Ṣadrā’s Use of Fakhr al-Dīn al-Rāzī’s al-Mabāḥith al-mashriqiyya in the Asfār


 The aim of this paper is to show that, in several chapters of the Asfār, Ṣadrā’s use of Fakhr al-Dīn al-Rāzī’s al-Mabāhith al-mashriqiyya is far more extensive than what a superficial reading, focusing only on explicit quotations, might indicate, and to explore what knowing this can bring to our reading of Ṣadrā’s text. It analyzes what Ṣadrā does with his source and examines a few examples to illustrate why it matters to know.


bonmariage
Oriens 48 (2020) 188-218 writes here is often mainstream philosophical discourse and could as well be borrowed from a number of other texts. But it is still possible to determine that, even if it is undoubtedly accompanied by other tools, Fakhr al-Dīn al-Rāzī's Mabāḥith is used here.6 In the middle of pages of summary or paraphrase of Rāzī's text, suddenly his name appears, showing that the Mabāḥith is clearly on Ṣadrā's desk. For instance, in the introduction, an "Imām al-Rāzī says" opens a paragraph that is in fact surrounded by texts already heavily depending on Rāzī.7 Rāzī's name appears again later, to introduce an objection raised against a demonstration of the finitude of dimensions, and its refutation.8 Again, the beginning of the section on the void is explicitly a quotation from Rāzī,9 but it is unclear where it ends (his name appears also in the next section, on quality, a section that is often a reworked paraphrase of the Mabāḥith).10 In the section where the question of the number of categories is addressed, there is a brilliant passage when, speaking about the third question raised by Rāzī ("there is no category outside these ten [categories]"), Ṣadrā refuses to quote the argument given by Avicenna, and here mentioned by Rāzī, writing: "To demonstrate this, the Shaykh gave an argument that is weak and bad, as he himself admits [the very words used by Rāzī at the end of the quotation], and thus we skipped it."11 Ṣadrā skips here two full pages in Rāzī's Mabāḥith, before resuming his paraphrase. This shows again that Ṣadrā has the text of the Mabāḥith in mind when he writes these pages.

2
What Is Ṣadrā Doing with the Mabāḥith?
In the passages we are referring to, Ṣadrā uses the Mabāḥith as raw material for his own text. What he does is not follow uncritically his source but rather use 6 On the Mabāḥith as a philosophically inclined text, and its relation with the Avicennan corpus, see Ayman Shihadeh, "From al-Ghazālī to al-Rāzī: 6th/12th Century Developments in Muslim Philosophical Theology," Arabic Sciences and Philosophy 15 (2005) it, along with other texts, while remaining always attentive to his own purpose and what he wants to say. We will turn shortly to what this can bring to our reading of the Asfār. But first, we would like to look closer at what Ṣadrā does when he uses the Mabāḥith. The use made by Ṣadrā of Rāzī's text can be divided into three main categories.
(1) Text 1 in the Appendix (Asfār 8:28-9) is an example of the first type. It follows Rāzī's text with some additions and changes-some noticeable such as the use of 'animal' where Rāzī had 'human being'-but on the whole, the text is similar to that of the Mabāḥith. (2) An example of the second type is Text 2 in the Appendix (Asfār 8:93-7): Ṣadrā presents what he borrows from Rāzī as the view generally held, inserting words such as yuqāl ('it is said') or ʿalā mā huwa l-mashhūr ('according to what people generally hold') in the Rāzīan discourse (Texts 2 (A) and (D)). It is often Rāzī's text that is meant when Ṣadrā alludes to what thinkers "such as Avicenna and the like" hold. An example of that can be found at the end of the section on perception by external senses, where Ṣadrā writes after a portion taken from the Mabāḥith: "following what is recorded in the books of the qawm, such as the Shifāʾ and the like" (on odors).12 This happens quite often: another example is found in the section on the relative (muḍāf ), where Ṣadrā labels what he borrows from Rāzī again as "what can be learned from the discourse of the Shaykh and others," adding: "and I have objections."13 Again, on substance: "This is what the ḥukamāʾ say in such a place."14 Quotations of Avicenna attributed to an anonymous author (with introductory sentences such as "One of them writes that the Shaykh al-Raʾīs said") are also often borrowed from Rāzī.15 (3) In other passages, Ṣadrā engages in a continuous discussion with Rāzī, without mentioning his name, turning his discourse into a commentary on Rāzī's text of some sort. These are portions of text constructed as a dialogue ("someone says" or "one of the eminent (thinkers) says" … "I say"), where the text put in the mouth of the "someone" or of the "eminent" is that of Fakhr al-Dīn al-Rāzī. An example is found in Asfār 4:242, where Rāzī's words are attributed to "one of the eminent (people)" (baʿḍ al-fuḍalāʾ). 