit that the eye is capable of admitting light and that the eye is the proper site of the conjunction of the power of vision with what it seen-something that according to the Platonic extromissionist theory, required the emission of light outside the organ of the eye itself).

In line with the Aristotelian view, when commenting on the first quoted text above, Zabarella writes that the illuminated medium of vision is that “per quod fit actio obiecti in sensum hoc enim erit progredi ab obiecto ad operationem” (in Textus 68, page 119 (pdf pp. 336ff.)). Here the medium might be thought of as a necessary condition for action on the senses to take place and for them to be made active. Later in commenting on the same text, he writes that a property of color is that it is able to set a transparent medium in motion. Interestingly, he goes on to say that the precise way in which it causes this motion is by imprinting its “species,” its (potentially visible) form upon the medium of vision:

Movere autem perpsicuum nil aliud eft, quam imprimere in eo speciem suam, quae est alteratio perspectiva, de qua in praecedentibus dictum eft

Zabarella’s commentary therefore follows closely the Aristotelian version of the intromissionist theory  of vision, according to which the “eidos” (image, idea, form) of an image is transfered to the eye itself.

Attention to the elements in this theory (agent, patient, object) helps to clarify the way in which sight is analogous to thought and is Zabarella’s main preoccupation here.  The analogy is directly applicable to De Anima III.4, where Aristotle, in comparing the mind to the sense organs, says that it must be  capable of receiving the form of an object, so that it is potentially identical to its object without being its object (cf. 429a15-17):

ἀπαθὲς ἄρα δεῖ εἶναι, δεκτικὸν δὲ τοῦ εἴδους καὶ δυνάμει τοιοῦτον ἀλλὰ μὴ τοῦτο, καὶ ὁμοίως ἔχειν, ὥσπερ τὸ αἰσθητικὸν πρὸς τὰ αἰσθητά, οὕτω τὸν νοῦν πρὸς τὰ νοητά. (See the tlg for an online greek text)

I translate:

[The potential intellect] “Must be impassive, but receptive of the form and potentially such [i.e. as the form] while not becoming the former and to be of such a nature that, just as the sense organs are related to what is sensible, so the intellect is related to what is intelligible.”

In the text that follows, Zabarella distinguishes the role the agent intellect plays in the process of making potentially intelligible objects actually intelligible. His primary claim here is that, while it may be considered an agent, its action upon potentially intelligible objects is such that it involves the contribution of a form-light-that makes them actually intelligible.

(1) These are translated by J.A. Smith, in The Complete Works of Aristotle, The Revised Oxford Translation, Edited by Jonathan Barnes, vol.1, Bollingen: 1995, pp. 666-667.
https://www.princeton.edu/~cggross/neuroscientist_5_99_fire.pdf

http://nivea.psycho.univ-paris5.fr/FeelingSupplements/AncientVisions.htm#_edn6

Note 2: Zabarella is here referring to De Anima III.5 430a15, where Aristotle writes that the agent intellect that makes all things is kind of “disposition,” like light:

καὶ ἔστιν  μὲν τοιοῦτος νοῦς τῷ πάντα
γίνεσθαι δὲ τῷ πάντα ποιεῖνὡς ἕξις τιςοἷον τὸ φῶς·

The term habitus translates into Latin the Greek term ἕξις.  One difference between the two is that “ἕξις” is a substantive formed from the future tense of the verb “to have,” (thus “will have”) whereas habitus is a substantive apparently coined by Aristotle’s Latin translators directly from the past participle of the verb “to have, hold” (thus “has had”)  and has “clothing” or “garment” one puts on as a synonyms, whence we get the word “habit.” Thus, in Greek it signifies a “will have” and something one “has obtained” or “holds” or perhaps metaphorically, “puts on” in the Latin. In either case, in ethical contexts, the term ordinarily conveys the idea of an disposition to act in a certain way that is not easily acquired and that is kind of addition to one’s given nature. Aristotle commonly uses the term διάθεσις as a synonym (cf. for example Metaphysics V.20 where Aristotle discusses the definition of ἕξις), which can be analyzed morphologically into the notion of “directedness to an end” (διά) and a “placing,” “setting down,” or “positioning,” thus conveying the idea of an orientation to something.

In the context of Aristotelian ethics, the term fits neatly into the framework of character development as a fulfillment of one’s nature. For this reason, some have taken issue with “habit” as a translation in such contexts, since it does not adequately distinguish between the kind of “habit” (if the term fits at all) that is acquired by accident and one that is established deliberately in the attempt to attain some particular kind of excellence (See, for example, J.J. Glanville’s discussion of the term in The Material Logic of John of St. Thomas, p. 611, note 5, pdf page 641).  In the present context, a habitus refers not to an acquired, but to an innate disposition, however, making “disposition,” or perhaps even “faculty” or “power” better choices.

But the term can also be used more broadly to indicate the polar opposite of a privation (cf. Metaphysics, Book 10.4, 1055a33, whether referring to a natural state or to an acquired one.  In Categories 12a36, Aristotle uses the term to describe the possession as opposed to the privation of sight. There may be a sense in which the semantic richness of the term can be brought out with reference to the contrast between the potential and agent intellect in this respect: one represents the possession of a disposition toward a certain kind of activity, whereas the other represents its polar opposite: a privation of any ability to actualize knowledge on its own.  Furthermore, in the soul as a whole, the agent intellect is a habitus in the sense of being a disposition like sight, but also represents a natural ability that is the basis for further acquired abilities that lead to excellence.  In this sense it is a natural disposition such as the hand naturally has, to make things, being its intellectual counterpart, as Aristotle writes in De Anima III.8, 432a1-2,

ὥστε ἡ ψυχὴ ὥσπερ ἡ χείρ ἐστιν· καὶ γὰρ ἡ χεὶρ ὄργανόν ἐστιν ὀργάνων, καὶ ὁ νοῦς εἶδος εἰδῶν καὶ ἡ αἴσθησις εἶδος αἰσθητῶν.

