¹ This has been a topic of a remarkable amount of discussion, together with ironic kinds of inattention to those discussions, over the last hundred or so years. Shute, Richard in ‘Aristotle's Physics, Bk. VII’, Anecdota Oxoniensia, Vol. I, pt. III (1882)Google Scholar opened the recent discussion of the interrelation of the texts and the traditions with the interpretation of the role of this book in Aristotle's philosophy, and continued it in his posthumous On the History of the Process by which the Aristotelian Writings Arrived at their Present Form (Oxford, 1888)Google Scholar, in which he maintained that both versions are ‘un-Aristotelian.’ Hoffmann, Ernst, De Arist. Phys. septimi libri origine et auctoritate (Berlin, 1905)Google Scholar, offered a comprehensive discussion, in which he maintained that both versions of Book VII must have been student notes. Ross, W. D. gives critical attention to these and other earlier studies, while adding his own account, in Aristotle's Physics (Oxford, 1936, 1979)Google Scholar. Gohlke, P., ‘Die Entstehungsgeschicte der naturwissenschaftlichen Schriften des Aristoteles,’ Hermes, 1924Google Scholar, develops a thesis, elaborated in Die Entstehung der aristotelischen Prinzipienlehre (Tubingen, 1954)Google Scholar, that Aristotle's ‘doctrine of potency’ was a watershed in his work, with the β version of VII before and the a version after. Runner, Howard E., The Development of Aristotle Illustrated from the Earliest Books of the Physics (Amsterdam, 1951)Google Scholar, makes the issues of Book VII the centerpiece of his dissertation, claiming it separate, early, and the only book of the Physics ‘which can be characterized as Platonic’ Ingmar, During, Aristoteles: Darstellung und Interpretation seines Denkens (Heidelberg, 1966)Google Scholar, offers a students' notes thesis similar to Hoffmann's, but with an earlier dating. Lang, Helen S., ‘God or Soul: The Problem of the First Mover in Physics VII,’ Paedeia, 1978Google Scholar, Special Aristotle volume, while not attending to questions of order or dating, and offering both a ‘chain of change’ interpretation of Chapter 1 and an account of the book's coherence that are similar to Wardy's, focuses on the stated problematic of the book to show how it is dissimilar to and conceptually independent from the arguments of Book VIII and of Metaphysics XII. Her Aristotle's Physics and Its Medieval Varieties (Buffalo, 1992)Google Scholar follows this up with more detail. Bemd Manuwald attends to much of the same material on the soul that catches Lang's eye (but with no note of Lang's work), but argues against the unity of the work on the grounds that Chapters II and III are not consistent with Aristotle's mature psychology. Of all these, Wardy attends only to Hoffmann, Manuwald and Ross in his book. Verbeke, G., ‘L'Argument du livre VII de la Physique, une impasse philosophique’, in Naturforschung bei Aristotles und Theophrast, During, I. (ed.), (Heidelberg, 1969)Google Scholar, gives some broader background to the perplexing variety of interpretations of the book. Many others, such as W. Jaeger, A. Mansion, A. Torstrik, F. Nuyens, F. Solmsen, D. Graham and J. Rist have had something to say on the issues. All except Wardy, Graham and Lang make explicit claim to the separate status of VII, and most all of those except Shute (1888) would place this as somehow a product of Aristotle's own work, but earlier than the hylomorphic accounts of Book II.

² Ross, W. D., Aristotle's Physics (Oxford, 1936), pp. 8–19Google Scholar. In examining the MSS that Ross consulted, I found a number of interesting irregularities regarding Book VII which I think strongly corroborate the Ross thesis. These findings are to appear in my ‘The Bastard Book of the Physics.’

³ Wardy, Robert, The Chain of Change (Cambridge, 1990), p. 89Google Scholar. Helen Lang (1978, pp. 90–94) had already made much the same point.

⁴ Wardy, p. 84.

⁵ Wardy, p. 89, n. 13. Wardy here stops short of the claim that VIII ‘depends’ upon VII, though he does note Hoffman's claim that VIII–5, 256a 13–29 depends upon VII–1 as evidence that his own claim is not original.

⁶ This claim as it stands does not forge any logical link between VII–1 and VIII–5, since it does not take the argument in VII to be presupposed by that in VIII. Rather, Wardy's point is that the proposition ‘baldly stated’ in VIII–5 is the same as the proposition in VII–1 for which an argument is given. That lends plausibility to supposing that VII–1 is the proximate antecedent to VIII–5 by way of seeming to be taken for granted in the latter. That Aristotle had this argument—or any other—in mind when writing VIII–5 is mere speculation. Wardy simply maintains that ‘if it is right that VIII–5's thesis cries out for a supporting argument which VII very conveniently formulates, one reasonably associates the discussion of the two texts despite the absence of encouragement which explicit cross-references would provide’ (p. 89, n. 13).

