3.1 The triple primacy of the First Principle.

3.2 O Lord, our God, you have proclaimed yourself to be the first and last. Teach your servant to show by reason what he holds with faith most certain, that you are the most eminent, the first efficient cause and the last end. 3.3 We would like to select three of the six essential orders referred to earlier, the two of extrinsic causality and the one of eminence and, if you grant us to do so, to demonstrate that in these three orders there is some one nature which is simply first. I say one "nature" advisedly, since in this third chapter these three ways of being first will be shown to characterize not a unique singular or what is but one in number, but a unique essence or nature. Numerical unity, however, will be discussed later. 3.4 (First conclusion) Some nature among beings can produce an effect. 3.5 This is shown to be so because something can be produced and therefore something can be productive. The implication is evident from the nature of correlatives. Proof of the antecedent: (1) Some nature is contingent. It is possible for it to exist after being nonexistent, not of itself, however, or by reason of nothing, for in both these cases a being would exist by reason of what is not a being. Therefore it is producible by another. (2) Some nature too is changeable or mobile, since it can lack some perfection it is able to have. The result of the change then can begin to be and thus be produced. 3.6 In this conclusion, as in some of those which follow, I could argue in terms of the actual thus. Some nature is producing since some nature is produced, because some nature begins to exist, for some nature is contingent and the result of motion. But I prefer to propose conclusions and premises about the possible. For once those about the actual are granted, those about the possible are also conceded, but the reverse is not the case. Also those about the actual are contingent, though evident enough, whereas those about the possible are necessary. The former concern the being as existing whereas the latter can pertain properly to a being considered even in terms of its essentials. The existence of this essence, of which efficiency is now established, will be proved later. 3.7 (Second conclusion) Something able to produce an effect is simply first, that is to say, it neither can be produced by an efficient cause nor does it exercise its efficient causality in virtue of anything other than itself. 3.8 It is proved from the first conclusion that something can produce an effect. Call this producer A. If A is first in the way explained, we have immediately what we seek to prove. If it is not such, then it is a posterior agent either because it can be produced by something else or because it is able to produce its effect only in virtue of some agent other than itself. To deny the negation is to assert the affirmation. Let us assume that this being is not first and call it B. Then we can argue of B as we did of A. Either we go on ad infinitum so that each thing in reference to what precedes it in the series will be second; or we shall reach something that has nothing prior to it. However, an infinity in the ascending order is impossible; hence a primacy is necessary because whatever has nothing prior is not posterior to anything posterior to itself, for the second conclusion of chapter two does away with a circle in causes. 3.9 An objection is raised here on the grounds that those who philosophize admit that an infinity is possible in an ascending order, as they themselves were wont to assume infinite generators of which none is first but each is second to some other, and still they assume no circle in causes. In ruling out this objection I declare that the philosophers did not postulate the possibility of an infinity in causes essentially ordered, but only in causes accidentally ordered, as is evident from Avicenna's Metaphysics, B. VI, chapter five, where he speaks of an infinity of individuals in a species. 3.10 But to show what I have in mind, I will explain what essentially ordered and accidentally ordered causes are. Here recall that it is one thing to speak of incidental causes (causae per accidens) as contrasted with those which are intended to cause a given effect (causae per se) . It is quite another to speak of causes which are ordered to one another essentially or of themselves ( per se) and those which are ordered only accidentally (per accidens). For in the first instance, we have merely a one-to-one comparison, [namely] of the cause to that which is caused. A per se cause is one which causes a given effect by reason of its proper nature and not in virtue of something incidental to it. In the second instance, two causes are compared with each other insofar as they are causes of the same thing.