“So that the soul is like the hand: for even as the hand is the tool of tools, so too the intellect is the form of forms and perception is the form of  perceptions”

(My trans.)

Extending this consideration to the ethical realm, it is a possession of the soul that, in making all things, is of even more fundamental importance than the hand, as that without which wisdom, practical or otherwise, could not be acquired.

References:

*The Greek texts quoted from the de Anima and Metaphysics here are taken from Ross’s critical editions, while the Categories quotes from the 1949 Oxford edition of Aristotle’s works.

L. Minio-Paluello, Aristotelis categoriae et liber de interpretatione, Oxford: Clarendon Press, 1949 (repr. 1966): 3-45 (1a1-15b32).

W.D. Ross, Aristotle. De anima, Oxford: Clarendon Press, 1961 (repr. 1967)

W.D. Ross, Aristotle’s metaphysics, 2 vols., Oxford: Clarendon Press, 1924 (repr. 1970 [of 1953 corr. edn.]): 1:980a21-1028a6; 2:1028a10-1093b29.

Note 3: In his treatise, De Speciebus Intelligibilibus (see pdf p.730), Zabarella attributes the view that species impressa, i.e. the forms of things that have been “impressed” upon the intellect in the same way that visual forms affect the eye, act as an efficient, motive cause of intellectual activity (sp. apprehending the “forms” of things) to Aquinas and Zimara, who he says followed him in his opinion on the issue:

sic igitur species intelligibilis impressa in intellectu est ratio formalis, qua intellectus operatur.; non operatur autem, nisi a phantasmate moueatur, ideo phantasma lo- cum habet obiecti mouentis

sed opinionem Thomae sequi videtur Zimara in theoremate. 82. vbi afferit fpeciem in intellectu impressam esse principium effectivum intellectionis, hoc eft rationem formalem, qua intellectus, qui possiblis dicitur , est intellectionis productivus .

For Aquinas, he references Commentary on the De Anima, Book III, Commentary 8 and the Summa Theologiae, I,85 as textual sources.

Thus, Zabarella distinguishes the causal role of the agent intellect from that of the object of thought, which might be supposed to move one to think. In fact, it acts upon the object in such a way that, according to the analogy, it brings to it the form of light, a “perfection” of the colors themselves, making potentially visible colors into actually visible (i.e. intelligible). Such a light would not thereby alter the nature of such colors, but make them actually visible inasmuch as they are already potentially visible (Aristotle says that it is specifically being colored that makes objects visible, with light as a further necessary cause-see references to note 1 above).

References:

See pdf pg. 730 or pp. 686-7 of Zabarella’s De Speciebus Intelligibilibus  in the collection, De Rebus Naturalibus,  Venice: 1590.

Zabarella’s On the Actualizing Mind, Chapter 2: Various Opinions Concerning the Proper Activities of the Actualizing Intellect


Chapter 2: Various Opinions Concerning the Activities Proper to the Actualizing Intellect

CERTUM est officium mentis agentis esse agere, hoc est trahere de potestate ad actum, sed in quodnam agere dicatur, & quomodo, obscurissima res est, & maxime controversa: alii namque dicunt ipsam agere in phantasmata, non in mentem patibilem; alii in mentem patibilem; alii in mentem patibililem, non in phantasmata; alii vero in ambo simul.

 

 

Prima sententia Latinorum fuit, praesertim D. Thomae, qui in 3 Libro de anima, & in prima parte summae quaest. 79, artic. 3 & 4, and in quaestionibus disputatis de spiritalibus creaturis articulo decimo & in locis fusissime de hac re loquitur, asserit rationem agentis in hoc esse constituta, ut agat in phantasmata:

 

 

eiusdem sententiae est Ioan. Bacconius in primo sententiarum  q. 2. prologi, ubi dicit officium intellectus agentis esse propter phantasmata, & totam eius actionem in phantasmatibus terminari, neque ulterius progredi.

Pro hac opinione videtur argumentum sumi ex verbis Aristot. in contex. 13, tertii, libri de anima, ubi declarans officium intellectus agentis inquit ipsum esse sicut lumen, nam lumen facit colores, qui potestate sunt, esse actu colores: quemadmodum igitur lumen non agit in oculum, sed in obiectum colorem, & ipsum ducit de potestate ad actum; ita intellectus agens agit in phantasmata, non in intellectum patibiliem;

 

 

ideo potest inde colligi argumentum tale: officium omnis agentis est trahere de potestate ad actum; sed Arist. hoc officium tribuit intellectui agenti ratione obiectorum, facit enim de intellectis potestate actu intellecta; ergo in phantasmata agit, non in intellectum patibilem.

 

 

 

Contra vero Simplicius videtur ei total actionem tribuere respectu intellectus patibilis; putat enim utrunque esse unam & eandem substantiam, & unum intellectum, qui ut in se manens dicatur agens, & ut progressus dicatur patibilis; quoniam ipse ut in se manens seipsum ut progressum ducit de potestate ad actum, de imperfectione ad perfectionem. Hanc sententiam Aver. in commentario quinto tertii libri de anima [cf. pdf page 411] Themistio attribuit, attamen non satis liquet Themistiarum fuisse huius opinionis (Note 2).

 

Hanc eandem sequitur Ioannes Gandavensis in quaestiones 24 & 25 tertii libri de anima, ubi totam actionem intellectus agentis inquit esse in intellectum patibilem, non in phantasmata, & ipsum in intellectu patibili producere actum intelligendi.

 

Pro hac sententia sumitur argumentum ex Arist. in contex. 17 eiusdem libri, ubi ex eo quod detur intellectus patiens, infert dari etiam intellectum agentem propterea quod omni patienti respondet aliquod agens: vult igitur Arist. intellectum patientem, & ut agat in eum:

 

 

ratio nanque illa vana esset, nisi agens ageret in illudmet patiens, cui respondere debet:

hoc idem Arist. exempla declarant; inquit enim intellectum agentem ita se habere ad intellectu patibilem, ut ars ad materiam se habet, ars autem in materiam agit; ita materia prima est patiens, in quod agunt omnia agentia naturalia; quare etiam intellectus patibilis dicitur patiens respectu agentis: nomen quoque ipsum hoc ostendit ; non enim phantasmata vocavit patientia, sed ipsum patibilem intellectum, in hunc itaque voluit agere intellectum agentem.

