2.1 In this chapter we offer arguments for the aforesaid fourfold division of order and for the interrelations that exist between essentially ordered terms.

2.2 When the venerable doctor Augustine, writing about your triune self, declared (in the first book On the Trinity): "Nothing whatever begets itself," you, O Lord our God, were his infallible teacher. Have you not impressed upon us with equal certitude this similar truth? (First conclusion) Nothing whatever is essentially ordered to itself. 2.3 For what is more impossible in an order of eminence than that one and the same thing be essentially greater than itself. As for the other six orders, if dependence be taken in the sense defined above, is there any greater impossibility than that one and the same thing depend essentially upon itself? that it exist without itself? 2.4 This too is in accord with truth: (Second conclusion) In any essential order a circle is impossible. 2.5 For if anything precedes the prior, it also precedes the posterior. Deny this second conclusion and you must admit the opposite of the first. Besides, the same thing will be essentially prior and posterior to one and the same thing, and so be both more perfect and less perfect or be dependent and independent of the latter, which is anything but true. In the first book of the Posterior Analytics Aristotle excluded this circle from demonstrations and it is no less possible in the order of reality. 2.6 As I shall make use of it later, I next present a third conclusion, which like the second is proved from, and sufficiently contained in, the first. (Third conclusion) What is not subsequent to the prior is not subsequent to the posterior. 2.7 This is entailed by what we have affirmed earlier. From this it follows also that whatever does not depend upon the prior does not depend upon the posterior; and further, what is not caused by the prior cause is not caused by the posterior cause, for the latter in the very act of producing its effect depends upon the causality of the prior cause. 2.8 Under your guidance, O God, we shall compare the aforesaid six orders with one another beginning with the four orders of cause to what is caused. However, I shall not discuss their various differences nor the adequacy of the division. since both seem to be sufficiently known. For such a discussion could become prolix, and besides it is not necessary for our purpose. In six conclusions I shall compare the aforesaid orders but only as regards their overlap or logical ties on the side of what is caused. 2.9 (Fourth conclusion) What is not ordered to an end is not an effect. 2.10 The first proof is this. There is no effect which does not stem from some proper efficient cause; if something is not ordered to an end, it does not originate with a proper efficient cause; therefore, etc. The major is proved as follows. In no type [of causality] is the incidental first. Aristotle adequately expresses this in the second book of the Physics where he says that intelligence and nature as proper causes are necessarily prior to the incidental causes of spontaneity and chance. But what does not depend upon what comes first does not depend upon what is posterior (from the third conclusion above). I am referring to positive things which alone are properly capable of being effected. The major then is evident. The minor is proved thus: Every proper agent acts for the sake of an end, for it does nothing in vain. Aristotle settles this point as regards nature where it is even less apparent [than it is as regards an intelligent cause]. Consequently, no proper cause effects anything save for the sake of an end.