16 Another example is what Ṣadrā writes in faṣl 7 of the fann on quantity, on intensification, where Ṣadrā discusses extensively the two arguments presented here by Rāzī ( faṣl 9 of his chapter on quantity), and introduces his words by a "They say."17 In a more subtle manner, there are passages where Ṣadrā discusses a quotation made by Rāzī (often from Avicenna) and comments on Rāzī's opinion upon it, such as this example where, after quoting Avicenna, Ṣadrā concludes by saying: "There is thus no confusion in what he says, as some of them thought," referring again to Rāzī in the Mabāḥith.18 This comes in addition to the passages where Ṣadrā offers an explicit commentary of Rāzī, often with his text summarized or paraphrased, as is the case at the end of fann 3 in the discussion of categories, as mentioned earlier, or a little later in the same text, in a section where objections to the supporters of the void are discussed.19 In making use of Rāzī's Mabāḥith, Ṣadrā always appropriates the material: he reorganizes paragraphs, skips entire portions in Rāzī's text, he summarizes and paraphrases, interrupts the flow of Rāzī's discourse with a paragraph of his own, criticizes his source-text, changes the order and the wording-in a word, uses Rāzī's text for his own purpose.
An example of alteration in the order of the text on a small scale can be seen in the treatment of the arguments for the tashkīk, or the analogical character, of the concept of being: the arguments mentioned in the Asfār are taken in part from Rāzī's Mabāhith but presented in a different order and rearranged to fit Ṣadrā's discourse.20 Another example is found in the section on the per- ception of particulars by the soul, where Ṣadrā brings together the objections and their answer, putting the answer just after the corresponding objection, when Rāzī had all the objections first, followed by the answers to each of them (with also different answers at times).21 But sometimes the modification in the organization of the text is more radical. Thus on chance (ittifāq), we see Ṣadrā rearrange portions of text from the Mabāḥith as elements for his own text, putting together paragraphs coming from different chapters in Rāzī's treatment of the question in his chapter on the final cause: in the beginning of the chapter (mabḥath 2: Fī l-ittifāq), Ṣadrā brings together the position attributed to Democritus and that attributed to Empedocles, presented by Rāzī in faṣl 6 and faṣl 4 of his chapter on the final cause respectively, while the end of the mabḥath corresponds to the beginning of faṣl 4.22 There seems also to be places where Ṣadrā adds portions of text coming directly from Ibn Sīnā's al-Samāʿ al-ṭabīʿī (the part of the Shifāʾ on physics), a text to which Rāzī himself is close, as can be seen in the passage put by Ṣadrā between the demonstrations in favor of chance and their refutation, that seems to be closer to the Shifāʾ than to what Rāzī writes in the Mabāḥith on similar questions.23 It seems, and this could support our claim here, that Ṣadrā follows up on quotations that Rāzī makes of Ibn Sīnā, for he writes after giving such a citation: "This is what al-Fakhr al-Rāzī quotes from the Shaykh, but I did not find these words in his writings."24 This is the only case we found so far however.25 In some places, Ṣadrā follows the structure of the text as presented by Rāzī in the Mabāhith, but not what Rāzī says. Thus in the third part of the discussion on qualities, "On the qualities that exist in what has a soul," while following the frame he finds in the Mabāḥith, and keeping the sequence of topics (qudra, irāda, etc.), Ṣadrā chooses to provide an entirely different content under these titles.26 He plays here with different sources as well: at the beginning of this passage, Ṣadrā states that far more is said on these questions in the books of the Sufis, such as ʿAbdullāh Anṣārī's (396-481/1006-1089) Manāzil al-sāʾirīn and the last two sections of Ibn Sīnā's Ishārāt, known independently as Maqā-

21