⁷ Wardy, p. 89.

⁸ Lang's thesis (1978) seems most pertinent here. She advocates that we take seriously the problematic as Aristotle has explicitly set it, and that this book is ‘set against Plato's view that soul originates all motion.’ (p. 86) Even though Aristotle does not himself set his problematic in terms of Plato's views about the soul, Lang's thesis seems more consistent with his setting than does Wardy's. She maintains that the proof of Physics VII is valid on its own conceptual and substantive grounds, and ‘[t]he proof of Physics VII is unique in that it alone considers indirect relations among movers and moved things.’ Her point, in short, is that this book has a separate, independent and autonomous task, without reference to Book VIII. She gives a detailed analysis to show the argument as a whole valid, given its premises. In developing her thesis about the unity of Book VII, she holds no brief for the relation of the book to the whole treatise, or for its relative dating; only for its conceptual autonomy. Runner (1951), too, makes similar claims for unity in his analysis of the argument, but exceeds the Ross thesis, arguing that the book is not only early, but a conceptual watershed between the more platonic views of nature that precede it and the hylomorphic views that follow it.

⁹ Wardy, p. 87, n. 7.

¹⁰ Wardy, p. 94.

¹¹ See his On the Nature of Things. In my ‘Thomas' Conception of Causation,’ Nature and System, 1980, I wrongfully chided him for the ‘something else’ reading of 242a17. It now seems clear to me that the β text was all that he had to work from, and that his extrapolations seem more reasonable for him from that base.

¹² This again stands in counterpoint to Lang's contentions (1979, 1992) that the arguments of each book are to be understood according to their own subject-matters, and stand on their own explicit premises.

¹³ Ross (1936) interprets Simplicius as maintaining that Eudemus ‘omitted book vii’ (Ross, p. 15) on the basis of 1036.13ff., that he ‘passed over’ it. (Ross, p. 3) Runner complains of this interpretation, translating ‘τοτο παρελθὡν ὡc περιττ’ as ‘passed it by as being superfluous.’ (Runner, p. 81) This he takes to presuppose that Eudemus not only had it to take into account, but that it was already in its present position, and Eudemus merely regarded it as not worthy of comment. However, we could just as well take ‘περιττν’ to mean here ‘beyond the regular number’ or ‘extraordinary’ and not conclude with Runner that it had already found its place in the corpus as we know it. This would give us a very different sense of ‘passing by’ from Runner's reading of it.

¹⁴ Shute (1888, p. 119) acknowledged ‘it is more than probable that in the later chapters (3, 4, 5) of the textus receptus, we have still either the second text, or at least a mixture of the two.’ This was not in the 1882 piece that Ross considered.

¹⁵ Erasmus, in his Basel editions, actually got to this realization before Ross, calling the first four books ἣ περ φɛσικν ảρπχν’ and the last four ‘ἣ περ κι7nu;σεω’

¹⁶ This work was based on research done at the Aristotelesarchiv in the Institute fur Altertumskunde of the Freie Universitat in Berlin. I am grateful to my hosts at the Institute, and for travel grants from the University of Kentucky Research Foundation and the Southern Regional Education Board, that helped make that research possible.

¹⁷ Ross, p. 16.

¹⁸ α version, 242b42. Here, Hardie and Gaye give the unfortunate translation, ‘We have dealt with this question above’, implying the structure of the treatise as we now have it. If we took the reference to be regarding place rather than time, the plural would require more than one place. Theβ version at 242b8 has ‘τατα δ' εἲρηται κα 7nu; τοῖc πρτερον’ ‘But this also was spoken to in prior times.’

¹⁹ Aristotle's Physics, p. 11.

²⁰ Who the editor was need not here concern us. We can imagine that Aristotle himself was the editor in the Physics 1–8 passage, from the standpoint of the pedagogue rather than the composer. It seems less likely in the VII–1 passage, since VII seems so neglected in its Peripatetic use. It is not impossible that both were put into place as late as Andronicus' editing. Nothing is lost in either context by simply excising the references. That is, in neither case does the allusion enhance the effectiveness of the argument, in spite of Ross' claims for VII–1 to the contrary.

²¹ Solmsen, Friedrich, Aristotle's System of the Physical World, (Ithaca, 1960), Chapters 2,4, 9, 16Google Scholar.

²² An example of the opposite, strengthening kind of moves can be seen in the claims to necessity regarding the accompanying of generation and destruction by qualitative change, even though generation is not an alteration. This apodictic force appears only in the a version (246b14; 247a6; a13), but in two out of three instances it is accompanied by added argument not in the other version which lends force to the stronger claim. By contrast, the only instance, other than the ‘by something else’ in VII–1, where a parallel section in β is longer is in the midst of this same discussion about generation and alteration, where the differences can be attributed to different ways of talking about matter, perhaps reflecting an improved economy of discourse in the a version, but no discernable difference in the import of the argument.