2.11 A second proof for the main conclusion is this. The end is the first cause in causing, wherefore Avicenna calls it the cause of causes. Reason confirms this, for the end moves metaphorically in so far as it is loved, and this is why the efficient cause gives form to the matter. But it is not in virtue of some other cause's causing that the end as loved moves [the efficient cause]. The end then is essentially the first cause in causing. 2.12 Another proof is this. Aristotle in the fifth book of his Metaphysics shows that the end is a cause for it represents an answer to the question: Because of what? which question calls for a cause. Now since the end provides the first answer, it will be the first cause. That it really does so is evident, for if we ask: Why does something produce an effect? the answer is: Because it loves or intends the end. But if we ask: Why does it love or intend the end? it is no answer to say: Because it produces the effect. 2.13 From the primacy of the end, proved in this threefold way, our main conclusion follows. For according to the third conclusion above, if something is not caused by the prior cause then neither is it caused by the posterior [2.7]. 2.14 (Fifth conclusion) What is not an effect is not ordered to an end. 2.15 The proof consists in this that the end is a cause only to the extent that the existence of what is ordered to an end depends upon this end as upon something essentially prior. This is clear since every cause qua cause is prior in this way. Now this situation obtains if, and only if, the end as loved moves the efficient cause to give existence to the effect in question, so that the efficient cause would not give existence if the end were not simultaneously contributing its measure of causality. Hence only what the efficient cause brings into existence for love of the end is caused by the end. 2.16 A corollary follows at this point. One should not fail to mention a false opinion concerning the nature of the end, namely, that the final cause of a thing is its last operation or the object attained through this operation. If one were to think that this as such is the final cause, he would be wrong, because this follows the existence of the thing ordered to the end and the latter's existence is not essentially dependent upon it But it is precisely that for the love of which the efficient cause brings something to be that, as loved, is the final cause of what was made, for it is to the beloved that the latter is ordered. At times, it may well be that the object of the ultimate operation is something loved in this way and therefore it would be the final cause. But it would not be because it is the term of such a nature's operation, but rather because it is loved by that which causes this nature. Nevertheless, it is not without reason that the ultimate operation of a thing or the object attained thereby is at times referred to as an end, for it is ultimate and is in some way the best and as such verifies some of the requirements for a final cause. 2.17 Consequently, Aristotle would not maintain that the Intelligences, while lacking an efficient cause, nevertheless have an end in the proper sense of the term. If he would admit that they had only an end, however, it would be in an improper sense where end is understood as the object of their most perfect operation. Or if he would grant them a proper efficient cause, the latter would not be one which produces movement or change, because the four causes are treated in metaphysics where abstraction is made from any physical considerations concerning them. If he assumes them to be eternal and necessary, he would not admit that the First Being gives the Intelligences being after non-being, at least if "after" is taken in a temporal sense. "After" could only be taken in the sense of posterior in the order of nature, according to Avicenna's explanation of the meaning of creation in the sixth book of his Metaphysics, chapter two. Whether or not there is something incompatible about the idea of a thing being caused necessarily does not affect the point of our argument. For if some efficient cause could cause in a way that is simply necessary, and if some end could move the efficient cause in a necessary manner instead of the way that it does, every effect would still be possible not merely in the sense of being opposed to what is impossible, but also because it is not of itself necessary existence, for it is caused. According to the philosophers, however, it would not be possible in the sense that possible excludes any kind of necessary existence, be it caused or uncaused, for they deny that the separate substances are contingent in this sense. 2.18 Another evident corollary is that the end is the final cause of the effect and not of the efficient cause. Consequently, if an agent is said to act for the sake of an end, it is not the end of the agent but the end of the effect that is referred to. 2.19 (Sixth conclusion) What is not an effect is not made of matter. 2.20 Proof: since matter as such is in contradictory potency to form, it does not actually form itself, but requires something else to actuate its potentialities. And this is none other than the efficient cause of the composite [ of matter and form], for to make the composite and to actually form the matter are one and the same thing. The first consequence is clear, for no passive or contradictory potency actualizes itself. And if you say that the form actuates this potency, this is true, but it does so as a formal cause. Before this actual formation takes place, then, matter and form must be regarded as separate, and that which unites them has the character of an efficient cause.

