Is human a homonym for Aristotle?

Ward, Julie K.

I Some Puzzles about Human Capacities

By combining specific claims that Aristotle makes about human nature and capacities with other claims about substances, two aporiai arise that require an investigation and if possible, a resolution. The general issue to be taken up concerns whether the term 'human' or the characteristic of being human is synonymous, or said in the same way, across various social and political groups (Greek, barbarian, free, slave, man, woman, etc.). (1) While the answer might seem obvious given the supposition that human is a univocal characteristic, the textual support for this view is neither clear nor consistent. As a consequence, we need to weigh whether the term should be considered to refer to a univocal characteristic shared among human beings (be synonymous) or to a non-identical characteristic (be homonymous). In this section, I shall describe the two problems briefly and give an explanation why I take them to be difficulties that merit attention.

The first aporia to be examined arises from certain assumptions related to the idea that human beings constitute the same natural kind, or are the same qua substance. Given that they are identical as substance, they cannot differ in degree insofar as no substance, and therefore, no human being as substance is more or less a substance than another (cf. Cat 3b33-4). As Aristotle states in Categories 5, 'If this substance is human, it will not be more or less human, either than itself or than another human' (Cat 3b37-8). (2) While this denial clearly carries an anti-Platonic sentiment, the implication is that human signifies an invariant, non-scalar characteristic. (3) In this regard, if we consider various human beings as primary substances, there is no difference of degree among them with regard to their substantiality, their humanness. Nor is the notion that co-specific individuals share synonymous characteristics challenged by the central books of Metaphysics (e.g., Z) that weigh the possibility of non-synonymy for substance (ousia) insofar as it signifies, among other things, form, matter, or the composite. (4) Metaphysics Z 3 and 10, for example, support the notion that members belonging to the same kind of substance are formally identical, or possess the same species-form. (5) So, it seems uncontroversial to understand Aristotle as being committed to holding a formal equality, what we might term 'metaphysical equality', in regard to human beings as co-specific members. Having noted the evidence with regard to the univocity of substance, there is a different set of views about humans that seems to overturn the present data. Specifically, he seems to contradict or at a minimum restrict his view about metaphysical equality in Politics. For example, in Politics I 7 and I 13, he rejects the idea that all humans possess qualitatively identical rational capacities. As he states in I 13, 'all have the same parts of the soul, but have them in different ways' (Pol 1260a10-12), having in mind the deliberative faculty (to bouleutikon) which, it is implied, is not universally present among all humans. Here lies the rub. Assuming that the capacity for deliberation belongs to humans in virtue of having the capacity for reason, it must be per se, or essential, to humans. (6) By connecting this with the previous claim, we find that some humans cannot lack, and yet do lack, a per se property of being human. We thus arrive at one aporia about human nature that may be stated as follows, 'How can humans considered as substances possessing the same species-form fail to have the same per se properties?' The corollary to this aporia concerns a question about predication, namely, whether the term 'human' can be applied synonymously across all cases, or with regard to all human beings. Alternatively, assuming Aristotle's distinction in Categories 1 between synonymy and homonymy, we may ask whether the term 'human' or the real property human is supposed to be taken as homonymous, not by chance (apo tuches), (7) but in some systematic fashion. (8)

The second aporia to be discussed develops from claims centering on the disputed status of what may be termed the psychological, rather than the metaphysical, equality of human beings. This problem arises from a tension between statements about the human capacity for moral development as sketched in Nicomachean Ethics II and those made about the variations in deliberative capacity already described from Politics I. One half of the present inconsistency stems from the familiar account about moral habituation from Nicomachean Ethics II 1, where Aristotle asserts that humans develop a fixed moral ability, or hexis, by way of practical training, as in the arts. In Politics, however, he appears to neglect the thesis about the general human capacity for moral virtue, arguing, in effect, that some groups of humans are prevented from the outset in attaining it. So, in Politics I 7, 13, and VII 7, Aristotle excludes citizen women, barbarians (non-Greeks), and natural slaves (both men and women) from being considered potentially virtuous citizens on the grounds that they are unable to deliberate either at all or properly. In this way, we arrive at a second aporia about human nature that may be expressed as the following question: 'If human nature is neither good nor bad in itself but capable of moral virtue depending on habituation, why it is impossible for some human beings to become virtuous even if they are exposed to the proper methods of moral habituation and training?' The two problems are inter-related by virtue of a shared minor premise: the first one uses a major premise about a formal feature of substance and a minor about differing deliberative capacity, and the second, a major premise about moral habituation and the minor premise about differing deliberative capacity.

Before opening investigation of these aporiai, I wish to raise and answer two general objections, or issues, that may be considered problems with the project overall. First, it might be objected that Aristotle is concerned with strictly metaphysical problems in the Categories, Metaphysics, and de Anima, and that these stand in contrast to the moral and political problems addressed in Nicomachean Ethics and Politics, with the result that cross-textual comparisons are either compromised or futile. Although there are, admittedly, certain difficulties involved with the kind of investigation proposed, it seems fair to expect consistency among texts where the discussion is unified by subject-matter. In the case of the aporiai, these claims share a common subject-matter insofar as they are concerned with one natural kind, human beings, and their natural abilities. Thus, to find the thesis about metaphysical equality as it applies to substance so conspicuously absent from Aristotle's discussion of human nature and abilities in Politics seems to suggest a lapse of consistency and deserves an explanation.

Another criticism that might be thought to diffuse the aporiai I have posed concerns postulating extra-theoretical claims to account for Aristotle's inconsistencies. In general, I find this line of interpretation unsatisfactory in that it serves, essentially, to block the possible inconsistency and avoid a genuine explanation. It may be noted, as well, that postulating extra-theoretical claims for theoretical conclusions is not uncommon in the scholarship surrounding Aristotle's views about human nature with regard to women and slaves. One exception in the recent scholarship is an essay by Malcolm Schofield on the foundations of Aristotle's theory of natural slavery. Herein Schofield considers the role of extra-theoretical elements, such as cultural elitism or tacit ideology, rather than theoretical commitments as being determinants in Aristotle's theory. (9) The point is that if the deciding cause for natural slavery were provided by tacit assumptions about class superiority or sheer prejudice against non-Greeks, the conclusion would not be genuinely theoretical in the sense of following from Aristotelian premises. Equally, to account for the connection that Aristotle draws between women and barbarians in regard to deliberative and political capacities simply by posing sexism or cultural elitism as the cause is different from being theoretically adequate. The present essay rejects the kind of explanation that introduces non-theoretical claims as the decisive factors for explicit conclusions in two ways. By formulating the issues of deliberative and rational capacities against the larger backdrop of being human, the paper seeks to avoid covering the by-now familiar ground that has been discussed by other scholars in regard to the issues of Aristotle's view on women, slaves, and deliberation. (10) Additionally, and more to the point, it is theoretically unsatisfactory to supply non-theoretical bases instead of premises as reasons for Aristotle's explicit conclusions. So, to postulate a tacit prejudice against women or non-Greeks as a reason for Aristotle's conclusions resolves the problem in advance so that no aporia can arise. Alternatively, the present essay assumes that the inconsistencies in the aforementioned aporiai result from genuine theoretical commitments and deserve explanations of the same kind.