 

 

Averroes autem varius fuisse videtur: quandoque enim asserit actionem intellectus agentis requiri propter intellectum patibilem, ut ipsum ad actum ducat, atque perficiat; quandoque propter phantasmata, ut ea transferat de gradu in gradum, hoc est de materialibus faciat immaterialia, & de intellectis potestate actu intellecta:

 

quare videtur Averroes existimasse officium intellectus agentis requiri propter utranque actionem ductus

 

fortasse utrisque argumentis ex verbis Aristotelis sumptis, quibus duas priores sectas usas esse diximus; nam Aristotelis in contextu decimoseptimo tertii libri de anima dicere videtur intellectum agentem agere in intellectum patibilem, deinde in decimooctavo videtur assere ipsum agere etiam in phantasmata.

 

 

Ideo sententiam hanc nonnulli recentiores sequuti sunt, qui eam magis declarantes dixerunt intellectum agentem esse idem re cum intellectu patibili, & esse cognoscentem, & eatenus in illum agere, quatenus tribuit illi cognitionem ut hac ratione dicatur intellectionem producere, quia intellecui patibiili tribuit cognitionem, quam ipse agens prius habebat.  Haec sunt, quae ab aliis dicuntur, a me brevissime collecta.

It is certain that the activity proper to the actualizing mind is to act: that is, to draw from a potential state to an actual one. But what it is said to act upon and how is a most obscure matter and generates the greatest controversy.  For some say that is acts upon images, (see note 1) and not upon the mind in a state of potentiality; others, that it does act on the mind in a state of potency; and, to be sure, some say that it acts upon both together.

Thomas Aquinas

The first opinion was proposed by the Latin authors, but most notably by Saint Thomas, who, in his commentary on Book 3 of the de Anima and in the first part of the Summa, Q. 79, articles 3 and 4, and in Disputed Questions on Created Spirits, art. 10 and in very widely scattered places discusses this topic. He asserts that the nature of its agency consists in this: that it acts upon images.

 John Baconthorpe (John Bacon)

John Bacon was of this same opinion. In the first book of his Commentary on the Sentences, Q. 2 he writes that the activity proper to the actualizing intellect is in relation to images, and that the entirety of its action terminates in images and does not proceed further.

In favor of this opinion, it appears that the argument is taken from Aristotle’s words in Contextus 13, of Book 3 [de Anima 3.5], where, when he is speaking of the the activity proper to the actualizing intellect, Aristotle says that it is like a light, for it makes colors, which were in a state of potentiality, to exist as actual colors.  Therefore, just as light does not act upon the eye but on the colored object and takes it from a state of potentiality into actuality, so the actualizing intellect acts upon images and not upon the intellect in a state of potentiality.

It follows that an argument can be gathered therein of the following kind: [1] the activity proper to every actualizing element is to draw from a state of potentiality to actuality; moreover, [2] Aristotle attributes this function to the actualizing intellect on account of its objects, for it makes its objects actually understood from ones that are potentially understood. Therefore,  [3] it acts upon images and not in the passive intellect.

Simplicius

Simplicius, to the contrary, seems to attribute its entire activity to the passive intellect, for he thought that each was one and the same substance and one intellect, which, as abiding in itself  is called an agent, but as a process is called “passive.” As abiding in itself he considered it a “self-same” thing; as a process it moves from a state of potentiality to one of actuality, from an unfinished state to a state of completion. Averroes attributes this opinion to Themistius in his comment 5 on the third book of the de Animaalthough, it is not quite clear that this was the opinion of Themistius (Note 2).

Jean of Jandun (Johannes Gandavensis)

Jean of Jandun follows the same view in questions 24 and 25 in the third book of the Quaestiones de Anima, where he says that the entire action of the actualizing intellect is upon the passive intellect, not upon images, and that it brings about the act of understanding in the passive intellect.

In support of this view, an argument is taken from Aristotle’s works in context 17 of the same book, where, from the fact that a passive intellect is present, Aristotle infers that an actualizing intellect must also be present, accordingly, since to every potentiality there corresponds an actualizing agent, Therefore, it is claimed that Aristotle intends for there to be a passive intellect so that the actualizing intellect might act upon it. That reasoning would be pointless unless the actualizing agent were to act upon that very passive element to which it must correspond. When Aristotle discusses this same point he uses examples: for instance, he says that the actualizing agent is related to the potential intellect as a craft is related to matter and moreover, as craft acts on matter, : in the same way prime matter is a passive principle on which all natural agents act; wherefore, the passive intellect is also called “passive” with respect to an agent; [3] its very name indicates this too: for he did not call images passive, but the passive intellect itselfand so he intends for the agent intellect to act upon the latter.

Averroës

But Averroes seems to held a different opinion: for sometimes [a] he asserts that the action of the actualizing agent is required for the passive intellect, so that it might bring it to actuality and bring it to completion; and sometimes [b] he says that the activity of the agent intellect it is for the sake of images, so that it may transfer them from one step to the next-that is, so that it makes immaterial objects out of material ones and and bring potential intellgibiles to actual intelligibility. For that reason, Averroes seems to have supposed that the proper activity of the agent intellect is required for each action of taking from potentiality to actuality.

Both arguments may perhaps be taken from Aristotle’s words we said the two prior sects made use of: for Aristotle in contextus 17 of book III [De Anima III.5] of the De Anima seems to say that the actualizing intellect acts acts on the passive intellect, but then, in contextus 18 [cf. note 3], he seems to assert that it also acts upon images. Therefore, several recent authors have followed this opinion, who, expounding it to a greater extent, have said that the agent intellect is the same thing as the passive intellect, that it is a knows, and insofar as it acts upon it, gives it knowledge. By this reasoning it is said to produce intellectual activity, since it gives knowledge to the passive intellect that the agent intellect first posessed. These are the opinions which are given by others, which I have very briefly collected.