3.11 Per se or essentially ordered causes differ from accidentally ordered causes in three respects. The first difference is that in essentially ordered causes, the second depends upon the first precisely in the act of causing. In accidentally ordered causes this is not the case, although the second may depend upon the first for its existence or in some other way. The second difference is that in essentially ordered causes the causality is of another nature and order, inasmuch as the higher cause is the more perfect, which is not the case with accidentally ordered causes. This second difference is a consequence of the first, since no cause in the exercise of its causality is essentially dependent upon a cause of the same nature as itself, for to produce anything one cause of a given kind suffices. A third difference follows, viz. that all essentially ordered causes are simultaneously required to cause the effect, for otherwise some causality essential to the effect would be wanting. In accidentally ordered causes this simultaneity is not required. 3.12 What we intend to show from this is that an infinity of essentially ordered causes is impossible, and that an infinity of accidentally ordered causes is also impossible unless we admit a terminus in an essentially ordered series. Therefore there is no way in which an infinity in essentially ordered causes is possible. And even if we deny the existence of an essential order, an infinity of causes is still impossible. Consequently in any case there is something able to produce an effect which is simply first. Here three propositions are assumed. For the sake of brevity, call the first A, the second B and the third C. 3.13 The proof of these: first, A is proved. (1) If the totality of essentially ordered causes were caused, it would have to be by a cause which does not belong to the group, otherwise it would be its own cause. The whole series of dependents then is dependent and upon something which is not one of the group. (2) [If this were not so], an infinity of essentially ordered causes would be acting at the same time (a consequence of the third difference mentioned above). Now no philosopher assumes this. (3) Thirdly, to be prior, according to Bk. V of the Metaphysics, a thing must be nearer the beginning. Consequently, where there is no beginning, nothing can be essentially prior to anything else. (4) Fourthly, by reason of the second difference, the higher cause is more perfect in its causality, therefore what is infinitely higher is infinitely more perfect, and hence of infinite perfection in its causing. Therefore it does not cause in virtue of another, because everything of this kind is imperfect in its causality, since it depends upon another to produce its effect. (5) Fifthly, inasmuch as to be able to produce something does not imply any imperfection—a point evident from conclusion eight of chapter two—it follows that this ability can exist in some nature without imperfection. But if every cause depends upon some prior cause, then efficiency would never be found without imperfection. Consequently, an independent power to produce something can exist in some nature and this is simply first. Therefore, such an efficient power is possible and this suffices for now, since we shall prove later from that that it exists in reality. And so A becomes evident from these five arguments. 3.14 Proof of B: If we assume an infinity of accidentally ordered causes, it is clear that these are not concurrent, but one succeeds another so that the second, though it is in some way from the preceding, does not depend upon it for the exercise of its causality. For it is equally effective whether the preceding cause exists or not. A son in turn may beget a child just as well whether his father be dead or alive. But an infinite succession of such causes is impossible unless it exists in virtue of some nature of infinite duration from which the whole succession and every part thereof depends. For no change of form is perpetuated save in virtue of something permanent which is not a part of that succession, since everything of this succession which is in flux is of the same nature. Something essentially prior to the series, then, exists, for everything that is part of the succession depends upon it, and this dependence is of a different order from that by which it depends upon the immediately preceding cause where the latter is a part of the succession. Therefore B is evident. 3.15 Proof of C: From the first conclusion, some nature is able to produce an effect. But if an essential order of agents be denied, then this nature capable of causing does not cause in virtue of some other cause, and even if we assume that in one individual it is caused, nevertheless in some other it will not be caused, and this is what we propose to prove to be true of the first nature. For if we assume that in every individual this nature is caused, then a contradiction follows if we deny the existence of an essential order, since no nature that is caused can be assumed to exist in each individual in such a way that it is included in an accidental order of causes without being at the same time essentially ordered to some other nature. This follows from B. 3.16 (Third conclusion) If what is able to cause effectively is simply first, then it is itself incapable of being caused, since it cannot be produced and is independently able to produce its effects. 