In addition to the consideration of the passages about deliberative capacities, which are clearly pivotal, two other areas seem central. These include the theory of predication involving non-univocal characteristics as developed in Metaphysics [TEXT NOT REPRODUCIBLE IN ASCII.], Z, and H, and the tripartite distinction concerning actuality and potentiality in de Anima II 2, and the related one about natural and rational potentialities in Metaphysics [TEXT NOT REPRODUCIBLE IN ASCII.]. (11) Mentioned previously en passant, the first discussion arises, as is well-known, in regard to being in Metaphysics [TEXT NOT REPRODUCIBLE IN ASCII.] 2 (subsequently, Z 1, H 1) where being is afforded sufficient unity of subject-matter to allow for scientific study. Given that common predicates other than being--such as one, same, nature, medical, healthy, friendship--are thought to exhibit systematic homonymy, the possibility that the predicate human falls into this kind requires consideration. (12) It is to the theory of predication concerning things neither synonymous nor fully homonymous that I now turn. (13)

II The Theory of Related and Core-Related Homonymy

II 1 Overview

The question about how human is predicated admits of various possibilities of synonymy and homonymy, including what we might term simple or restricted synonymy, in contrast to chance or systematic homonymy. Before considering the various possibilities, let me quickly summarize the initial account of synonymy and homonymy from Categories 1. In this chapter, synonymous things must have the same name and account of the shared characteristic (cf. Categories 1a6-7). (14) To use Aristotle's example, a human being and an ox are animal in the same sense in that we apply the term 'animal' and they share the same characteristic in being animal; hence, they are called 'animal' synonymously (cf. 1a6-8). In contrast, homonymous things have the same name but not the same account of the common feature (cf. 1a1-2). Employing his illustration, a human being and a picture of an animal are not animal synonymously inasmuch as their natures differ (cf. 1a1-3). For, while we might use the term 'animal' in both cases, in the one, we refer to a living thing that is animal, and in the other, to a drawing, or likeness, of such a thing.

Departing from the account in Categories 1, we find homonymy being put to good use with regard to things that have related but non-identical natures in texts outside the Organon. In the most well-known of these passages, Metaphysics [TEXT NOT REPRODUCIBLE IN ASCII.] 2, Aristotle employs the theory of homonymy to distinguish the unity afforded by things that exist, or have being. Preferring to keep (in part, anyway) to his denial in Posterior Analytics II 7, 92b14, that being is not a genus and so has no special science, (15) he admits that the term 'being' (to on) is not synonymous. If existing things qua existing do not have the same property said in the same way (kath' hen), they lack a single nature as a group and cannot constitute a genus. But to the claim that being is said in many ways (legetai pollachos), he now adds that its instances are related to one thing (pros hen), substance, and thus being comprises a quasi-genus.

Abstracting from the case of being in [TEXT NOT REPRODUCIBLE IN ASCII.] 2, systematic homonymy is thought to be present among things that have a common term, or predicate, and share some defining characteristic; they do not, however, share precisely the same characteristic or in the same way. Let us consider one of the canonical examples of related homonymy, the medical, from Metaphysics [TEXT NOT REPRODUCIBLE IN ASCII.] 2, 1003b1-4. As he puts it, things that are medical are so called by being related to medicine in a certain way: one thing is a medical instrument, another, a medical operation, and another, a medical practitioner. All are related to a primary case of the medical, the medical art. The items that are so called in virtue of having some relation to the primary thing, medicine, are the secondary items considered to be medical. In this way, the medical and similarly, the healthy exhibit a specific type of systematic homonymy. In sum, for things to be related in this way, they possess the following features: (i) one term is applied to them, (ii) they have related, but non-identical defining characteristics, (iii) they have related, but non-identical definitions.

A more restrictive kind of systematic homonymy that involves the inter-relationships among the relata appears in the cases of being, medical, and healthy. According to the taxonomy of kinds of homonymy, this feature depends on the presence of some kind of inter-dependence among the related items. Specifically, the secondary cases will bear a non-symmetrical kind of relationship to the primary case; the relationship may, for example, be a causal relationship, as in the cases of the medical or the healthy. In such cases, we find that the primary instance is the source of the characteristic for the things that are so called or is the reason that the secondary cases are so designated. To put the thesis as a formal condition, we may state that two things, a and b, are causally core-dependent homonyms of F if and only if: (i) they share the same name, (ii) the definitions of a and b are not identical, and (iii) if a is the primary instance of F, then b's being F stands in a causal relation to a's being F. (16) To return to the case of the medical, for example, medical items meet the requirements for core-related homonymy: the instrument, operation, and doctor are all termed 'medical' in virtue of being related in different ways to the medical art, the primary medical item in virtue of which the secondary things are considered medical.

Armed with these distinctions about systematic homonymy, we may consider whether the term 'human' and the property of being human can be considered to involve homonymy and if so, whether they count as systematic homonymy. The starting point consists in determining if human names a synonymous characteristic, and if not, whether it is something that applies to individuals in a loose or systematic fashion. Additionally, if it should turn out that being human is not a univocal predicate or characteristic, this might provide an account as to why some properties that belong to or follow from the form human could be absent from some humans. This, in turn, might be used to explain some of the anomalies about human beings and their capacities, such as those concerning the capacity for deliberation. But if the predicate 'human' is taken as referring to a homonymous characteristic, certain difficulties follow, such as having to deny the intuitively plausible idea that 'human' refers to a univocal characteristic across all individuals.