Notes to Chapter 2 of the Translation of Zabarella’s De Mente Agente


Main Topics Covered: the intellect and its relationship to images, forms and images, De Anima III.7, outline of primary theses covered in Averroes account of the Intellect with textual references, the influence of Themistius upon Averroes

Note 1: It might be wondered whether a broader term than “images” should be required here.  After all, the mind forms its concepts on the basis of mental content besides images. However, the phrase nequaquam sine phantasmate intelligit anima, (“the soul never understands without an image”) as a translation of Aristotle’s De Anima III.7, 431a15, was often commented on and understood in terms of the mind having the productions of the imagination as its objects of contemplation (For Scotus’ text, Quaestiones de Anima on this topic, click here).  In De Anima III.7, Aristotle distinguishes images from sense perceptions (cf. 431a14) in describing images as the proper objects of the mind. Aristotle may have had a case in mind such as the way a student of geometry might imagine various shapes in forming ideas about them.  An idea of importance to Thomists was that, in doing so, the student might be able to transcend images and attain pure ideas.  A clue that Aristotle may have had something like the latter in mind is when he says at the end of De Anima III.7 that the mind “thinks the forms in images,” and goes on to describe different types of images: natural (a snub nose presumably belonging to Socrates) and mathematical (the curvature of the nose). In commenting on this in his commentary on the De Anima Thomas writes:

Abstrahit tamen circa naturalia intellectus universale a particulari simili modo, inquantum intelligit naturam speciei sine principiis individuantibus, quæ non cadunt in definitione speciei (cf. the full text here)

The term “species” in Aquinas’ commentary refers to the form the intellect receives from the image having universal content.

Note 2: The text of Comment 5 contains the textual source for Averroes’ thesis that the material intellect (1) does not have a material form; (2) is conjoined with the agent intellect; and (3) is one for all humans. He can be seen attributing the first opinion to Themistius on pdf page 413 in the Crawford edition of the Long Commentary. Brief textual sources for all three are presented here for convenience:

Et hoc idem induxit Theofrastum et Themistium et
plures expositores ad opinandum quod intellectus materialis
est substantia neque generabilis neque corruptibilis.
60 Omne enim generabile et corruptibile est hoc; sed iam
demonstratum est quod iste non est hoc, neque corpus neque forma in
corpore. Et induxit eos ad opinandum, cum hoc, quod ista est sententia
Aristotelis.

The second reads as follows:

Et ideo opinatus est Themistius quod nos sumus intellectus
agens, et quod intellectus speculativus nichil est aliud
nisi continuatio intellectus agentis cum intellectu materiali
tantum. Et non est sicut existimavit, sed opinandum est quod in
anima sunt tres partes intellectus, quarum una est intellectus 570
recipiens, secunda autem est efficiens, tertia autem factum.
Et due istarum trium sunt eterne, scilicet agens et recipiens;
tertia autem est generabilis et corruptibilis uno
modo, eterna alio modo (cf. pdf pg. 430)

Finally, the third opinion (not attributed directly to Themistius) is here:

Quoniam, quia opinati sumus ex hoc sermone quod 575
intellectus materialis est unicus omnibus hominibus, et etiam                                          ex hoc sumus opinati quod species humana est eterna, ut
declaratum est in aliis locis, necesse est ut intellectus
materialis non sit denudatus a principiis naturalibus communibus
580 toti speciei humane, scilicet primis propositionibus
et formationibus singularibus communibus omnibus; hec enim
intellecta sunt unica secundum recipiens, et multa secundum
intentionem receptam. (cf. pdf pp. 430-431)

For further discussion on the relationship between Themistius and Averroes on the intellect, see “Themistius and the Development of Averroes Noetics,” by Richard C. Taylor.

Note 3: See pp. 461ff. in Crawford.

The Genesis and Motivation for Plato’s Theory of Forms: Summary and Comments on Harold Cherniss’ “The Philosophical Economy of the Theory of Ideas”


Harold Cherniss’ “The Philosophical Economy of the Theory of Ideas”: Summary and Comments  

Pdf of Cherniss’ article available here: http://uwch-4.humanities.washington.edu/Texts/CHERNISS/Cherniss%20Philosophical%20Economy%20of%20the%20Theory%20of%20Ideas.pdf

Key terms: A Priori, Recollection Argument, Eristic Argument, Hypothesis, Early Dialogues, Middle Dialogues, Fifth Century, Theory of Forms

Topics: The hypothetical method as leading to the theory of forms (the intellectual genesis of the theory of forms), the 5th Century background to the development of the hypothetical method, the importance of and need for an absolute ethical standard/underivative standard of reference,  the result of the “dialogues of search,” the eristic argument in the Meno, the recollection argument, teaching and knowing, the epistemological necessity of the theory of ideas

Outline

  1. Introduction
  2. Cherniss’ Account of the Fifth Century Background to Plato’s Discussion of Ethics in the Early and Middle Dialogues
  3. Plato and Relativism/Plato’s Motivation for Developing his Theory of Forms
  4. The Eristic Argument
  5. Comment: Text and Formalization of the Argument
  6. Comment: Evaluation
  7. Plato’s Response to the Eristic Paradox: the Recollection Theory of Knowledge
  8. Final Comments and Further Remarks on the Application of the Eristic Argument and Recollection Theory to the Contemporary Scene

Introduction

Harold Cherniss’ “Philosophical Theory of Ideas,” originally published in 1936 in the American Journal of Philology, concerns the genesis of Plato’s theory of Ideas or Forms, its rationale, and Plato’s own discussion of its attendant epistemological difficulties. The article may be divided into two parts, one that concerns the genesis of the theory in relation to ethics in the early and middle dialogues, and a second part that concerns epistemological questions in Plato’s later dialogues. The following will discuss the first part of the article (pp. 445-448).