3.17 This is clear from the second conclusion, for if such a being could cause only in virtue of something else or if it could be produced, then either a process ad infinitum or a circle in causes would result, or else the series would terminate in some being which cannot be produced and yet independently is able to produce an effect. This latter being I call "first," and from what you grant, it is clear that anything other than this is not first. Furthermore, it follows that if the first cannot be produced, then it has no causes whatsoever, for it cannot be the result of a final cause (from conclusion two of chapter two)—nor of a material cause (from the sixth conclusion of the same)—nor of a formal cause (from the seventh conclusion there). Neither can it be caused by matter and form together (from the eighth conclusion there). 3.18 (Fourth conclusion) A being able to exercise efficient causality which is simply first actually exists, and some nature actually existing is capable of exercising such causality. 3:19 Proof of this: Anything to whose nature it is repugnant to receive existence from something else, exists of itself if it is able to exist at all. To receive existence from something else is repugnant to the very notion of a being which is first in the order of efficiency, as is clear from the third conclusion. And it can exist, as is clear from the second conclusion. Indeed, the fifth argument there which seems to be less conclusive than the others established this much. The other proofs there can be considered in the existential mode—in which case they concern contingent, though manifest facts—or they can be understood of the nature, the quiddity and possibility, in which case the conclusions proceed from necessary premises. From all this it follows that an efficient cause which is first in the unqualified sense of the term can exist of itself, for what does not actually exist of itself is incapable of existing of itself. Otherwise a nonexistent being would cause something to exist; but this is impossible, even apart from the fact that in such a case the thing would be its own cause and hence could not be entirely uncaused. Another way to establish this fourth conclusion would be to argue from the impropriety of a universe that would lack the highest possible degree of being. 3.20 As a corollary of this fourth conclusion, note that not only is such a cause prior to all others, but that it would be contradictory to say that another is prior to it. And insofar as such a cause is first, it exists. This is proved in the same way as was the fourth conclusion. The very notion of such a being implies its inability to be caused. Therefore, if it can exist, owing to the fact that to be is not contradictory to it, then it follows that it can exist of itself and consequently that it does exist of itself.

3.21 (Fifth conclusion) A being unable to be caused is of itself necessarily existent. 3.22 Proof: By excluding every cause of existence other than itself, whether it be intrinsic or extrinsic, we make it impossible for it not to be. Proof: Nothing can be nonexistent unless something either positively or privatively incompatible with it can exist, for one of two contradictories is always true. But nothing can be either positively or privatively incompatible with a being which cannot be caused, because it would be either of itself or from another. Not the first way, for then it would exist of itself—from the fourth conclusion,—so that there would be two incompatible things, and for that reason neither would exist, since you admit that the uncausable is nonexistent because of this incompatible element and vice versa. Neither can the incompatible be from another, because nothing caused has a more intense or potent existence from a cause than an uncausable thing has of itself, since the former is dependent in existing whereas the uncausable is not. Furthermore the possibility of the causable being does not entail its actual existence as is the case with the uncausable. Nothing incompatible with what is already a being can come from a cause unless it receive from that cause a being more intense or powerful than is the being of that which is incompatible with it. 3.23 (Sixth conclusion) It is the characteristic of but one nature to have necessary being of itself. 3.24 This is proved thus: If two natures of themselves could be necessary being, then this necessity of existing would be a common feature. And this they would share by reason of some essential or generic kind of entity in addition to which they would differ by reason of their ultimate actual formalities. Now two inconsistencies follow from this. To begin with, each will be a necessary being first of all through that common nature which is the less actual, rather than through that distinctive nature which is the more actual. For were it necessary being also by reason of its distinctive nature, then it will be necessary being twice over, because that distinguishing nature does not formally include the common nature, even as a [specific] difference does not include the genus. It seems impossible however, that the less actual be the primary reason why something is necessary, and that it is neither primarily nor per se necessary by reason of what is more actual. The second impossibility is that neither of the two would necessarily exist by virtue of that common nature which is presumed to be the primary reason why each is necessary. For that nature is insufficient to account for the existence of either nature, since every nature is what it is by reason of its ultimate formal constituent. But it is precisely what—to the exclusion of all else—accounts for a thing's actual existence, that is the reason for its being necessary. If you say that the common nature suffices for existence apart from the distinguishing natures, then it follows that the common nature of itself exists actually and without any distinguishing features, and therefore cannot be distinguished, since the necessary being already existing is not in potency to being [different kinds of things] in an unqualified sense [in the way] that the generic being in a species is simply that kind of thing. 