Asfār 8:230-41 ( faṣl 5); Mabāḥith 2:351-7 (= 2:337-44). 22 Asfār 2:253-9; see Mabāḥith 1:653-9 (= 1:531-7). 23 Asfār 2 māt al-ʿārifīn. But he adds that what he is going to provide here is a chapter on each of the qualities mentioned in the books that we can label as pertaining to philosophy, except for knowledge, a topic so important it needs a separate chapter.27 As for changes in the wording, these are often without consequences, like when Ṣadrā uses "those who deny the unicity of the soul" (al-munkirūn liwaḥdat al-nafs)28 where Rāzī had "those who hold that the souls are many" (al-mukaththirūn li-l-nafs),29 or "the arguments of those who support chance" (ḥujaj al-qāʾilīn bi-l-ittifāq), where Rāzī had "the proofs of those who deny aims" (adillat munkirī l-ghāyāt).30 The same holds for places where Ṣadrā alters the sequence of words in propositions such as "that which increases the dimensions of the natural body, so that it reaches the perfection of its growth according to natural proportions," for "that which increases the dimensions of the body according to natural proportions so that it reaches the perfection of its growth" in the Mabāḥith (Appendix, Text 2 (C)).31 Ṣadrā almost systematically avoids the precise numbers given by Rāzī in his divisions, a clever practice when using another text to build one's own (see for example Appendix, Text 1 (A) and (A')). This is often the case, as Ṣadrā rarely follows all the arguments or questions listed by Rāzī.32 But changes in wording may also be an expression of something more important, for instance when Ṣadrā changes the words introducing paragraphs such as taḥqīq dhālika ("the reality of this [is shown by]," "what this is really"), which becomes in the Asfār: wa-mimmā yadullu ʿalā dhālika ("one of the things that show this") (Appendix, Text 2 (E)).33 Or when, speaking about what accounts for the conservation of the mix of elements, Ṣadrā widens the question to include all animals by just changing one word in Rāzī's text (Appendix, Text 1 (B)).

27
Asfār 4 Again, Ṣadrā's use of Rāzī's text does not mean he agrees with what is said by Rāzī or endorses his teaching: Rāzī's text is rather taken as the expression of the basic understanding of a topic, or the ground of a thorough critique. Thus on "when" in the discussion on categories, Ṣadrā writes: "This is what has been said, and it is not correct," followed by one page of explanation of his own point of view on the question.34 And on being in a position, after a broad definition based, but with other words, on Fakhr al-Dīn, Ṣadrā adds "and what it is really for us is rather (…)."35 Again, on qualities, Ṣadrā writes at the end of his paraphrase of Rāzī: "We have something higher to say than this on these sensible qualities; we will come back to it later."36

Why Does It Matter to Know?
The fact that we can identify Rāzī's Mabāḥith as the subtext to a number of passages in the Asfār, besides offering a much larger textual basis for an analysis of Ṣadrā's understanding and critique of Rāzī than that provided by the explicit quotations, allows us to better understand what appears often in Ṣadrā's text as short questions and lengthy answers. Now that we know that Rāzī is Ṣadrā's interlocutor or target, we can find out what is the context of the debates and understand allusions that could remain incomprehensible otherwise. But the most interesting benefit of our reading lies perhaps in the quick access it allows to the gist of Ṣadrā's thought, to what Ṣadrā considers as keynotions, which is easily found in the portions of texts where he stops following, even loosely, his source-text. This is obvious in passages where he interrupts his summarized paraphrase of Rāzī's text by a paragraph entitled ḥikma mashriqiyya, as he does in faṣl 2 of the chapter on quantity,37 or, on a smaller scale, when he introduces a portion of text starting with a "for us" (as in Appendix, Text 3 (C)). But other changes are more subtle, as we will see shortly in examples taken from texts where Ṣadrā remains closer to his source. At a bigger scale, the same question can be asked: why does he stop using the Mabāḥith on such and such question? What happens in the third part of the discussion on qualities, "On the qualities that exist in what has a soul," where Ṣadrā follows, as we said earlier, the order of the topics in the Mabāḥith, but changes entirely 34 Asfār 4:219-20; Mabāḥith 1:581 (= 1:454-5). 35 Asfār 4:221-2; Mabāḥith 1:581-2 (= 1:455). 36 Asfār 4:104; after Mabāḥith 1: 426 (= 1:311). 37 Asfār 4:12-3. The portions of texts from the Mabāḥith used before and after correspond to Mabāḥith 1:280 (= 1:175) and 1:284 (= 1:178).