2.21 A second proof of the conclusion is this. In the order of relative priority the efficient cause is next to the final cause and therefore it precedes the material cause. But what does not stem from a prior cause is not caused by a posterior cause. Proof of our first proposition. The causation of the end consists in this that by being loved it moves metaphorically. Now it is only the efficient cause that is moved in this fashion. 2.22 A third proof is this. The composite, being truly one, has some unitary entity which is neither the entity of the matter nor that of the form. Now the primary cause of this unity is not the two entities as such, for if one thing is produced from several it is only because of something which is itself one. Neither is the primary cause the matter or the form taken singly, for each is less than the total entity. Their unity then must stem from an outside cause. 2.23 (Seventh conclusion) What is not made of matter is not formed, and vice versa. 2.24 Proof: What is not made of matter is not composed of essential parts, for in every such composite that is essentially one, there is one part that is potential, because to constitute an essential unity according to the seventh and eighth books of the Metaphysics, a thing must be made of potency and act. Anything that does not possess a potential part as an essential constituent, therefore, is not composed. Consequently, it is not formed, for to be formed means to have form as a constituent part. What was said of matter and form could also be applied in corresponding fashion to subject and accident. 2.25 A confirmation of this proof is to be found in what Aristotle says in the seventh book of the Metaphysics. "If anything were compounded of but one element, that one will be the thing itself." And what is more, such a thing could not be an element according to our first conclusion in this second chapter. By using the topical rule a simili, therefore, we can argue: If anything has but one essential part, then it is just that part. And what is more, the latter is not really a part nor a cause, because of the first conclusion just referred to. Consequently, anything that has one intrinsic cause will also have another which exercises its causal influence along with the first. And so the conclusion we set out to prove becomes clear. 2.26 (Eighth conclusion) What is not caused by extrinsic causes is not caused by intrinsic causes. 2.27 This is sufficiently clear from the four conclusions just mentioned. However, there are special proofs for it. The first is that the causality exercised by extrinsic causes expresses a perfection which does not of necessity entail any imperfection, whereas intrinsic causes of necessity imply imperfection. Consequently, in the act of causing extrinsic causes are prior to intrinsic causes in the sense that the perfect is prior to the imperfect. This in conjunction with the third conclusion entails what we set out to prove. 2.28 The second proof is this that intrinsic causes can themselves be caused in turn by extrinsic causes, and are therefore posterior to them in causing. The antecedent is evident in the case of the form. It is also clear in the case of matter to the extent that the latter is a part. We shall show below that it also holds good for matter itself. 2.29 (Ninth conclusion) The four kinds of causes are essentially ordered in their causation of one and the same thing. 2.30 This is clear enough from the preceding five conclusions. But it seems to be reasonable enough in itself that if something essentially one depends upon more than one cause, some order should prevail so that it does not proceed from them in haphazard fashion. For a plurality of causes which are not related to one another as act and potency or possess no unity of order whatsoever cannot be expected to produce something essentially one. Since the four causes are not constituent parts of a composite of act and potency, how then will they produce the same thing if they do not at least cause together? In so far as they are causing the effect, then, they possess a unity of order. By reason of this order they become a functional unit as regards causation even as many things in the universe become an essential unit through order.