3.25 Besides, two natures included under a common class are unequal. Proof of this is to be found among the different kinds of things into which a genus is divided. But if the two such natures are unequal, one will be of a more perfect being than the other. Nothing however is more perfect than a being having necessary existence of itself. 3.26 Moreover, if there were two natures having necessary being of themselves, neither would depend upon the other for existence and consequently no essential order would exist between them. One of them, therefore, would not belong to this universe, for there is nothing in the universe which is not related by an essential order to the other beings, for the unity of the universe stems from the order of its parts. Here it is objected that inasmuch as each is related to the parts of the universe through the order of eminence, this suffices for unity. To the contrary: One is not so ordered to the other, for a more perfect existence characterizes the more eminent nature. Nothing however is more perfect than a being having necessary existence of itself. What is more, one of two is not ordered to the parts of the universe, because if the universe is one, then it is characterized by a single order and this obtains where there is but one first. Proof: If you assume there are two first natures, since there is a dual term of reference, the nature next to the first has no unique order or dependence and the same is true of each subsequent nature. And thus through the whole universe there will be two orders, and hence two universes. Or else where will be an order only to one necessary being, but not to the other. If one proceeds reasonably, then, it seems he ought not to postulate anything for no apparent need, or whose entity is not clearly revealed by reason of some order to other things,—for, according to Physics, Bk. I, more than one thing should not be postulated where one suffices. Now we show there is a necessary being in the universe from the uncausable, and this in turn from what is first in causing, and the latter from what is caused. But from these effects there is no apparent necessity for assuming several first causing natures; furthermore, this is impossible, as will be shown later in the fifteenth conclusion of this third chapter. Therefore it is not necessary to assume that there are several things which are uncaused and necessarily exist. With reason, then, they are not postulated. 3.27 Concerning the final cause I propose four conclusions similar to the first four in this chapter about a being able to produce an effect. They are also proved in a similar way. The first of these is this: (Seventh conclusion) Among beings some nature is able to function as final cause. 3.28 Proof: Since something is producible (from the proof of the first conclusion of this chapter), something is able to be ordered to an end. The implication is clear from the fourth conclusion of chapter two. That an essential order is involved is even more evident here than in the case of the efficient cause (from conclusion nine of chapter two). 3.29 (Eighth conclusion) Something able to be an end is simply ultimate, that is to say, it can neither be ordained to something else nor exercise its finality in virtue of something else. 3.30 This is proved by five arguments similar to those advanced for the second conclusion of this third chapter.

3.31 (Ninth conclusion) Such an ultimate end cannot be caused in any way. 3.32 This is proved from the fact that it cannot be ordained for another end; otherwise it would not be ultimate. It follows further that it cannot be caused by an efficient cause (from conclusion four of chapter two and also from what was said above in the proof for the third conclusion of the present chapter). 3.33 (Tenth conclusion) The being which can be an ultimate end actually exists, and that this primacy pertains to some actually existing nature. 3.34 The proof for this is like that used for the fourth conclusion of chapter three. Corollary: It is first to such an extent that it is impossible that anything should be prior to it. This is proved in the same fashion as the corollary to the fourth conclusion above. 3.35 Having set down four conclusions about both orders of extrinsic causality, I submit four like conclusions about the order of eminence, the first of which is this: (Eleventh conclusion) Among beings, there is some nature which excels. 3.36 This is proved from the fact that something is ordered to an end (from conclusion seven of this chapter); therefore it is excelled (from conclusion sixteen of chapter two). 3.37 (Twelfth conclusion) Some eminent nature is simply first in perfection. 3.38 This is clear because we have an essential order. As Aristotle points out in the Metaphysics, Bk. VIII, forms are like numbers. And in such an order, an ultimate nature is to be found. This is proved by the five reasons given above for the second conclusion. 3.39 (Thirteenth conclusion) The supreme nature cannot be caused . 3.40 Proof is found in the fact that it cannot be ordained to an end (from conclusion sixteen of chapter two), and therefore it cannot be caused by an efficient cause (from conclusion four of the same chapter! . The other ways of being caused are excluded as in the proof for the third conclusion of this chapter. An additional proof that the supreme nature cannot be caused by an efficient cause is to be found in the reason given for proposition B, in the proof for the second conclusion of this chapter [DP 3.14]. For whatever can be produced has some cause to which it is essentially ordered.

