The Definition of Nature in Aristotle, Physics II.1

I don't know how interesting this will be to others, but I found it surprising. (Also, I don't have any commentaries with me, so this definitely counts as seat-of-the-pants scholarship. But that's how all true scholarship starts, isn't it?)

I had always taken it for granted that, at the beginning of Physics II.1, Aristotle presents a definition of 'nature'. But yesterday I began to suspect that he is defining at least three distinct things, and maybe four:

• nature (phusis)
• to come to be by nature (phusei)
• to happen in accordance (or as a result of) a nature (kata phusin)
• (perhaps) to have a nature (phusin echein)

If this is true, it would be important to distinguish these, and to understand their relation to one another, because so often for Aristotle everything hinges on which notion is taken to be primary and which derivative.

I came suspect this following the maxim (which I stress to my students) that Aristotle should be read backwards. At 193a2 he says, ti/ me\n ou}n e0stin h( fu&sij, ei1rhtai, kai\ ti/ to_ fu&sei kai\ kata_ fu&sin. And then this can be matched fairly well to the text. In what follows I include serially all the lines of the opening of Physics II.1, but I divide them into sections as I think is appropriate.

fu&sij
(Note that, if this parsing is correct, then, in defining phusis, what Aristotle is really wishing to define is a particular manner of constitution, and when he defines phusei, below, what interests him is how something with that sort of constitution perpetuates itself by bringing about things that are like it.)
Tw~n o1ntwn ta_ me/n e0sti fu&sei, ta_ de\ di' a1llaj ai0- (8)
ti/aj, fu&sei me\n ta& te zw|~a kai\ ta_ me/rh au)tw~n kai\ ta_
futa_ kai\ ta_ a(pla~ tw~n swma&twn, oi[on gh~ kai\ pu~r kai\ (10)
a)h_r kai\ u3dwr (tau~ta ga_r ei]nai kai\ ta_ toiau~ta fu&sei
fame/n), pa&nta de\ tau~ta fai/netai diafe/ronta pro_j ta_
mh_ fu&sei sunestw~ta. tou&twn me\n ga_r e3kaston e0n e9autw|~
a)rxh_n e1xei kinh&sewj kai\ sta&sewj, ta_ me\n kata_ to&pon,
ta_ de\ kat' au1chsin kai\ fqi/sin, ta_ de\ kat' a)lloi/wsin: (15)
kli/nh de\ kai\ i9ma&tion, kai\ ei1 ti toiou~ton a1llo ge/noj
e0sti/n, h|{ me\n tetu&xhke th~j kathgori/aj e9ka&sthj kai\
kaq' o3son e0sti\n a)po_ te/xnhj, ou)demi/an o(rmh_n e1xei meta-
bolh~j e1mfuton, h|{ de\ sumbe/bhken au)toi=j ei]nai liqi/noij h2
ghi5noij h2 miktoi=j e0k tou&twn, e1xei, kai\ kata_ tosou~ton, w(j
ou1shj th~j fu&sewj a)rxh~j tino_j kai\ ai0ti/aj tou~ kinei=sqai kai\ (20)
h)remei=n e0n w|{ u(pa&rxei prw&twj kaq' au(to_ kai\ mh_ kata_ sumbebhko/j

fu&sei
(I include in this section the end of the clause of the previous one, highlighted. I believe that these are linked by a phenomenon in Aristotle that I like to call 'chaining', that is, where Aristotle uses the conclusion of one line of thought to begin a seamless transition into another line of thought. The comparison with the physician is sometimes taken to be offered as an illustration of what it is for something to have a nature; I take it to have a more limited purpose, viz. to clarify how something that has a nature perpetuates itself, by causing something similar to exist.)
e0n w|{ u(pa&rxei prw&twj kaq' au(to_ kai\ mh_ kata_
sumbebhko&j (le/gw de\ to_ mh_ kata_ sumbebhko&j, o3ti ge/-
noit' a2n au)to_j au(tw|~ tij ai1tioj u(giei/aj w2n i0atro&j: a)ll'
o3mwj ou) kaqo_ u(gia&zetai th_n i0atrikh_n e1xei, a)lla_ sumbe/- (25)
bhken to_n au)to_n i0atro_n ei]nai kai\ u(giazo&menon: dio_ kai\ xwri/-
zetai/ pot' a)p' a)llh&lwn). o(moi/wj de\ kai\ tw~n a1llwn e3ka-
ston tw~n poioume/nwn: ou)de\n ga_r au)tw~n e1xei th_n a)rxh_n e0n e9au-
tw|~ th~j poih&sewj, a)lla_ ta_ me\n e0n a1lloij kai\ e1cwqen, oi[on
oi0ki/a kai\ tw~n a1llwn tw~n xeirokmh&twn e3kaston, ta_ d' e0n (30)
au(toi=j me\n a)ll' ou) kaq' au(ta&, o3sa kata_ sumbebhko_j
ai1tia ge/noit' a)\n au(toi=j.

fu/sin e/)xein
(This notion is not mentioned in the summary at 193b2, yet it seems to get distinct treatment. It is a new notion, once hupokeimenon and ousia are introduced, because then the phusis becomes something which something else has.)
fu&sij me\n ou}n e0sti\ to_ r(hqe/n: fu&sin de\
e1xei o3sa toiau&thn e1xei a)rxh&n. kai\ e1stin pa&nta tau~ta ou)si/a:
u(pokei/menon ga&r ti, kai\ e0n u(pokeime/nw| e0sti\n h( fu&sij a)ei/.

kata_ fu&sin
kata_ fu&sin de\ tau~ta& te kai\ o3sa tou&toij u(pa&rxei kaq' (35)
au(ta&, oi[on tw|~ puri\ fe/resqai a1nw: tou~to ga_r fu&sij me\n ou)k
e1stin ou)d' e1xei fu&sin, fu&sei de\ kai\ kata_ fu&sin e0sti/n. ti/ me\n
ou}n e0stin h( fu&sij, ei1rhtai, kai\ ti/ to_ fu&sei kai\ kata_ fu&sin
(I understand the highlighted kai/ to be exepegtical, sc. " 'by nature', in the sense of 'in accordance with nature'".)