2.31 The type of order that obtains in the case of these various causes should be clear from what was said of the mutual relations of end and efficient cause. (See the second proof of the fourth conclusion and the second of the sixth conclusion. Also look there and in the eighth conclusion to discover how final and efficient causes are related to the other causes.) 2.32 Here I am not much concerned with investigating how the intrinsic causes are ordered to one another as I make little use of them in what follows. As for one being independent of the other, however, it would seem that matter is prior, for the contingent and informing cause seems to depend upon the permanent and informed, since we think of what can be formed before we think of its form. That is how some explain Augustine's remarks in the Confessions regarding the priority of matter to form. And if you ask, according to what order is it prior? I reply: as the more proximate effect of the same remote cause—more proximate, I say according to that order which necessarily obtains inasmuch as the present production of the form [presupposes the matter is produced]. So far as the order of eminence is concerned, however, it is form that is prior [to matter] since it is the more perfect. Though Aristotle assumes this to be evident, when he compares the two in Bk. VII of the Metaphysics, it can be proved from other statements of his about act and potency in Bk. IX of the Metaphysics. 2.33 Keep in mind, however, that it is one thing for causes to be essentially ordered in causing or as regards causation and quite another for the things which are causes to be essentially ordered, as is clear from Avicenna in the sixth book of the Metaphysics, chapter five. For the first is true and has been shown to be such. Otherwise the following propositions would be false: "Because something loves the end, therefore it causes the effect" and "Because it produces the effect, therefore the form informs and the matter functions as material cause"; nevertheless these propositions are commonly admitted. The second, however, is false, for the end is not the cause of the efficient cause; neither is the converse ever true. Ordinarily, however, the efficient cause is not the cause of the matter since it presupposes it. 2.34 Having completed the comparison of the members of the fourth division, I pass on to a brief consideration of the third, since it is clear that the members of this division are mutually exclusive and exhaust what is divided, for (Tenth conclusion) if two effects are compared to the same cause, the latter must be either a proximate or a remote cause. 2.35 As for the second division, I propose two conclusions, the first of which has to do with the distinction of the members. (Eleventh conclusion) If one and the same cause produces two effects, one of which is more immediate than the other, it is not always the case that the more immediate effect is the cause of the more remote effect. Consequently, one effect can be prior to another, but not because it is the cause of the other. 2.36 The antecedent of this conclusion is proved by an example and by reason. The example offered is this. Quantity is an effect more immediate than quality, yet it is not the cause of quality, as is evident if we go through the causes. It is also proved by reason. Look [in 1.131]. 2.37 The second conclusion has to do with the adequacy of the division. (Twelfth conclusion) Nothing is essentially dependent except upon a cause or upon a more immediate effect of some cause. 2.38 Here is the proof. If it did depend upon something else, let that something be A and let the dependent be B. Now if A does not exist, neither will B. But it is possible for all the essential causes of B to be present as well as all the effects of these causes which are more immediate than B, while A is still nonexistent, for A [by stipulation] is not included among them. Consequently, B will not exist, even though all its essential causes concur and all the more proximate effects are present. All these essential causes, then, do not suffice as causes, even when the more immediate effects are produced. It is clear that this follows, for once the more proximate effects are present, the causes if they are sufficient can cause the more remote effect. 2.39 If you say the argument does not imply that they cannot cause but merely that they do not, you are missing the point. For if you stipulate A cannot exist, then neither can B. The existence of all the aforesaid causes and their more immediate effects cannot guarantee the existence of A, since A is none of these, neither can it be caused by them. Consequently, neither can B be caused by them. For if one thing is a necessary condition for another to be, then nothing which is powerless to bring about the first can account for the existence of the second. 2.40 Neither is it relevant to object that a natural agent can produce something made of matter without being able to create the matter which is indispensable to its existence, for a natural agent is not the total cause of the composite as would be the case if it could produce the latter even were everything else excluded. The reason I bring up the point is this. Even if I were to assemble all the various kinds of causes B possesses, plus those more immediate effects prerequisite for B, A's existence is not thereby entailed, for A is neither a cause nor is it one of the effects enumerated. Still B can exist only if A does. If this be so, then it follows that the aforesaid combination as a whole cannot produce B and in consequence is not its total cause, which is the very opposite of what we assumed.