Cherniss’ article has the special merit of providing a clear explanation for why the theory of Ideas or Forms was developed at all-a discussion that may well be very important for a proper appreciation and enjoyment of their content. Cherniss locates the primary texts for his theory within the dialectical battlefield of the Meno, specifically in two arguments labelled the “eristic argument” given by Meno (Socrates’ opponent in the debate) and Socrates’ response, an argument that produced his theory of “recollection.” After a brief discussion of the fifth century background to the discussion and the development of a criterion for an ethical absolute, the stage is set for the discussion in the Meno. The stage is then set for a summary of what Plato hoped to show and for further consideration of its significance.  

Cherniss’ Account of the Fifth Century Background

Plato’s development of his theory of Ideas or Forms grew out of his concern with ethical absolutes.  Cherniss describes Plato as having been concern to defend the possibility of such absolutes against a tide of relativism that had surfaced in the 5th century. Cherniss refers to such 5th century sources as the “dissoi logoi,” which he describes as discussing the relativity of such concepts as good and evil, the just and unjust, “fair” vs. “foul,” and even the possibility of teaching virtue (Cherniss 446). A further source is a papyrus fragment attributed to Antiphon the Sophist which Cherniss reads as contrasting conventional and natural justice, where conventional justice is “adventitious” and natural justice is simply what is “truly advantageous to each individual.  Those familiar with the Platonic dialogues will, of course, recognize these points of debate as their main themes.  The fact that they were so prominent as themes in the dialogues gives further evidence of the influence they likely had upon Plato.  

Plato and the Relativists/Plato’s Motivation for Developing the Theory of Forms

In seeking to contest the influence of these fifth century sources, Cherniss describes Plato as having recognized that an independent measure of goodness, justice, etc. would be necessary if he was to succeed.  Cherniss, considers that the early dialogues provide clear enough evidence of this point to venture to remark that the point ought not to be labored (Cherniss 447). Socrates does, in fact, repeatedly lead his interlocutors back to the methodological aim of discovering what justice, piety, love, etc. themselves are as opposed to their particular instances of them. Using individual examples of goodness or justice as standards of goodness and justice merely leads back into the trap of relativism: to do so is merely to “measure individuals against one another” leaving us without an independent standard for determining what makes some things, for example, just while others are unjust, and without a way to determine the similarities and differences among things one might want to classify as just or unjust (Cherniss 447). It might be considered that the refusal to accept particular instances amounts to Socrates’ way of funneling his interlocutor into a non-relative conception of ethical standards even as they fail to arrive at one.   

Thus, we arrive at the first step in the development of what Cherniss sees as Plato’s motivation for the theory of Forms or Ideas: “The possibility of ethical distinctions, then, implies objective differences which can be accounted for only by the hypothesis of objective ideas” (Cherniss 447). The justification for such an hypothesis is that it makes ethical absolutes possible, thereby making the determination of the true nature of justice, goodness, etc. possible; however, the defense and development the theory requires that it face certain epistemological (as opposed to methodological) questions-questions that properly concern knowledge itself as a subject of investigation.

One such question arises in the Meno, where the possibility of knowing such ethical paradigms as the Forms are supposed to represent is raised in the context of a discussion of whether virtue may be taught. After all, can what Socrates is searching for be discovered at all? The question is certainly worth asking. Cherniss comments that “a consistent and practical ethical theory depends upon an adequate epistemology” writing that the discussion of the nature of virtue underway in the Meno demands “a prior determination of the nature of virtue itself” (Cherniss 447-448). The dialogue brings the reader to consider these points when, after failing to give Socrates a satisfactory definition of virtue, he introduces a skeptical paradox designed to show that the search for knowledge is self-defeating in nature.    

The Eristic Argument

Enter the “eristic” argument. When Meno is unable to give a proper definition of virtue, he presents one that one might be recognized as being very much in tune with the skeptical tendencies of Plato’s 5th century adversaries.  It may be expressed concisely as the self-defeating proposition, “It is impossible to search either for the known or for the unknown” (cf. Cherniss 447).

Comment: The Text of the Eristic Argument

Going beyond Cherniss for a moment, It might be said, by way of initial clarification, that, while (a) it defies common sense and even logic to imagine searching for something one already has (since searching seems to imply something that is, at least in some sense, lost); on the other hand, (b) if one has no knowledge of what one is looking for, there is no means of finding it or even of knowing when one has found it (some means of discovery and of proof are equally essential for real  knowledge). Thus, it appears impossible to search either for what one already knows or for what is yet unknown. The Greek text and my own literal translation are placed below for the reader’s convenience.

The Greek text (Meno 80 d-e) may be translated as follows:

Μένων

καὶ τίνα τρόπον ζητήσεις, ὦ Σώκρατες, τοῦτο ὃ μὴ οἶσθα τὸ παράπαν ὅ τι ἐστίν; ποῖον γὰρ ὧν οὐκ οἶσθα προθέμενος ζητήσεις; ἢ εἰ καὶ ὅτι μάλιστα ἐντύχοις αὐτῷ, πῶς εἴσῃ ὅτι τοῦτό ἐστιν ὃ σὺ οὐκ ᾔδησθα;

Meno

And how will you look for something when you do not in any way know what it is?  What sort of thing is it you do not know that you are making the object of your search? Or above all, even if you should find it, how will you know that the former was what you did not know?

Notice that Socrates’ gloss of Meno’s objection is different:

Σωκράτης

μανθάνω οἷον βούλει λέγειν, ὦ Μένων. ὁρᾷς τοῦτον ὡς ἐριστικὸν λόγον κατάγεις, ὡς οὐκ ἄρα ἔστιν ζητεῖν ἀνθρώπῳ οὔτε ὃ οἶδε οὔτε ὃ μὴ οἶδε; οὔτε γὰρ ἂν ὅ γε οἶδεν ζητοῖ—οἶδεν γάρ, καὶ οὐδὲν δεῖ τῷ γε τοιούτῳ ζητήσεως—οὔτε ὃ μὴ οἶδεν—οὐδὲ γὰρ οἶδεν ὅτι ζητήσει.