the content: why (what is so unpalatable here for Ṣadrā?) and to replace it by what, are questions worth investigating. Reading the Asfār with an eye on the Mabāḥith helps also identify references made in the text. For instance, in Asfār 9:77, where Ṣadrā had "one of the commentators of the Qānūn said," the text of the Mabāḥith provides us with a more specific reference, for Rāzī writes: "We explained in our commentary on the Qānūn."38 The opposite is also true: Ṣadrā provides in some places more details than what is found in the Mabāḥith. An example of this is found in faṣl 2 of fann 1 in the categories, on quantity, "On the distinction between dimension and corporeity," where Rāzī had just "someone might say." Ṣadrā is here more precise, and writes: "the author of the Ḥikmat al-ishrāq objected that […]."39 The same happens later in the discussion on categories: "It has been answered" in Rāzī becomes in the Asfār: "It has been answered, as appears in the Shifāʾ."40 Again when listing the three elements that prepare the soul to joy, Ṣadrā adds the following information: "as the Shaykh mentioned in his Epistle Fī l-adwiya alqalbiyya." Here we see also Ṣadrā give a much larger quotation than what Rāzī had.41 A parallel reading of the Asfār and the Mabāḥith has thus something to offer to the reader of Rāzī as well. This is also true for the edition of the texts (even if we found only minor cases so far). The examples here mentioned are from one perspective, from the point of view of Ṣadrā's text, but the same must happen for the text of the Mabāḥith. One example can be seen in Text 2 (B) in the Appendix, where the editors of the Mabāḥith made a better choice than what appears in Ṣadrā's text: the reading ‫غ‬ ‫ا‬ ‫ي‬ ‫ة‬ (goal) fits better in the sequence "form, matter, goal" (and corresponds to the enumeration made just before in the text, where one reads: "There is a form and a matter for this change; and for the agent, there is a goal in its actions"), than ‫غ‬ ‫ا‬ ‫ذ‬ ‫ي‬ ‫ة‬ (the nutritive).42 But then again, this might be a voluntary change by Ṣadrā. It might happen that some of these variants may not be in manuscripts of one of the two textual traditions, making it thus important to compare the two texts.
Reading the Asfār alongside the Mabāḥith clarifies also details of Ṣadrā's text, such as when Ṣadrā summarizes so much that something goes wrong, 38 Asfār 9:77; Mabāḥith 2:420 (= 2:409). 39 Asfār 4 In the portions of text where Ṣadrā follows closely Rāzī's Mabāḥith, the benefit of a reading attentive to both texts is even greater: we can focus on the passages where the texts are dissimilar (here even more easily apprehensible), where the important points made by Ṣadrā are situated. I will explore here two examples in more details to show how this works (examples 1 and 2), followed by one example to illustrate how focusing on dissimilarities between the texts can make some passages more interesting than they initially look (example 3).

Asfār 8:93-7 // Mabāḥith 2:271-3 (= Appendix, Text 2)
We referred to this section earlier regarding two points. Ṣadrā presents here what he borrows from Rāzī's text as that which is usually said ( yuqāl, dhukira)43 or what is commonly held (ʿalā mā huwa l-mashhūr)44 (Texts 2 (A) and (D)). But he is careful not to endorse Rāzī's teaching too eagerly: this is the passage where we see Ṣadrā prefer "one of the things that show this" (wa-mimmā yadullu ʿalā dhālika) when Rāzī used "what this is really" (or "the reality of this [is shown by]," taḥqīq dhālika) (Text 2 (E)).45 This passage deserves also our attention for what happens at the end. This follows the allusion to a first wrong supposition: that held by those who consider the nutritive power to be fire. This is already found in the Fī l-nafs of the Shifāʾ,46 but the treatment of the question and the wording is clearly that of Rāzī in the Mabāḥith. Ṣadrā considers next another (wrong) supposition: that there is a void in the members of the body that the power of growth would fill. There, Ṣadrā objects to the text by Ibn Sīnā presented by Rāzī as the truth (alḥaqq), before giving his own explanation (Text 2 (F)).