II 2 Problems for the Homonymy Thesis

There are clearly some problems facing the hypothesis that human is homonymous. As a first strategy, we might think to employ one of Aristotle's tests for homonymy suggested in Topics I 15, such as detecting ambiguity among cases of contraries, privations, contradictories, comparables or paronymous terms. (17) The difficulty is that none of these will work for the predicate 'human': as we may expect in view of the claim about substance in Categories 5, 'human' has no contrary, privation or contradictory, does not admit of comparison, and paronymy does not make it off the ground. (18) By these results, human resists the standard diagnostic tests for multivocity that snag many other common terms in Topics I 15. So, we return to test the possibility that 'human' is, after all, synonymous. This alternative is initially supported by the claim in Categories 5 that one substance is not more or less what it is than another substance (cf. 3b37-8), and also by the requirement in Posterior Analytics I that common terms signifying per se properties are used synonymously across such predications (cf. 77a5-9, 83a30-5, 85b15-18). For example, if animal is per se to horse, ox, and bird, these possess the same feature qua animal, and the term 'animal' is used synonymously across such predications. If we employ 'human' as we do 'animal', that is, as a term that signifies a per se property of some kind, there is a prima facie case for this use being considered synonymous. A further confirmation is that throughout many discussions about human beings and human capacities, Aristotle never mentions the idea that 'human' is homonymous rather than synonymous. Where he considers examples of systematic or core-related homonymy, such as the central books of Metaphysics, he never finds 'human' to be homonymous nor does he mention it in conjunction with homonymous instances in which the relation of shared characteristics is harder to discern, as with nature (physis) or friendship (philia). (19) The absence of any explicit claim that human is homonymous, taken together with the notion that predicates signifying per se properties are used synonymously and that substances do not admit of degree present considerable difficulty for the alternative involving homonymy.

In fact, it is more natural to assume that human is taken to be synonymous, given that a name that can be applied with the same sense across individuals of a kind, or genos, can be used in a precise sense. The lack of textual evidence that the term 'human' has different senses or is highly variable in everyday usage (excluding the metaphorical kind) when applied to human beings seems to show it functions as an ordinary natural kind term. In this function, it must have a unified sense and signify a synonymous property among human beings, in the same way that a common species-name such as 'horse' or 'dog' does. But perhaps this inference is too easy. After all, other common predicates that seem to indicate one characteristic and to be used synonymously do not signify such on further inspection. (20) In what follows, I present some evidence that seems to indicate homonymy arises as part of being human, and so the possibility of 'human' being non-synonymous remains open, at least at this juncture.

III Homonymy, Synonymy, and Human: Deliberation as Human Potentiality

The main argument supporting 'human' as homonymous and against it as synonymous stems from Politics I where Aristotle denies that the deliberative capacity (to bouleutikon) is present in all humans or present in the same way. Specifically, he claims that the deliberative faculty is present and operative in male citizens, present but inoperative (akuron) in female citizens and not present at all (ouk echei to bouleutikon) in slaves (Politics 1260a10-14). So, in regard to levels of deliberative potentiality, adult male citizens have the most complete, or fullest, deliberative ability and natural slaves (men and women) have the least complete ability (i.e., non-existent), with women citizens and immature male citizens falling somewhere in between these two extremes. Accordingly, Politics 1260a12-14 implies that human deliberative capacity is non-universal, as justified by the following ranking: immature male citizens possess deliberative capacity but in an incomplete (ateles) way, women, in an inoperative (akuron) way, and slaves simply do not possess it (ouk echein). Taken in this way, the various levels of deliberative capacity at Politics I 13 (1260a12-14) appear to be pivotal in assessing the lack of synonymy in being human. It would seem that if someone lacks deliberative ability, then he or she lacks reason or some constitutive part of reason, given that the two capacities are connected in definition and being. As we know from texts throughout de Anima III, as well as from Nicomachean Ethics VI, the deliberative capacity is found in the rational part of the soul, and both works consistently maintain that only rational animals deliberate. (21) On the assumption that deliberation belongs to reason, the absence of one implies that of the other. So, someone who lacks deliberative ability lacks reason as well, and this contradicts the linkage between being human and being rational. If this inference is correct, 'human' cannot be considered synonymous for all individuals: it does not refer to the same characteristic or in the same sense when applied to male citizens, female citizens, and natural slaves, for example.

However, we might propose an alternative to the previous by restricting the predicate to smaller groups: for example, if we apply 'human' specifically to mature, male, Athenian citizens, it becomes synonymous. This alternative preserves synonymy through restricting the range of application of the term, and will be called RRS ('restrictive range synonymy') in what follows. One disadvantage of RRS is that 'human' becomes multi-vocal if we consider it as a global predicate taken across all humans: the synonymy of 'human' collapses into homonymy when it is being taken as a predicate at a higher level of abstraction.

It may be suggested, as an alternative to RRS, that deliberation and reason are not precisely co-extensive capacities, and this raises the possibility that someone can have reason, not deliberation, and be considered human. We might refer to this as the 'de-linking' strategy, and this, it may be proposed, provides a solution to the aporia that was mentioned above in the following way. According to the de-linking proposal, human beings as such would possess a general rational capacity in virtue of their human form, and in addition, a subset of humans would possess a more specific kind of rational capacity, namely, deliberative capacity. On this view, possessing a rational faculty simpliciter is sufficient for a living thing to be considered human, and so, having a rational capacity does not strictly imply a deliberative capacity. Hence, it seems that on the first cut 'human' is being applied synonymously across all human beings. This proposal has a distinct conceptual advantage over RRS by preserving the synonymy of being human at the broadest level, and it has, as well, support from crucial passages in Politics, such as that in I 13, concerning the restriction of full deliberative capacity to adult, male citizens. After all, it would be implausible to read passages such as Politics I 13, 1260a10-14, as implying that free women, natural slaves, and children are not human because they do not deliberate properly or at all. In fact, Aristotle rejects the view of those who think slaves lack reason (logos) and deserve only admonition (cf. Pol 1260b5-6), indicating his disagreement with those who think that slaves are lacking in rational capacity altogether.

Weighing against the de-linking strategy just proposed is other evidence to the effect that Aristotle thinks of deliberative capacity as belonging to human beings as such, as being part of the essence of being human. If so, it would be impossible for someone to have reason and be unable to deliberate, that is, to be unable to deliberate at some time or other. (22) So, if being able to deliberate is essentially connected to being human, and if deliberative ability is denied to some humans, the term 'human' cannot be used synonymously across human beings. It would appear, then, de-linking, the option of disconnecting the rational from the deliberative capacity does not have the desired result of preserving the synonymy of human with reference to men, women, and slaves taken together. For it will turn out that the sole application of 'human' that preserves synonymy is that application to adult male citizens able to deliberate fully.