I understand what you want to express, Meno.  Do you see what a contentious argument you are introducing, that is it not possible for anyone to inquire either into what he knows or into what he does not know? For he would neither seek what he does, in fact, know-for he [already] knows, and no one goes searching for that kind of thing-nor for what he does not know-since he knows nothing about what he will be looking for.

The gloss given above attempts to capture all the elements. It can be seen, in summary, that once Socrates’ own version of it is added to Meno’s, three basic elements of inquiry are in play: the starting point of an inquiry, the end point, and how one knows when the end has been reached.  These elements might, in turn, be related to Plato’s favorite key epistemological terms: a belief as a starting point, knowing as an end-point, and justification as a way of knowing one’s search is complete.  

[The following may be skipped if the reader is not inclined toward symbolic logic]

The search itself is paradoxical in nature because if one knows beforehand, there is no need to search, but if one doesn’t, one cannot find what one is looking for. It may be expressed succinctly in symbolic form as the proposition

                                                        ~K -> ~F & K -> ~F

where K represents “knowing” and F represents “finding knowledge.” As the formula shows, whether one knows or not, finding knowledge does not occur.  Assuming that one must be in either state (K v ~K), one cannot but help to fall into either side of the dilemma. This may be shown (purely for the reader’s enjoyment as it may happen) as follows:

  1. K V ~K
  2. ~K -> ~F & K -> ~F
  3. Assume K is true, then
  4. ~F
  5. Assume ~K is true
  6. Then ~F
  7. Therefore, K v ~K -> ~F

This amounts to a proof that, if the primary premisses (1 and 2) are true, it is useless to search for knowledge (self-defeating) because one either begins with knowledge or without knowledge-and it is impossible to avoid being either state.  

Comment: Evaluating the Eristic Argument

At this point, the reader may be experiencing a sense of exasperation. It appears that the premise, “We must either know, or not know” creates a false dilemma. The argument seems to overlook the distinction between partial and full knowledge. When the claim is made that in order to search for some x one must “know” x, apparently only the strong sense of knowing is considered. One might both have partial knowledge of what a Red Tailed Deer is, for example, and yet not “know” what it is in the stronger sense of the term. Not initially knowing in the strong sense does not appear to present any difficulty for really knowing what a Red Tailed Deer is later.  

In other words, it appears that the solution is simply to acknowledge that some form of induction or perhaps Aristotelian epagoge (lit. “going up”) is a valid method for arriving at full knowledge of a given thing. Setting aside the regress problem of numerical induction or verification (always another case), perhaps the best way to see why the eristic argument should have been taken seriously is to consider it in the context of an axiomatic model of knowledge.  Suppose we begin with partial knowledge and proceed on that basis: according to an axiomatic model, the partial knowledge we have won’t count as real knowledge until some principle is formulated that secures that what we think we know can really be considered knowledge at all.  We may, in other words, proceed inductively (or by epagoge) on the basis of a belief that we are right-but we might also be quite mistaken in that belief.  Examples of proceeding in the wrong direction on the basis of a false belief are not hard to find in the history of science: alchemy or ether theory in the early 20th century come to mind. In contemporary physics, string theory may or may not take us anywhere.  It is not until partial knowledge, based upon hypothesis or belief, is secured as an axiom that we can claim any real knowledge. That, in turn, leaves in doubt whether our starting point will lead us where we want to go.  

In such a context, really knowing will allow us to deduce whatever consequences we have not already drawn, and thus, to really know them too. Nevertheless, it might be insisted that unless some sort of “way up” involving induction or generalization of some kind is allowed, the search for knowledge really will be impossible. Certainly this is right.  But, by the same token,what the eristic argument points out is right.  It points out a problem with knowledge: we don’t really “know” anything until we know it, making it quite possible that, until we have arrived at a principle/axiom that secures our knowledge as knowledge, we may be mistaken.  In such a case, prior knowledge is vital to the search, inasmuch as it helps us to disregard beliefs that are likely to be false before pursuing them. Without prior knowledge, one might very well wonder how anyone can hope to find what they are looking for without prior knowledge of some kind to guide them.

Meno seems to find himself in such a case.  He believes he knows what virtue is, but his accounts of it, playing the role of axioms in Socrates’ critique, fail to secure his accounts as  knowledge.  They run into contradictions with other things we believe we know, making them appear unreliable-and knowledge, more than belief, ought to be reliable. Moreover, nothing he has said has even appeared to be partially correct, which would seem to leave him at a loss as to how to proceed. When he asks how it is possible to look for something lacking any knowledge of what it is, Meno may be read as having reached a point of exasperation either a) as to how to proceed b) as how to proceed lacking any idea of whether he is heading in the right direction (because he lacks any real knowledge of it) or c) as comprehending the problem of how the two are intertwined.  Socrates does intertwine them, by re-framing his question as the eristic argument formalized above with its apparently self-defeating results. What may be seen now is perhaps the extent to which it is, as representing a genuine problem about the search for knowledge-and at the same time isn’t.  

The Recollection Theory of Knowledge

Whatever possible solutions might be given for the dilemma, Socrates’ response is to hypothesize the “recollection” theory of knowledge.“Recollection” involves a prior direct knowledge of the Forms of things (absolute standards for things such as justice, goodness, virtue, etc.) that were available to known prior to one’s birth. One might alternatively think of such knowledge as “innate,” but Cherniss emphasizes that such knowledge of the Forms or Ideas was “direct” in nature, citing Meno 81 D 4-5 (Cherniss 448). Its application to the eristic argument might be given as follows:

  1. Such innate knowledge would give one a prior knowledge of virtue.
  2. The search, in this case, is for a definition of virtue, which would require a knowledge of its essential nature-a non-relative kind of knowledge of what it truly is.
  3. The eristic argument states that one cannot search either for what one knows or for what one does not know, but (1) presupposes that even the essential nature of virtue actually exists, was directly available, and may be known (in the strong sense of the term).