²³ In spite of his own call that we attend to Aristotle's attention to phenomena, none of these absences are noted by Wardy. On the contrary, he presumes the standard homogenous view of Aristotle's accounts of motion. See p. 83ff. as a lead into the thesis of his book. These presumptions on his part clearly prejudice the development question.

²⁴ Manuwald takes this to be the import of the discussion in the first part of VII–3. See his C–l.

²⁵ Barnes follows the Hardie and Gaye translation of ν οἷc τɛγχ7nu;ο7upsiv;σιν οὖ7sigma;αι πρώτοιc’ as ‘or the elements, whatever they may be, on which the states primarily depend’. This gives the misleading appearance of elements talk where there is none, even though it is in a place where one would ordinarily expect it to be.

²⁶ Solmsen, Friedrick, Aristotle's System of the Physical World (Ithaca, 1960), p. 179Google Scholar, takes both versions to be Aristotle, and attributes the inconsistency to the influence of ongoing Academic discussions. Lang (1979) follows Solmsen on this, and maintains that the whole of the book is devoted to a critique of Plato's conception of souls as self-movers.

²⁷ Whether this satisfies the Manuwald concerns about discrepancies is another story. He cites not only passages outside of VII at odds with the treatment of alteration in VII–3, but also passages within. He notes that at β 242a6–9, change from good to bad is a qualitative change, and the same is true for the restoration of health at β 249a29ff. But the reference in VII–1 has to do with exemplar contraries relative to change, and as I have already noted, these are the same examples he used in Categories 11 in his discussion of number, species and genus. Given that link, it seems a slight presumption that Aristotle simply carried over the same examples without thought to the treatment of qualitative change, but just to the contrariety. The passage in VII–4 embeds becoming healthy as undergoing alteration in the protasis of a conditional, which Aristotle often does when he is not committed to the truth of the clause. A comparable case is near at hand at 249b23, where he entertains, ‘If substance is number…’ Besides, the way this clause is structured would allow us to read the becoming and the alteration as constantly conjoined instead of identical, which would be consistent with the arguments in the a version of VII–3. These considerations may not be satisfying to the partisans of the Manuwald thesis, but they at least ‘save the phenomena’.

²⁸ This seems already on the path to his notion of proximate matter.

²⁹ Here, the conclusion is interestingly different in the two versions. The β version concludes, ‘it is evident that alteration occurs only in perceptible qualities’. The argument is also different: ‘that from which come the form and the shape and what has come to be is not referred to by the same name as the shapes that emerge from it’. The a version has ‘in the case of shape and form the thing which has come to be wherein there is a shape is not referred to by means of its material’. This is a difference not noted in Wardy's list, and it is the one place where the β version takes more words. Even though they appear to be talking about the same things, there seems to be a firmer move toward a nascent hylomorphic treatment in a, propelled by Aristotle's beginning to speak of the out-of-which things are formed in terms of matter.

³⁰ Wardy gives the phrase the translation, ‘substantial and geometrical form’ with no apparent excuse.

³¹ Wardy, except for the instance of ‘substantial form’ at 246b 15, translates ‘εἷδοc’ as ‘species’ throughout Book VII. It is interesting to translate it that way before the VII–3 passage, and as form afterward. This seems particularly pertinent where the locution ‘ἕχ7epsiv;ι εἲδη’ at 249b 12 and εἲδη ἓ7chi;ει’ at b16 in VII–4 require a translation of having εἷδοc In the Categories account, individuals do not have species (as a property), but rather belong to species (as members). On the other hand, in later writings, individuals do have forms, not as properties (supervenient or otherwise), but as the constitutions of their actual beings. This locution in VII–4 seems to invite this form-interpretation, and serves as a follow-up to the introduction of formas-constitution in VII–3. Such an interpretation not only serves as a link between the uses εἷδοc in the Categories and those in the physical works, but it also serves as a link between VII–3 and VII–4, thus reinforcing the unity thesis (against Manuwald). It also gives us an enriched understanding of Aristotle's appropriation of one of Plato's terms for form for a contrasting conception of a unit of being.

³² Manuwald, pp. 82; 128–32.

³³ This helps explain the peculiarities of another ‘un-Aristotelean’ expression in β that bothers both Ross and Wardy: ‘λγω δ τ βλτιστον σωζιατι7theta;ν πρἱ τν φσ7iota;ν.’ (‘I say the best is that which maintains and manages according to the nature of the thing.’ 246b23, my translation) If we do not assume that Aristotle's philosophy of nature is already in place, we can see this as a nascent formulation of what eventually becomes his definition of happiness in his mature account of ethics. Something of this sort has already appeared in the Protrepticus.