To review at this point, the alternatives for human being synonymous include: (i) restricted range synonymy, or RRS, the option that preserves the synonymy of human by restricting the range of application by class or group, and (ii) de-linking, the option that separates the rational and deliberative capacities so as to find only the former necessary to being human. On the first option, human is synonymous within each group taken as such (viz., only free men or only free women or only slaves), but not as a global common predicate; on the second option, human is synonymous and refers to the same capacities when applied to rational animals as such. What arises as a concern is that, on either alternative, human becomes non-synonymous at a higher level, although not in the same way. According to RRS, human becomes non-synonymous as a global predicate, although synonymous in each restricted application; in contrast, according to de-linking, human is synonymous when referring to being rational simply, but non-synonymous across different groups. On the assumption of de-linking, the use of human applied to adult male citizens, adult female citizens, and slaves is homonymous since the deliberative capacity is not present in the same way among them. Since the characteristic of rational-plus-deliberative capacity is lacking as a common characteristic, human becomes homonymous. (23) But since there is good reason to think that for Aristotle human beings share a common nature which encompasses rational and deliberative capacities, the de-linking option appears to lose favor. If we exclude accidental homonymy, we seem to be led to the conclusion that human is homonymous unless another proposal for synonymy can be offered. Since the case for human being homonymous presents various initial difficulties, another line of argument for preserving synonymy seems warranted.

A third option concerning being human preserves synonymy by enlarging the notion of what deliberative activity is. On this alternative, being human encompasses rational and deliberative capacities, but allows for some plasticity in the attribution of deliberation. Briefly put, this option consists in: (i) predicating rationality of all humans, (ii) preserving the link between rationality and deliberation, and (iii) allowing for different activities to count as deliberation. The distinctive feature of this option depends on deliberation being considered as referring to different range of activities, at one end, a general means-ends reasoning, and at the other, a highly specialized kind of practical reasoning involving a view of the good. This possibility will be referred to as 'dual deliberation synonymy,' or DDS. On this view, when Aristotle claims male citizens deliberate while female citizens, children, and slaves do not (cf. Pol 1260a12-14), he means that male citizens deliberate in some more complete way than women or children do. On this interpretation, it is possible for a human being to deliberate in one way, but not necessarily in another more specialized way. The present alternative is attractive in that it would allow us to rest with the natural assumption that when Aristotle claims that humans have a capacity to deliberate about future alternatives in Nicomachean Ethics III and de Anima III, he means that humans as such are the kind of animal that has a capacity for de liberation and practical reasoning. (24) In this case, being human ensures that the animal is capable of deliberating in a general sense, although not necessarily in a narrow, or specialized, sense. In addition, it allows that from being human and having rational capacity, it follows that any subject has a deliberative capacity. More to the point, since it allows deliberative capacity to be present among all humans, the term 'human' remains synonymous as referring to a common complex characteristic, that of having rationality plus deliberation. Furthermore, a potential objection against this option, namely, that human becomes homonymous by the homonymy introduced by deliberation filtering upwards, does not find a firm footing. (25) As will be developed presently, the ways of deliberating share basic, common features; it does not appear either that these cases involve accidental homonymy or that the conditions for core-dependent causal homonymy are in place, as in the medical and the healthy. (26)

Let us recapitulate the alternatives raised in the previous discussion. At present, there are four alternatives of how to consider human with regard to homonymy or synonymy, including: (i) restricted range synonymy, or RRS, according to which human is synonymous in each special sphere of application (though not globally); (ii) de-linking, according to which human is synonymous as a generic property, but homonymous if considered as referring to reason plus deliberation; (iii) dual deliberation synonymy, or DDS, according to which human is synonymous given the assumption that deliberation admits of different kinds; (iv) related homonymy, RH, as a possibility if the first three options fail. As has been observed with regard to the first three, the option involving de-linking (the second option) has the result that the sole group capable of deliberation is that comprised of adult, male citizens. One rather unhappy consequence here concerns that this seems to imply that human signifies a generic property (like animal), and so, the various groups of humans seem to be considered sub-kinds, or species, of humans. In this regard, the option involving de-linking, while initially plausible, will have the following undesirable consequences.

Let us suppose for the moment that being human is a genus; if so, there will be species answering to the several groups mentioned, adult male citizens, female citizens, children, slaves, and so on. Now while this alternative might be used to explain the reason for which Aristotle finds demonstrable differences in deliberative capacities among humans in Politics I, it would not be consistent with other claims he makes. In fact, the counter-evidence for supposing Aristotle postulates more than one species of humans is two-fold. First, like Plato, Aristotle rejects the notion that human differences, e.g., cultural, geographical, or sexual differences, are sufficient to account for real, biological species differences. (27) For example, in weighing the political capacities of non-Greeks, or barbarians, where Aristotle follows the Hippocratic medical tradition in citing climate differences for differences in dispositions, he does not infer species-differences as a result. (28) So, the difference in deliberative capacities cannot map directly onto a specific difference among human kinds. Also, as we have noted, Aristotle's claim about the non-comparability of substance in Categories 5 matters. (29) Aristotle holds that things identical in substance do not differ in degree: 'for example, the same substance, human, cannot be more or less human, than itself or than another human, just as a pale thing is more [pale] than another or a beautiful thing more [beautiful] than another' (Cat 3b37-4a1). In contrast to qualitative ascriptions like color, 'human' does not admit of comparison or degree: one human is not more human than another. In this regard, it cannot be the case that the free male citizen is more human than the female, the child or the non-citizen. So the hypothesis that human groups exist as ranked kinds within a genus conflicts with Aristotle's claim that 'human' signifies an infima species, not a genus.