3.41 (Fourteenth conclusion) The supreme nature is something actually existing 3.42 The proof of this is like that of conclusion four of this chapter. Corollary: It is contradictory that any nature should be more perfect or higher than this. The proof for this is like that for the corollary of the fourth conclusion above. 3.43 (Fifteenth conclusion) In some one and the same actually existing nature, there is the triple primacy in the aforementioned triple order, namely, of efficiency, finality and eminence. 3.44 This fifteenth conclusion comes as the fruit of this chapter. It clearly follows from what has been shown in this way. If to exist necessarily of itself is characteristic of but one nature (from the sixth conclusion of this chapter) and if such existence is proper to whatever possesses a primacy in any of the aforementioned three ways (from the fifth and third conclusions for one primacy and from the fifth and ninth for the second and from the fifth and thirteenth for the third), then each of the aforesaid primacies belongs to that unique nature to which the others belong. For each is actually in some -nature (from conclusions four, ten and fourteen), and they are not split up; therefore they are in one and the same nature. Proof of the minor: Otherwise there would be many necessarily existing natures (from the second proposition of the argument just cited). Our point is also proved from the fact that what is first cannot be caused, for only such is first. But everyone of the aforementioned firsts is uncausable; therefore, etc. Proof of the major: How will a multitude arise of itself? 3.45 This is a very fertile conclusion, containing, as it does, six others virtually. Three of these are about the aforementioned orders considered singly, viz. that at the head of each is one nature. Three more identify the first nature in one order with the first nature in the other orders. And this conclusion which is so fertile has been shown through the sixth conclusion alone, as through a kind of major premise. It is good to express the proper majors for these, if they can be found. 3.46 For the first two conclusions I suggest this simple conclusion as a premise. (Sixteenth conclusion) It is impossible for the same being to depend essentially on two things in such a way that its total dependence terminates with each. 3.47 Proof of this: If for a given causal category one cause is the total cause of something, then it is impossible that another cause of the same category should cause the same thing, for then one and the same thing would be twice caused or neither would be the total cause. In like fashion, one could argue that in such a case something would be caused by an agent which, even if it did not cause the thing in question, would leave the latter caused by it, which is absurd. So too it is impossible for the same thing to depend, by any dependence, on two beings when one of them totally terminates its dependence. For if it still depends upon the other, the first does not sufficiently terminate its dependency. Similarly then it would depend upon something without the existence of which it would still exist, and in the same order of being. But this runs counter to the very notion of dependence. 3.48 Having established this conclusion, I now propose the first ones [virtually] included in the fifteenth conclusion in this fashion: (Seventeenth conclusion) There is but one nature which is first as regards any given type of extrinsic causality. 3.49 Proof: If such a primacy pertains to more than one, then it does so with respect either to the same or to a different set of secondary beings. But the first is not the case (from the sixteenth conclusion just established). What is more, in such a case in every one of the secondary things there would be a twofold dependence of the same type, since where two first beings are involved, there is no single dependence. This consequence, however, is unacceptable. The other way is also out, for if the other first and its secondaries form a distinct set, then they will constitute a distinct universe, because the members of the two sets are neither ordered to each other nor to the same thing. Without unity of order there is no unity to the universe. Yet Aristotle assumes the principal goodness of the universe to consist in its single end. And since there is but one order to one supreme [thing] it is enough for me to speak of this universe alone and not to fabricate another for no reason whatsoever; or rather, for which there are reasons to the contrary. 3.50 We also add some probable proofs. The ascending progression in an essential order is from more to less. Consequently, it terminates with one.