The way in which these premises lead to a solution might be made more explicit than the way in which Cherniss presents the argument as follows:

  1. Such knowledge has a dual nature that defies the paradoxical/self-defeating nature of the eristic argument: firstly, (1) secures the assumption that a true, non-relative definition of virtue really does exist; secondly, (1) allows for both potential and actual knowledge of that definition. This is illustrated by the example of Socrates helping the slave boy to discover knowledge of geometry within himself.
  2. According to the innate hypothesis (1), it would be possible to search for the definition of virtue because, although one does not know what one is searching for, in another sense there is the potential for it to be discovered since:
    1. One once had actual, direct, prior knowledge of it.  
    2. Finally, the process of acquiring such knowledge can be furthered by the efforts of a figure like Socrates, who can help to “learner” to recall it by giving clues that assist in the recollection process. This same process is sometimes described as Socratic midwifery.

It might be noticed that by posing the argument in this way, an overall coherence emerges among the elements of Cherniss’ depiction of the 5th century background, the motivation Socrates had for developing the recollection hypothesis, and the introduction of the eristic argument and its solution.  

The Recollection Hypothesis and A Priori Knowledge

A further implication is that a priori knowledge appears to be required in the search for genuine knowledge of virtue. Plato does not explore the path of reasoning on the basis of prior knowledge through generalization in order to arrive at full knowledge of virtue. One suspects that Cherniss is on the the right track in taking the relativism of the fifth century to be a foil for the present discussion, since a non-relative answer to the problem of what virtue is could not reliably be deduced from experience, since experience is always bound to the time, place, and circumstances of those who are acquiring it. In that case, any real answer would require a priori knowledge of virtue that does not come from our experience. Innatism is at least a way of meeting that requirement. With the supposition of an innate idea of virtue, it will at least be possible, in theory, to proceed from our experience to a true definition, even if that search is not carried out in the dialogue. The recollection argument attempts to provide some basis for such knowledge, which, if the paradox is taken seriously, appears to be necessary for any knowledge-a result which appears to have motivated the theory of Forms.

Final Comments and Further Remarks on Contemporary Application  

In sum, Cherniss’ basic position regarding the progression from the early to middle dialogues is that they can be read together as laying the groundwork and finding a solution to the problem of ethical relativism. What is at stake is that unless some solution to the eristic paradox is given, we must “surrender all possibility of considering ethical problems” (Cherniss 448). Unless some solution can be offered, the relativists have the upper hand and morality is ultimately a matter of custom or persuasion. But the even more serious result is that without some means of acquiring a priori knowledge, even the attempt to discover a non-relativistic way of knowing is doomed to failure.

The supposition of such hypotheses as the recollection theory requires, namely, that we have an immortal soul and that it, in some sense, exists in a “realm” wherein it may acquire direct knowledge of the Forms, may well appear extravagant to contemporary readers. These additional hypotheses might even seem to fatally weaken Plato’s argument; nevertheless, together, they do at least provide a working solution to the problem of a priori knowledge developed thousands of years before Kant’s approach to the same problem and the development of modern genetics. But finally, through Cherniss’ analysis, the eristic argument and its response had three further results for the progress of Plato’s dialogues: (a) as providing a motive for Plato’s discussion of the soul; (b) as a solution to the problem of ethical relativism; and (c) as foundational material for understanding the theory of Forms and its development. Finally, (d) it may be considered that problems the eristic argument raises have led to many of the “perennial” problems of philosophy.

The study of the eristic problem helps illuminate when, where, and why innatism might become necessary as a hypothesis: when the foundations for knowledge, whether universal or absolute enter into our consideration; where the problem of discovering a foundation for our knowledge prior to our everyday experience comes into view, and because, unless it is taken seriously, no other explanation for universal knowledge can be found. Are there other ways to satisfactorily oppose ethical relativism? Is there any other way to approach the problem of absolute knowledge? At the very least, because of its preliminary nature, it seems easy to conclude that the problem of a priori knowledge can be ranked as a meta-ethical question.  

Chomsky’s hypothesis of a universal grammar and Kant’s hypothesis of a set of ultimate mental categories provide two applications. In the case of Chomsky’s “universal grammar,” the problem is not so much related to the problem of absolute knowledge but to an absolute way of processing language, which is encoded in our DNA. If correct, the hypothesis would account for our ability to learn and master language use more easily than we otherwise would be able to. The same attempt to discover a foundation for a universal way of knowing (as opposed to theorizing for specific content) may be seen in the Kantian approach to cognition.  In either case, a point was reached where empirical investigation gave way to supposition and hypothesis.  That point, in each case, was reached when an evident fact could not be directly explained by empirical means.

Whatever course research takes, it begins with hypotheses, some of which are of sufficient generality that they may serve to direct it. The eristic problem and Plato’s solution impact investigations directly where the search for knowledge is required to reflect upon itself, its ultimate directives, and foundations. It would seem that if the apparent result of the eristic problem is taken seriously, a priori knowledge will always be necessary as a foundation for empirical, scientific knowledge.    

Looking out a Window on a Cold Saturday Morning


Lakeside Pondering

Whither this sport-utility realism,

This preparing-to-be-there,

Amid reticence and lassitude,

Against a child’s fortitude?

Is this not sheer resolution?

And where do these preparations end?

However duty-bound they be,

Does this road lead

Where they might be free?

“Nonsense”

Behold these delights:

A fox running in the snow

And geese taking flight,

Some snapping as they go.

Yes, and would the dog really like

To chase them on the lake,

Just as the wind appears to do,

Sweeping up and creating a swell?

Ah, and the wind imparts a thrill,

So fresh as it enlivens anew,

We breathe, and then let go.