It seems justified to pass to option (iii) dual deliberation synonymy, or DDS, the option involving differences in the ways of deliberating. I have presented the option as one that preserves human as synonymous, but it might be argued that its synonymy is compromised inasmuch as deliberation invites homonymy by allowing for qualitatively different kinds of deliberating. First, we need to ascertain whether the texts support one kind, different kinds, or different ways, of deliberating. It might be argued, in fact, that Aristotle is using deliberation in a specific, narrow sense in Politics I 13 when he denies deliberation to free women and slaves. According to this view, he is using the term in the sense of signifying the kind of rational, calculative planning dependent on having a conception of the good, and this, it is argued, is available only to those who possess full moral virtue. (30) This reading would restrict genuine deliberation to the few citizens in the best regimes, and would likely turn out to be a capacity present as a first level potentiality only in adult, male, Athenian citizens. (31) It has already been observed, however, that there are certain problems in arguing this position, not the least of which is that it implies deliberation is rarely engaged in, and only by a few. Looking across his many discussions about deliberation in Nicomachean Ethics, such as II 2, II 5, and III 3, where he discusses deliberation as being a part of moral excellence, we do not find the narrow reading of deliberation being borne out. Nor do we find that Aristotle restricts genuine deliberation to the few in, for example, Nicomachean Ethics VI where he reflects on it in relation to the specification of phronesis as practical deliberation implied by moral virtue nor, again, in Nicomachean Ethics VII in relation to akrasia and the ways in which we may fail to deliberate properly. On the contrary, we find that in his central discussions of deliberation across Nicomachean Ethics, the activity is described as a common albeit distinctively human capacity, natural to humans in virtue of being the kind of animals who are able to reason about the good in relation to the future. In this regard, deliberation is a universal, human capacity for rational calculation concerning means and relative ends. The discussions about deliberation in Nicomachean Ethics do not, then, state that antecedent to practical reasoning we must have a full, correct conception of the human end, or good, in order to deliberate.

My present suggestion is to modify DDS, the dual deliberation view; instead of thinking there is one way of deliberating properly, let us suppose that are different ways of partaking in the same activity, one way with, and one without, a full conception of the good. If there are different ways in which humans can partake in the same activity, we need not assume that all kinds of people have to perform the same activity; we would need to see which kind seems more appropriate in the context of the discussion. In the passages about deliberation and moral virtue in Nicomachean Ethics II, III, VI, and VII, he appears to be referring to a type of means-ends reasoning about the day-to-day sort of good. This kind of deliberation is operable apart from having a complete conception of the human good, which only the fully virtuous possess. This garden-variety activity of deliberation is, I take it, the basic practical ability available to us by nature as we develop from children to adults. There is, of course, another activity of deliberation that is related to phronesis in a special sense, as being the best kind of practical reasoning allowable to virtuous, Athenian, male citizens. On the present reading, all humans, men and women, free and slave, engage in the everyday type of deliberation, but only a virtuous few engage in the specialized activity of deliberation in the narrow sense.

The dual reading of deliberation affords us the benefit of allowing Aristotle more consistency than we first considered. In addition, it holds the key to unlocking both the aporiai posed at the outset of the essay. The first aporia was, in effect, the result of combining the thesis about metaphysical equality following from Aristotle's claims about substance in Categories and Metaphysics with those about deliberative capacities among humans from Politics. What we have found is that Aristotle is able to maintain both sets of claims. On the one hand, humans belong to the same species or possess the same species-form, and so they possess the same essential properties. Assuming Aristotle holds, as is generally accepted, that rational capacity is an essential human property and that this capacity entails the capacity for rational calculation called deliberation, it follows that humans as such have deliberative capacity. What has been added as a qualification to the previous claim is that the generic kind of deliberation that humans possess by virtue of their species-form, namely, the everyday deliberative capacity, does not require having a complete conception of the human good. Rather, having a complete conception of the human good is only required to be able to engage in deliberative activity of the highest kind. It is, I think, from this panoptic perspective about practical reason and deliberation that in Politics I and VII Aristotle draws his conclusions about different political groups having or lacking a deliberative capacity. Thus, in Politics I 13, 1260a10-14, he finds that all humans have the same parts of the soul, but free women, children, and slaves do not have it or possess it fully, and in Politics VII 7, 1327b20-35, he claims that European and Asian 'barbarians' also lack deliberative ability (although for different reasons). (32) If we grant that he has different standards of what it means to deliberate, he may be able to claim without contradiction that virtuous male citizens alone deliberate and that other, less virtuous humans, slave or free, also deliberate. For it would be true to say that lacking a full conception of the good would imply that someone could not deliberate in the special sense but not imply that he or she could not deliberate in any sense. For purposes of analogy, we may consider the lesser, or 'second-best' sense of deliberation to the second-best good activity that Aristotle describes in Nicomachean Ethics X 7 and 8, 1177b29-8a14 where he seeks to compare the qualities of moral virtue (ethike arete) with those of study (theoria). The activity of moral virtue seems to pale in comparison with that of study precisely because study better fulfills the conditions Aristotle specified for eudaimonia in Nicomachean Ethics I. (33) Yet it hardly follows from this claim that moral virtue is not a eudaimon, good activity. In a similar way, I suggest that the everyday kind of practical reasoning that less than fully virtuous humans display does not fail to involve deliberation. From a wider perspective about deliberation and practical reason, it seems reasonable that free women, slaves, barbarians and perhaps children possess a basic capacity for deliberation: they actualize it in the sense of making decisions about their future ends and projects in spite of lacking a complete notion of the good that virtuous, male Athenians have.

Unifying the activities among the ways in which we deliberate suggests a solution to the second aporia posed at the outset of the paper. This problem concerned the issue of deliberation with Aristotle's teaching about moral habituation, the claim which seems to be extended to all human beings in Nicomachean Ethics II. What I suggest is that the capacity for deliberation exists as a natural human capacity in all humans (as a first level potentiality), but it is one that, like moral virtue, must be subjected to training by habituation in order to become effective as a hexis (a second level potentiality). In this discussion, I am assuming that the relation between the two potentialities is fully natural, or part of the thing's nature, even if one of the potentialities requires training; so the relation is not one of natural to artificial potentiality. (34)

IV Reason and Potentialities in de Anima II

The tri-partite distinction among levels of actuality and potentiality in the soul as sketched in de Anima II 1 and developed in II 5 may afford a tool for understanding the nature of the differences in deliberative capacity. Placing the tri-partite schema in de Anima II 5 against Aristotle's comments from Nicomachean Ethics II 1 about the nature and acquisition of moral virtue yields the following conclusions. (35) From Aristotle's discussion of capacities and actualities in de Anima II 5, 417a21-b2, we may say that in virtue of their kind, humans possess a first level dunamis for moral excellence (or vice) when they are born in the same way that they possess a first level dunamis for, say, knowledge of grammar. Following a period of training and habituation, some humans develop a second level dunamis, or hexis, with regard to moral excellence (while others do not). Once in possession of the second level potentiality for moral virtue, the moral agent, like the one schooled in knowledge of grammar, is able to exercise the capacity, as long as nothing impedes the exercise. It seems reasonable to extend this sequence of change with regard to the capacity for deliberation as well, especially since the two are related. For deliberation (boulesis) is essentially related to moral virtue (ethike arete) as a constituent element, and so, it is fully reasonable that for the initial capacity of deliberation to become a hexis, it must undergo a similar kind of habituation as that for moral virtue.