4.2
Asfār 8:205-8 // Mabāḥith 2:335-6 (= Appendix, Text 3) In the section on the sensus communis (al-ḥiss al-mushtarak) (Fī ʿilm al-nafs, bāb 5: On internal perceptions, faṣl 1), a lot of parallels with Rāzī's Mabāḥith are found, as in the whole section on internal senses. This is normal on such a topic. But Ṣadrā could be following other texts, and even Ibn Sīnā directly here (or Suhrawardī or a later Ishrāqī for some aspects of what he says). But it is again Rāzī's Mabāḥith that is on his desk, as witnessed by the wording, and 43 Asfār 8:94, l. 9 and 14. 44 Asfār 8 by the direct quotation given at the end of the section, yet another example of how Rāzī's name appears in the portions of text where Ṣadrā takes his text as the background of his own. What do we see when reading this passage along with Rāzī's text? First, Ṣadrā presents as that which is commonly held (ʿinda l-jumhūr), what Rāzī said about the sensus communis as "localized in the front of the brain" (Text 3 (B)). For Ṣadrā, it is only the disposition (istiʿdād) of this faculty of the soul that is in the front of the brain, not the faculty itself. He adds again a precision: "Or rather, in the spirit (rūḥ) that runs in it" (Text 3 (C)), a precision given by Ibn Sīnā at the end of the section on the soul of the Shifāʾ (maqāla 5, faṣl 8).47 The caution of Ṣadrā in using the proper language here explains also another change in wording (Text 3 (A)): even when relating the common view on this faculty, Ṣadrā prefers to use the word 'posited' (mūdaʿa), and not 'impressed' (murtasama) as Fakhr al-Dīn al-Rāzī had it (Ibn Sīnā used murattaba, 'disposed,' in the Shifāʾ and the Najāh).48 Again, the same caution can be noticed where Ṣadrā speaks of "the forms of the external sensations" being brought together to the sensus communis, where Rāzī had "the external sensations" (Text 3 (D)).
Another remarkable change of wording occurs where 'corporeal' is systematically replaced by 'particular' when speaking of the faculty or instrument needed to perceive sensations. It is through a "particular instrument," not a "corporeal faculty," that sensible perception occurs; the external sensations come together in a "particular perceptive faculty," not "an internal corporeal faculty" (Text 3 (F-F²)). And here again, Ṣadrā insists on the fact that it is the intellect that ultimately judges on sensible perceptions. Where Rāzī said that the judgment "this thing that is touched is (or is not) this thing that has such or such color" cannot be made by the intellect, Ṣadrā has rather that what makes this judgment cannot be "the intellect alone, with no intermediary sense, because its perception of the sensible occurs only with a particular instrument" (Text 3 (E)). In two remarkable additions, Ṣadrā makes clear what he himself claims, that is, that it is the intellect in human beings, and the estimative in animals, that brings together all the perceptive faculties and makes judgments (Texts 3 (E) and (G)).
As for the direct quotation of Rāzī ("the author of the Mabāḥith") at the end of this faṣl (Text 3 (H)), it is again made by Ṣadrā to criticize what Rāzī makes of a quotation (here from Bahmanyār) in his own text (note the similarity of method with the end of the text analyzed just before,

Asfār 3:507-19 // Fakhr al Dīn al-Rāzī, Mafātīḥ al-ghayb (= al-Tafsīr al-kabīr), on Quran 2:31, masʾala 8
To know the underlying layer of a passage, and to be thus able to determine quickly where Ṣadrā's text is different from Rāzī's, allows us also to make the best of texts that, at first sight, seem rather unlikely to yield any interesting element. The clearest example we found so far is not related to the Mabāḥith but to another text by Rāzī, his Tafsīr al-kabīr. It is a list of definitions of terms that Ṣadrā adds as a conclusion at the end of safar 1, marḥala 10, on knowledge, using the material he finds in Rāzī's tafsīr of Quran 2:31, "He taught Adam all the names."49 In fact, the title of the section in the Asfār ("Where one explains terms with similar meanings used in this domain") is that of the corresponding, and much shorter, section in Rāzī's Mabāḥith,50 but the content is closer to the Tafsīr al-kabīr. A version of the same passage is found in the third miftāḥ of Ṣadrā's Mafātīḥ al-ghayb, on knowledge (miftāḥ 3, mashhad 7).51 In this latter text, the borrowing follows a section where Rāzī's Tafsīr is already heavily present. This is not surprising: mashhad 6 consists of quotations from the Quran and hadiths in praise of knowledge and the search for it, where Ṣadrā uses what he finds in Rāzī.52 Here again, Ṣadrā not only reorganizes but appropriates this material: one telling example is where a saying attributed to ʿUmar b. al-Khaṭṭāb in Rāzī's Tafsīr is quietly anonymized by Ṣadrā, who attributes it to "one of the Companions" (baʿḍ al-ṣaḥāba).53

49
Asfār 3 2:31, masʾala 7)). He does this reluctantly, it seems, for in his introductory paragraph he writes (Shīrāzī. Mafātīḥ,99): "It is not in the habit of someone who is searching the truth to devote his attention to the sayings of those who have no opening and no vision for the perception of realities, like the usual mutakallimūn, masters in qalqala ["agitation"], linguists and specialists in rhetoric. There is no harm however in quoting what they say to get a picture of what the words convey […]." Ṣadrā appears here to get fascinated so to speak by his source, but only to a certain point: at the end of the miftāḥ (131; end of mashhad 6), he leaves aside entirely the stories and anecdotes with which Rāzī concludes his section (Rāzī, al-Tafsīr al-kabīr, 2:211-6, on Quran 2:31, masʾala 6).