 

 

 

 

 

The Best Way to Learn Ancient Greek


I have recently arrived at the point where I can more or less read almost anything at mid-level ancient Greek without too much difficulty.  I arrived at this point after many years of labor-the kind of labor I am sure anyone setting out with a grammar book and dictionary is undoubtedly familiar with.  Along the way, I was always thinking that there must be a way to learn ancient Greek that would see me through the labyrinth of forms and grammatical surprises.

One method I tried early on was one that helped me learn French in a much shorter period of time: reading along with  a translation.  The interesting thing is that this was very effective for one author: Aristotle.  I think I can attribute this to the fact that I was very familiar with Aristotle’s text before I started the work of translating it.  I reasoned that the situation might be the same as that with translating the New Testament: because of the familiarity I had with the translation I could almost guess at the Greek text and come out more or less right.  An added factor was I think that the range of vocabulary used in Aristotle’s texts is more or less very regular and predictable in most cases.  The latter was what helped with French.

My method was to look up one word on each line and simply slog through whatever was unclear until I had enough vocabulary to see my way through the text.  I worked well enough so that at the end of one summer I was able to read most of Aristotle’s corpus after spending around 3-4 hours per day on the project.  By building up my vocabulary, I eventually started to get an intuitive feel for the way the language worked.  Over time, I got better and better with it.

The difficulty was that the same method didn’t apply very well to Plato.  It applied somewhat well to the early dialogues, but not at all to the Republic.  The answer was simply that I lacked sufficient vocabulary, but a further problem was that the syntax remained more or less bewildering in ways that I felt should be worked out once I had enough vocabulary. Moreover, the vocabulary suddenly seemed far more difficult to learn.  I spent a long period of time attempting to work through to the same point I achieved with Aristotle, using the same method and telling myself that a tipping point was inevitable.

What finally produced the breakthrough was to finally spend a lot of time, not on vocabulary, but on the grammatical forms.  I don’t feel I actually became comfortable reading mid-level Greek until I knew not just some of the grammatical forms, but just about every list in the first half of Smyth to the point where I could recite them by heart.  It was only when I reached that point that the syntactical mysteries began to disappear and my level of reading reached what I achieved with Aristotle.  Once I did reach that point, using the method I used to read Aristotle helped me to make further progress.

Along the way, I was told that I should not rely on facing translations.  I have mixed feelings.  On the one hand, they certainly helped me read easier Greek in a short period of time and being able to read the translation kept the work from being absolute drudgery.  On the other hand, the method kept me far from the shores of mid-level Greek until I finally came home only after a serious dedication to learning every paradigm I could find. Suddenly, too, the vocabulary was far less mysterious.

Was the optimal thing to have begun with the massive effort at learning them from the beginning? That would certainly have smoothed the path to the next level. However practically speaking, when a beginner is confronted with a Greek text, the task of looking up many, many vocabulary words to find to right sense or to ensure that a preposition applies as it should can, if carried out long enough, be nothing short of stunting to the intelligence. One can get lost in a de-motivating maze of forms and mesh of vocabulary words to the point where all other worthwhile ends are finally sacrificed and the investment no longer appears to be worth the reward.

I believe the moral to this story, my own conclusion is that different levels of Greek demand different levels of familiarity with the grammar.  There is simply no way around learning every paradigm to the point where any one of them can be recited if mid-level Greek is to be really and truly attained.  I hope to move up to an advanced level in the near future organically by reading as much Greek as is necessary to feel like my vocabulary base at the mid level is sufficient, and along the way looking into every Grammatical nicety that comes my way.  I think I will have attained that level when I can read mid-level texts without translation.  I have an OCT text of the Republic that has beckoned for many years.

The Basic Conflict Between Phenomenology and Empiricism and Why it Matters


16195371_10154907549978187_4605671376749088166_n
Photo by the author, Gem Lapin Beaucoup
Perhaps someone can explain exactly how phenomenology is actually different from phenomenalism when it comes to the question of objectivity. I think that the only difference might be Huserl’s idea about intersubjectivity-but that is consistent with phenomenalism. At the core of each is the idea that perception is the only foundation for objectivity. The move that empiricists would want to make is to say that there must be some grounding for our subjectivity that must ground our perception of the world whether we perceive it or not.
The existence of things in the world, in other words, can’t just depend on our perception of them. Is this a trap that phenomenalism and phenomenology fall into? It does seem that intersubjectivity involves the presumption that someone must perceive an object in order for it to be considered a reality. I don’t think it manages to avoid Hume’s more objectionable skepticism about unperceived objects, in other words.
Intersubjectivity seems about as good as it gets if you want to take phenomena as the basis for reality, however, since it is basically consistent with science. And yet it does seem to fall into the trap of Cartesian skepticism that perception makes the world rather than the other way around, or that it is our perception that validates the existence of things rather than that it is the existence of things that validates our perception. The trap being simply that no actual world apart from my perception needs to exist at all for the phenomenological viewpoint to follow. In other words my reality is consistent with a dream if the phenomenologist is right.

Russia and the Emails as an Ethical Dilemma


Consider the following argument:
(1) Information is provided that appears to be evidence that someone has committed a crime or otherwise acted inappropriately in ways that are very significant. It may also provide evidence that exposes systematic corruption.
(2) But the information was provided by a source that is disreputable and possibly even hostile.
(3) Therefore the information should be ignored.

Some might say that the information itself and who is providing it are two separate issues. If you take the present actors out of the situation and look at it as a question of ethical norms, it does appear that a case could be made for either side and that there might actually be cases where you might agree with (3) for the good of the country. I think it poses an interesting ethical dilemma and not one easily resolved for or against (3).
In the case of the present election, I think it turns out that the further a Trump presidency fades from reality, the more it begins to look as if the Russians have actually done us a service by exposing possible corruption. It will hopefully make our political processes less suspect by exposing corruption where it exists and make us into less of a country where elites who are able to manipulate the system are able to get the upper hand. If there is a flaw it really is that Trump’s side is the side presenting the information and for that reason it will likely be ignored for partisan reasons. But it will be interesting to see what happens after the election.