What presents a problem concerns the precise nature of the change and the kind of difference existing between the first and second potentiality in deliberative capacity. On the one hand, the change from first level potentiality--from what we may call a 'bare' dunamis (36)--to a second level potentiality, the habituated state, involves a qualitative change, or alloiosis, in the affective part of the soul (cf. de An 417a31 2). (37) What Aristotle explicitly states at de Anima II 5, 417a31-2 is that the first change in potentiality-state in the soul, that from the bare dunamis to the hexis, comes about '... by a qualitative change through learning (mathesis) and often after changes from a reverse condition' (enantias hexeos). The change from first to second actuality, in contrast, does not involve the same kind of change in that in the latter case, the subject uses the capacity acquired without any change in the capacity itself. Aristotle expands on the two kinds of change involved by way of an explanation of two senses of paschein, 'to suffer,' or 'be affected': one sense indicates the destruction of one thing by its contrary, and another sense, the preservation of something in potentiality by something in actuality and like it (cf. 417b2-5). These two states are intended to map onto the two changes previously discussed, such that the kind of paschein involved in the destruction of one thing by its contrary describes the change from the bare dunamis to the hexis, and that involved in the preservation of one thing by another like it, the change from hexis to entelechy.

For the moment, it is the former kind of change that is of interest: what seems clear is that in de Anima II 5, the transition from bare dunamis to hexis involves a process of 'being affected' (paschein) which specifically involves a change in the pathe of the subject. While there is some plasticity in the terminology of pathe, in Categories 8, Aristotle distinguishes pathe as the third type of qualities (poiotetes), those which belong to a subject in virtue of being so qualified (cf. Cat 9a28-33). This class of qualities includes properties that belong to bodies, such as being sweet or being hot (cf. Cat 9a33-4), and also those qualities that belong to soul, such as being angry (cf. Cat 9b33-5). (38) In brief, pathe are the qualifications of body or soul that may be considered either as passive affections (the results of some process of affection) or as active affections (the bases for effecting changes in another) as heat affects a change in temperature in another thing.

With regard to how affections, or pathe, relate to the capacity for deliberation and moral excellence, the suggestion is that the first level potentiality (dunamis) for moral virtue is in fact a pathos of the subject and so, subject to change. For, we know that affections, after having undergone many changes from one contrary to another over time themselves change from being transitory affections to more permanent states in the same way as an unformed dunamis becomes a hexis. This change in the mode of how the subject is qualified helps to explain why the first actuality, or second potentiality, is a dunamis in another sense--as the thing's ability 'to exhibit what it really is' or to use what it has, rather than an ability to be changed in pathe. (39)

The capacity for deliberation exists in relation to that for moral excellence, and so exhibits a parallel transition in the affections underlying the change from bare dunamis to hexis. Whatever the basic contrary tendencies underlying deliberation are, for example, indecisiveness or rashness, on the one hand, and lack of concern or excessive feeling, on the other, such feelings have to be brought to a proper, permanent state of balance so that they are not fleeting states, as pathe are. What this may imply as an explanation for the differences in deliberative capacities among human beings is that the affections underlying the bare dunamis in some people are less malleable than those in others; if so, the transition to the second level potentiality would be diminished or blocked dependent upon some initial natural tendency. The failure of the underlying pathe to be affected in the right way as to be transformed into the qualities underlying a second level ability may, in fact, ultimately be due to a complex of innate and external causes, but this discussion would take us outside the confines of the present essay. (40)

V Conclusions

The position that best resists refutation stands on the side of human being considered to be synonymous. To sum up, we have seen that arguing that human is homonymous using the de-linking strategy on the grounds that some humans are said to have reason and deliberate while others do not (cf. Pol 1260a10-14) has deficits. The de-linking alternative fails to make sense of many passages in Nicomachean Ethics and de Anima about deliberation being a human capacity and specifically, one needed for choice, practical reason, and local motion. In contrast, holding human to be synonymous is made on various, uncontested grounds, including that: (i) substances cannot be compared, or do not differ in degree, (ii) being rational is taken as a common, per se property of every human, (iii) deliberation is a widely distributed human capacity following from being rational, (iv) slaves, too, are said to possess the rational part of the soul (to logistikon). But there remains the issue of whether we are speaking about a restrictive or a global kind of synonymy for the predicate human. According to RRS, or restrictive range synonymy, we have human applying synonymously to each separate group, but the view threatens to lead to homonymy if the predicate is applied to the whole class. For, homonymy in regard to deliberation seems to result from the following claims: (i) deliberative capacity is absent in some humans, (ii) rational and deliberative capacities are separable and not mutually implicative; (iii) being of two kinds, there is no common element across deliberation. As has been observed, finding deliberation to be homonymous is unwelcome as it threatens to filter upwards to human. However, by revising the initial interpretation of DDS, the upwards filtration is not necessary in that we preserve the synonymy of deliberation itself on this interpretation. As was suggested, it is plausible to argue that deliberation is a human activity that can be present in different ways, all of which employ rational activity involving practical reason, just as the activities of study and moral excellence may be considered to count as different ways of exercising human goodness and flourishing. Consequently, the difficult objection that human is not synonymous but homonymous based on the result of deliberation being homonymous is not warranted; the conclusion that human is synonymous remains intact and alive.

References

Achtenberg, Deborah. 1989. 'The Role of the Ergon Argument in Aristotle's Nicomachean Ethics'. Ancient Philosophy 9: 37-47.

Annas, Julia. 1999. 'Aristotle on Human Nature and Political Virtue'. In Lloyd Gerson, ed., Aristotle: Critical Assessments, Vol IV, 48-66. Boston: Blackwell.