3.51 Again, the higher the cause the more effects does its causality embrace. As one goes higher, then, fewer causes suffice, etc. This proof clarifies the previous one. 3.52 Again, it seems clear enough that the primacy of eminence pertains to one nature, for if two natures cannot be so ordered that one does not excel the other (for in this respect they are like numbers), then it is even more impossible that two different natures should be first to the same degree. 3.53 Then there is this argument about the end: No end would fully satisfy everything other than itself. Since this cannot be the case, the implication is as before. 3.54 What is more, no nature would contain virtually the perfection of every other nature. But without contradicting ourselves, we cannot think that no nature is most perfect. 3.55 As for the other three conclusions there are also special proofs for: (Eighteenth conclusion) That being which is first able to produce an effect is most actual, since it virtually contains all possible actuality. The first end is the best, virtually containing all possible goodness. The first of those beings which can excel is most perfect, containing eminently all possible perfection. 3.56 These three cannot be separated, for if one were in one nature and another in another, it would be impossible for one of them to be simply eminent. These three primacies, then, are seen to express three necessarily concurrent features of the supreme goodness, viz. the highest communicability, amiability and integrity or wholeness. For "good" and "perfect" are the same (from Metaphysics, Bk. V) and "perfect" and "whole" are the same (from Physics, Bk. III). From Bk. I of Ethics, however, it is clear that the good is desirable and from Bk. VI of Avicenna's Metaphysics, that good tends to communicate or give of itself. But nothing perfectly gives of itself unless it does so out of liberality. And this is surely a characteristic of the supreme good, since it expects nothing in return from its giving. And such, according to Avicenna (chapter five of the same book), is the property of one who is liberal. 3.57 (Nineteenth conclusion:) But one existing nature is first in the aforesaid triple way with reference to every other nature, so that any such is, therefore, posterior to it in a threefold way. 3.58 Some petulant objector, while admitting the fifteenth conclusion, could say that besides this nature there are many others which though not first in this [threefold] way, are posterior to this first nature according to some, but not all, of the aforesaid orders. They would be posterior only as regards the order of eminence, or of eminence and finality, but not that of efficiency, as some say Aristotle felt was the case with those intelligences that came after the first, or perhaps with prime matter. Now while this can be refuted from what has been said so far, still it is helpful to explain how. 3.59 In the first place the sixth conclusion disproves this, for if to have necessary existence pertains to [but] one nature and whatever is not posterior as regards any of the three orders is of itself necessary being, then there is but one nature which is not posterior by any kind of posteriority. Consequently every other nature is thus posterior in a threefold way. The second proposition of this argument [i.e. whatever is not posterior is a necessary being] is clear from the third, ninth and thirteenth conclusions of this third chapter (add to each the sixth conclusion of this same chapter). 3.60 Secondly, there are proofs for particular orders. Whatever is neither an end nor ordered to some end, exists in vain. Among beings, [however], nothing exists in vain. Every nature other than the first end, therefore, is ordered to some end, and if to some end, then it is ordered to the first end (from the third conclusion of the second chapter). There is a similar proof as regards the eminent. Whatever is neither supreme nor excelled by another has no degree of perfection whatsoever, and therefore is nothing. Whatever is not supreme, then, is excelled by something and therefore by what is supreme (from the third conclusion of the second chapter) .

3.61 From these we show what was denied about efficiency [by the petulant objector above, viz. that the secondary intelligences and primary matter are not ordered to a first efficient cause]. Everything is either the first end or is ordered thereto as we have just shown, therefore everything is either the first efficient cause or an effect thereof for the members of the latter disjunct are interchangeable with those of the former as regards posteriority. This is clear from the fourth and fifth conclusion of the second chapter so far as the posterior [portion of the disjuncts are concerned]; as for the prior portion, this is clear from the argument just above. 3.62 A similar proof is that based on the order of eminence. If everything is either supreme or excelled by what is supreme, then everything is either the first efficient cause or an effect, for these members are also interchangeable—from last and second last conclusions of the second chapter and the thirteenth conclusion of this one. What is more, it is quite irrational to assume the existence of some being which has no order [i.e. is essentially unrelated to anything in the universe], as has been shown by the second reason for the sixth conclusion and, to some extent, by the proof for the seventeenth conclusion of this chapter. 3.63 Indeed, O Lord, in wisdom you have made things so ordered that any reasonable intellect may see that every being is ordered. Consequently, it was absurd for the philosophers to deny order of some. From the universal statement "Every being is ordered," then, it follows that not every being is posterior and not every being is prior, since in either case an identical thing would be ordered to itself or else a circle in the ordered would be assumed. Consequently there is some prior being which is not posterior, and is therefore first. And there is some posterior being which is not prior. But nothing exists which is neither prior nor posterior. You are the unique first, and everything besides you comes after you by reason of a threefold order, as I have explained to the best of my ability.