As in other examples mentioned above, the name of Rāzī is not entirely absent in the passage of the Asfār. It appears in two cases where Ṣadrā wants to object to Rāzī's understanding of a word (under tadhakkur and fikr), as well as under dhikr, where Ṣadrā clearly shows that he is here reading Rāzī's text, for he summarizes a few lines he skipped from his source and writes: "The author of the Mabāḥith says, after repeating the objection he claims cannot be solved," followed by an explicit quotation of Rāzī.54 Note that here, Ṣadrā seems to willingly try to blur his source by calling Rāzī "the author of the Mabāḥith." In the parallel text in the Mafātīḥ, he chose "al-Fakhr al-Rāzī."55 Likewise, under fikr, he uses "Imām al-Rāzī" in the Mafātīḥ, but "the author of the Mulakhkhaṣ" in the Asfār.56 What does a reading of this passage with our "Rāzī lenses" bring to our understanding of what matters to Ṣadrā in this list of definitions of terms related to knowledge? The terms here defined are the following (* = different but on the basis of Rāzī's text; */ = wholly different): idrāk; shuʿūr; taṣawwur; ḥafẓ (*); tadhakkur (*); dhikr (*); maʿrifa; fahm; fiqh; ʿaql (*/); ḥikma; dirāya; dhihn; fikr (*); ḥads; dhakāʾ; fitna; khāṭir; wahm (*); ẓann; ʿilm al-yaqīn (*/); ʿayn al-yaqīn; ḥaqq al-yaqīn; bidāha (*); awwaliyyāt; khayāl (*); khubr; raʾy; firāsa.
The comparison with the underlying text shows some changes in the order of the entries, and places where Ṣadrā disagrees with his source. It allows to quickly determine that if we leave aside ʿaql and ʿilm al-yaqīn, where Ṣadrā offers something entirely different, the notions where Ṣadrā engages in a discussion with his source-text are taṣawwur, fikr, three terms that have to do with the notion of memory, both retentive and active (tadhakkur, dhikr, ḥafẓ), and the terms wahm and khayāl.
Ṣadrā gives a short discussion on taṣawwur and fikr. On taṣawwur, he attributes to the ʿāmmat al-nās the understanding of form (ṣūra) given by Rāzī, as "the corporeal quiddity actualized for the body with a shape," and prefers its definition as "that by which the thing becomes in act this thing."57 On fikr, Ṣadrā rejects the restriction here made by Rāzī of this activity to assent.58 54 Asfār 3:511. The quotation is from Rāzī, al-Tafsīr al-kabīr, 2:223. 55 Shīrāzī,Mafātīḥ,134. 56 Shīrāzī, Mafātīḥ, 138, and Asfār 3:516. In Ṣadrā's Mafātīḥ al-ghayb, the specific text of Rāzī is even mentioned, but in mashhad 1, where Ṣadrā rebukes the "author of the Tafsīr alkabīr" for his understanding of knowledge as "the relation (nisba) between the knower and the known," an understanding which is meant to counter the difficulties inherent to a definition of knowledge as "the actualized form in the soul," but which is actually "very foolish" (Shīrāzī,Mafātīḥ,107). 57 Asfār 3:508. 58 Asfār 3:516.