Code, Alan. 1984. 'The Aporematic Approach to Primary Being in Metaphysics Z'. In F. J. Pelletier and J. King-Farlow, eds., New Essays on Aristotle. Canadian Journal of Philosophy, Suppl Vol X, 1-20. Guelph, ON: Canadian Association for Publishing in Philosophy.

--. 1985. 'On the Origins of Some Aristotelian Theses about Predication'. In J. Bogen, ed., How Things Are, 101-32. Boston: Reidel.

Cohen, Sheldon M. 1996. Aristotle on Nature and Incomplete Substance. New York: Cambridge University Press.

Cole, Eve Browning. 1994. 'Women, Slaves, and "Love of Toil" in Aristotle's Moral Philosophy'. In B. Bar On, ed., Engendering Origins: Critical Feminist Readings in Plato and Aristotle, 127-44. Albany: State University of New York Press.

Cook, Kathleen. 1996. 'Sexual Inequality in Aristotle's Theories of Reproduction and Inheritance'. In J. Ward, ed., Feminism and Ancient Philosophy, 51-67. New York: Routledge.

Cooper, John. 1975. Reason and the Human Good in Aristotle. Cambridge, MA: Harvard University Press.

Driscoll, John. 1981. 'Eide in Aristotle's Earlier and Later Theories of Substance'. In D. O'Meara, ed., Studies in Aristotle. 129-159. Studies in Philosopy and the History of Philosophy, Vol IX. Washington, DC: Catholic University of America Press.

Fortenbaugh, W. W. 1977. 'Aristotle on Slaves and Women'. In J. Barnes, M. Schofield and R. Sorabji, eds., Articles on Aristotle, Vol II, 135-139. New York: St. Martin's Press.

Gill, Mary Louise. 1989. Aristotle on Substance. Princeton, NJ: Princeton University Press. Homiak, Marcia. 1996. 'Feminism and Aristotle's Rational Ideal'. In J. Ward., ed., Feminism and Ancient Philosophy, 118-137. New York: Routledge.

Irwin, Terence. 1981. 'Homonymy in Aristotle'. Review of Metaphysics 34: 523-544.

Kamtekar, Rachana. 2002. 'Distinction Without a Difference? Race and Genos in Plato'. In J. Ward and T. Lott, eds., Philosophers on Race: Critical Essays, 1-13. Boston: Blackwell.

Kraut, Richard. 2002. Aristotle's Political Philosophy. Oxford: Oxford University Press.

--. 1989. Aristotle on the Human Good. Princeton, NJ: Princeton University Press.

Lloyd. G. E. R. 1978. The Hippocratic Writings. Ed. Harmondsworth: Penguin.

Modrak, Deborah. 1994. 'Aristotle: Women, Deliberation, and Nature'. In B. Bar On, ed., Engendering Origins: Critical Feminist Readings in Plato and Aristotle, 207-222. Albany: State University of New York Press.

Roche, Timothy. 1988. 'Ergon and Eudaimonia in Nicomachean Ethics I: Reconsidering the Intellectualist Interpretation'. Journal of the History of Philosophy 26: 175-194.

Schofield, Malcolm. 2006. 'Ideology and Philosophy in Aristotle's Theory of Natural Slavery'. In R. Kraut and S. Skultety, eds., Aristotle's Politics: Critical Essays, 91-119. Lanham, MD: Rowman & Littlefield.

Shields, Christopher. 1999. Order and Multiplicity: Homonymy in the Philosophy of Aristotle. Oxford: Clarendon Press.

Smith, Nicholas. 1983. 'Plato and Aristotle on the Nature of Women'. Journal of the History of Philosophy 21: 467-78.

Tress, Daryl. 1996. 'The Metaphysical Science of Aristotle's Generation of Animals and its Feminist Critics'. In J. Ward., ed., Feminism and Ancient Philosophy, 31-50. New York: Routledge.

Ward, Julie K. 2008. Aristotle on Homonymy: Dialectic and Science. New York: Cambridge University Press.

--. 2005. 'Aristotle on Physis: Human Nature in the Ethics and Politics'. Polis: Journal of the Society of Greek Political Thought 22: 287-308.

--. 2002. 'Ethnos in The Politics: Aristotle and Race'. In J. Ward and T. Lott, eds., Philosophers on Race: Critical Essays, 14-37. Boston: Blackwell Publishing.

Whiting, Jennifer. 1992. 'Living Bodies'. In M. Nussbaum and A. Rorty, eds., Essays on Aristotle's De Anima, 75-92. Oxford: Oxford University Press.

(1) Aristotle typically uses a word to refer either to the characteristic signified by the word or to the term itself (cf. Categories 1a9, where both ox and human are called 'animal' and are animal); where the context requires a distinction, I use the word in quotation marks to refer to the term and the italicized word to refer to the property signified.

(2) Translations are my own.

(3) The term 'non-scalar' I owe to Chris Shields, my commentator on an earlier version of the paper that was presented at a conference entitled Aristotle and Life held at University of Alaska, Anchorage, August 7-10, 2007. In addition to benefiting from Shields' comments, I would like to acknowledge those offered by the other conference participants, and especially to thank John Mouracade for his dual role as conference organizer and gracious host.

(4) For the familiar tri-partite list, see, for example, Metaph Z 3, 1028a2-5; a chief difficulty in these books of Metaphysics arises from the fact that eidos, rather than ousia, turns out to be so highly variable in its applications, as Paul Studtman's paper on eidos in this volume makes clear.

(5) I am taking substantial form as picking out the shared characteristics of the specific kind, typically, the functional capacities of the thing (if living, taken as the composite). For related discussions on the nature of substance and form, see Code 1984, 1985, Driscoll 1981.

(6) The inference here needs expansion; I note the difficulty because the implication involves capacities or dispositional properties, not simple properties.

(7) Chance homonyms are things having only the name in common and no overlapping feature, as for example, at EN 1096b26-7, where Aristotle contrasts the good with things that are homonymous by chance.

(8) See section II in brief for kinds of homonymy; for full discussion, see Shields 1999, 9-41, and also Ward 2008, 77-102.

(9) See Schofield 2006, 94-7.

(10) See, for example, the discussions by Cole 1994, Cook 1996, Fortenbaugh 1977, Homiak 1996, Modrak 1994, Smith 1983, to name but a few.

(11) The account of potentiality (dunamis) in DA II 5 may be amplified by others, such as in Metaph [TEXT NOT REPRODUCIBLE IN ASCII.] 2 and 5 on rational and non-rational natural capacities, as well as the entry on dunamis in Metaph [TEXT NOT REPRODUCIBLE IN ASCII.] 12.