2.41 As for the first division, I present two similar conclusions, the first of which is that the members of the division are distinct. (Thirteenth conclusion) If one thing exceeds another, it does not follow that the latter depends essentially on the former. Consequently, the first member of this initial division does not necessarily imply the second. 2.42 Proof of the antecedent: the less noble type is exceeded by the more noble (e.g. the lesser contrary by the greater), but the latter is not the cause of the former as is clear inductively [if we run through the causes]. Neither is one a more immediate effect, since the causality of their common cause is not related to them as to essentially united effects, for if it were, then it could not produce the inferior effect unless it had first produced the superior effect. This is clearly not the case with any cause. For if the inferior contrary is produced by this cause, even if the superior has not been produced by any cause, then it follows that the two contraries are not essentially ordered to any cause. Furthermore, if the one of greater excellence is neither the cause of the other nor a more immediate effect of their common cause, then it follows that what is inferior is not essentially dependent upon everything which excels it. This is clear enough from what has just been shown [2.38]. 2.43 Though it be superfluous, I add the converse: (Fourteenth conclusion) It is not the case that every dependent thing is thereby inferior to that upon which it depends. 2.44 It is clear that what is composed of material is much more perfect than the material upon which it depends. Possibly the form depends in a similar way upon matter, as suggested in the ninth conclusion, even though the form is more perfect according to the seventh book of the Metaphysics. In the case of orderly changes, too, what is eventually generated depends upon what went on before, for the anterior developments are the more immediate effects of a common cause. Nevertheless, according to the ninth book of the Metaphysics, the later developments are more perfect. 2.45 To show that this division is adequate, I propose this third generalization well known to Aristotle: (Fifteenth conclusion) Plurality must never be assumed without necessity. 2.46 Since there is no apparent need, then, for assuming any primary essential orders beyond the two already mentioned, they alone are to be assumed. This same general principle also indicates that there are only six essential orders into which the second member is subdivided. This many we have shown, and there is no apparent need for assuming others. 2.47 Having made a general comparison of the members of the first division with one another, I compare in particular the posterior of the first order (viz. that which is excelled) with two special posterior members of the second order (viz. the effect and what is ordered to an end) . Here I propose one conclusion, which is as follows: (Sixteenth conclusion) Everything ordered to an end is excelled. 2.48 Proof for this is found in the fact that the end is better than anything ordered to it. This in turn follows from the fact that it is the end insofar as it is loved that incites the efficient cause to productivity. Let the end be A, the efficient cause B and the effect ordered to an end C. Then we argue in this fashion. A is no worse than B; neither is it equally good; consequently, it must be better. The second part of the antecedent (viz. that it is not equal) follows from this consideration. Whatever the reason might be why A would move an equally excellent B, for the very same reason B could move itself, since it is equally lovable and desirable. And so B could be its own final cause, which is contrary to the initial [ conclusion] of this second [chapter]. From this consideration it can also be inferred that A is not worse than B. 2.49 Moreover, according to the second book of the Physics, nature acts for the sake of an end as art would act if it were a natural function. But where products of art are concerned, the knowledge which is the art principle [or premise, if you will], is a knowledge of the end or goal to be achieved, whereas it is the conclusion which has to do with whatever is ordered to the end. Now the premise is truer. Consequently, the end, which includes the truth of the conclusion virtually, is more perfect than the subject of the conclusion. 2.50 You may object that at times voluntary actions are motivated by the love of some inferior good and in such a case the end is not as good as the action performed to obtain it. This is illustrated in every action which is good in itself but evil by reason of the end or purpose for which it was performed, for in this case the action is ordered by the agent to an end inferior to itself. Now my answer to this is that our conclusion holds for those ends which lie in the very nature of things such as is invariably the case with the goals which nature seeks or the aims of a well ordered will. But even the instance of the inordinate will does not really vitiate the conclusion, since in this case we are not dealing with the primary cause of the effect. For even granting that such a will does direct its action to some less perfect end, this action has been ordained by some higher cause for a more perfect purpose, for otherwise, it would not have been ordered, as the proof for the conclusion indicates. But if it has a more perfect end, to the extent it is produced by a higher cause, then it follows that something else is more perfect than it is, and consequently that everything ordered to an end is excelled by some end, even though, perchance, it may not be excelled by the immediate end for love of which some proximate agent produced it.

2.51 There is also another answer which might be given to the objection, viz. that the end which the inordinate agent intends is an end only in a qualified sense. But this is not a happy solution, because the efficiency of the inferior cause is a pure and simple case of efficiency. In other words, it does not effect something simply because it is moved as does a stick, which has no end of its own because it is not an agent on its own, but rather resembles a more immediate effect. If the inordinate will, I say, does not produce its effect in this fashion, then its end is an end in an unqualified sense, since there is some proper end in the case of every proper efficient cause.