(12) For discussion of these related homonyms, see Shields 1999, chs. 5, 6, 9 for analysis of being, one, and body, and see Ward 2008, chs. 4, 5, on being, nature, and friendship.

(13) The distinction between complete and related homonymy depends on the presence in the latter kind of certain shared, but non-identical, common features and overlapping definitions. For discussion on the foundational account of homonymy in Cat 1, see Irwin 1981, Shields 1999, 9-12, Ward 2008, 12-18.

(14) In Categories 1, Aristotle defines the term 'animal' as synonymous or homonymous by reference to the characteristic that he takes to be signified by the term, so synonymy and homonymy are categories spanning words and real characteristics signified by words.

(15) For a similar text, see EE I 7, 1217b25-35, where he reasons that since the good is said in as many ways as being (invoking the categorical account of being), there is no science of either one.

(16) This formal account is based on the definition of 'four-causal core primacy' given by Shields 1999, 118-19, although I am not presupposing a specifically four-causal mode of dependence in this discussion.

(17) For fuller discussion of these strategies, see Ward 2008, 56-75.

(18) In the sense that for paronymous terms, the central term must be non-synonymous to begin with, as is the case with 'health' (on which, see Top 106b29-7a2).

(19) For discussion of the various senses and unity of nature (physis), see Ph II 1, 192b21 ff.; Metaph V 4; for the parallel one on friendship (philia), see EE VII 2, 1236a15-32, and also, Ward 2008, 137-67.

(20) See Aristotle's comment with regard to justice in EN V 1, where he states that sometimes different uses of a term are closely related and so the homonymy escapes our notice (cf. 1129a26-7); similarly, at Ph VII 4 concerning the kinds of motion where he finds that some cases of homonymy are closely related and so, less evident, while others are farther apart and more evident (cf. Ph 241a21-5).

(21) For example, we know from EN III 3 that deliberation relates to choice (prohairesis) which is present only in rational animals; again, from EN VI 5 (cf. 1140a25-b6) that deliberation is bound up with practical reason (phronesis); furthermore, in de An III 3 (428a16-24) and III 7 (432b2) we find deliberation working with a rational sort of imagination (phantasia bouleutike) open only to humans.

(22) For, clearly substances can have essential properties and yet lack them at certain times: so, water is essentially liquid and yet becomes ice; children have reason and yet are unable to exercise the ability until maturity; as Cohen points out, Aristotle's essentialism involves dispositional per se properties: see Cohen 1996, 51-2.

(23) On this alternative, humans share central common characteristics but the one signified by 'reason plus deliberation' would be absent as a common feature; since there would be no shared feature, accidental homonymy, or 'by chance' (cf. EN 1096b27) would seem to result.

(24) Textual passages are numerous; see, for example, the discussion of deliberation: relating to choice (prohairesis) in EN III 3; relating to practical reason (phronesis) in EN VI 5 (1140a25-b6); relating to a reasoned kind of imagination (phantasia) in de An III 3 (428a16-24) and de An III 7 (432b2-12); relating to thought and movement in de An III 10-11, esp. 433a9-16, 433a31-b4, 433b28-9, 434a10-12.

(25) The 'upward contamination' objection arises from a line of criticism offered to me from Chris Shields in regard to an earlier version of the paper.

(26) The feasibility of establishing one case of deliberation as causally prior to the others in a systematic fashion as shown in the medical or the healthy seems slight; we might suggest the relation of the deliberative capacity of the immature male (ateles) to mature male citizen as related by final cause, but the remaining cases of deliberative capacity (women, slaves) defy similar treatment.

(27) See Plato, Plt 262d, where the Stranger points out that the notion of dividing human beings into two kinds, Greeks and barbarians, is incorrect insofar as the division does not mark off human differences according to natural kinds, or gene. For discussion on Plato's use of genos in regard to divisions among humans, see Kamtekar 2002.

(28) See Pol VII 7, 1327b20-30, where he gives a version of the climate theory of human difference based on the Hippocratic treatise, Airs, Waters, Places, chs. 12-24, on which see Lloyd 1978. For discussion of Aristotle's account of climate theory in Pol VII 7 in regard to slaves and barbarians, see Ward 2002, 20-3.

(29) Namely, where he claims 'No substance admits of a more and a less' (Cat 3b33).

(30) I take it this is Kraut's view on deliberation as it relates to the natural slave's ability (or lack of ability) in Pol I, for which see Kraut 2002, 289-304.

(31) I discuss development of the deliberative capacity in section 4; I take it that the natural capacity is not first present as a hexis, or second level potentiality, like sight, for example.

(32) It is interesting to note that while barbarians are groups from distinct geographical areas, they are defined as such in relation to their political abilities, not to their ethnicity as such; see Ward 2002, 17-23.

(33) Essays on the two kinds of good activities (and lives) are nearly too numerous to mention, but see, for example, Achtenberg 1989, Cooper 1975, Kraut 1989, Roche 1988.

(34) The relation of first to second level potentiality here is distinct both from that of essential to accidental potentialities, on which see Cohen 1996, 53-4, and from natural and artificial ones, as in Whiting 1992, 91-2.

(35) I argue for these results more fully in a paper on physis, see Ward 2005, 294-300.

(36) A similar term is used by Annas in her discussion of moral virtue, see Annas 1999, 50-1.

(37) So, this rules out cases of change to a potentiality that would involve substantial change; for example, earth is not potentially (even as a first level potentiality) a box or a human being, as Aristotle notes in Metaphysics VIII 7, 1049a16-18, because there would have to be several prior substantial changes, as, e.g., earth to wood or earth to seed. For discussion on passive and active potentialities, see Gill 1989, 175-80.

(38) Aristotle distinguishes the latter kind of pathe from psychological states of longer duration such as congenital conditions like being irascible and other conditions that are hard to erase or unchangeable (cf. Cat 9b36-10a4).

(39) For a similar sense, see Cohen 1996, 164, who defines a 'base dunamis' as that which 'marks a thing's ability to become different from what it is or change something else'.

(40) Thinking about the lack attributed to women and slaves, we might speculate about which cause is implicated; in regard to women's lack, the cause may be single or a complex: see for example, Cook 1996, Tress 1996; in regard to slaves and barbarians, see Ward 2002.