Wahm and khayāl receive a more extensive treatment. On wahm, Ṣadrā states that for him, the estimative power (al-quwwa al-wāhima) is not separate from the intellect or imagination, but it is rather the intellect when related to the imaginative or sensitive form.59 On khayāl, Ṣadrā takes this opportunity to express his two main objections: against the idea that the imaginative forms are imprinted in one of the powers of the body, as philosophers have it; and against the idea that they are separated from the soul, in a discrete world corresponding to the ʿālam al-mithāl al-muṭlaq, rather than in the world the soul creates for itself, a position Ṣadrā attributes to the Ishrāqī thinkers.60 The discussion on the possibility of recalling a form that would have slipped the mind (under tadhakkur61), is for Ṣadrā an occasion to reiterate his understanding of human soul as having several dimensions, or domains, of existence. Rāzī wrote that the experience of recalling is a mystery (sirr), since we cannot recall that which is represented to us nor that which is not represented to us, a version of a well-known paradox, and yet recalling is an experience we share, hence the mystery. Ṣadrā recognizes that in his understanding of intellective perception as "being only through the unification of the soul with the Agent Intellect, which is the form of the existents, or in which the forms of the existents exist," this difficulty is even harder to solve. This compels him to think of a solution that he finds in the multiple dimensions of existence and activity of the souls, and the souls' scale of perfection. It is the soul intermediate in its perfection-not so imperfect that it is restricted to the dimension of sensation (with just a faint experience of the imaginative life), but not so perfect either that one dimension would not veil it from another-that experiences this "recalling" when, leaving its intelligible dimension, and left with a weak image of its intelligible experience and the ability to recall and join the intelligible dimension again, it recalls that which was manifested to it. I am not convinced Ṣadrā would have made it as clear, had he not had Rāzī's text in front of him.62 Our reading of this passage with a focus on the sections where Ṣadrā engages with Rāzī's text, allows us thus to understand that the main focus of Ṣadrā in this list of definitions is wahm and khayāl on the one hand, and terms that have to do with memory and recalling on the other. These are precisely the 59 Asfār 3:517. 60 Asfār 3:518. 61 In Asfār 3:509, the heading of the paragraph reads al-dhikr and should be corrected to altadhakkur according to the parallel texts and to the fact that the next heading is precisely al-dhikr. 62 Asfār 3:509-11. key notions in Ṣadrā's personal contribution on internal senses. We might have seen this without knowing that the passage is written with Rāzī's Tafsīr as a background. But we could have missed the significance of some aspects of the discussion, and we certainly would not have grasped it so quickly.

Conclusion
No misunderstanding: the passages where Ṣadrā does not rely on Rāzī are far more extensive than the ones where he does. It is important to clarify the extent to which Ṣadrā borrows from Rāzī's Mabāḥith (and on which topics), and to explore the ways he uses and adapts this text. But it is also important to investigate if there is some significance in his not using Rāzī on subjects where the latter had something to offer (on metempsychosis, for instance).
There are several ways to consider what Ṣadrā does here. Most passages are found in what can be labelled as "unavoidable sections," where definitions are given, or well-known objections to such and such question, or again, in lists like the one we just analyzed or that of the characteristics of human beings.63 Ṣadrā's use of the Mabāḥith could be interpreted as showing a lack of interest in some questions that need to be addressed in a summa but are not particularly challenging for him. But it could also be interpreted simply as a way to save time. Why "reinvent the wheel" here instead of using a standard exposé as the basis of one's own text? Again, what Ṣadrā does here is not to blindly reproduce his source-text, but to rework it for his own purpose. And his use of Rāzī's text does not mean that he endorses what Rāzī says. About al-Rāzī himself, Ṣadrā has this to say: "My God! Has there been a human being (ādamī) in the world who reached the same point in the abundance of study and investigation, wrote as many books and was absorbed in thinking as much as he was, and was still so far from the truth?"64 As for Rāzī's Mabāḥith, that he uses so often as raw material for his own discourse, they are for him only "so-called" or "just only said to be" oriental (mashriqiyya).65 What I want to emphasize here is not what this tells about Ṣadrā's method or aims in writing, but rather the extremely valuable tool that knowing the 63 Asfār 9:78-83 (Fī ʿilm al-nafs, bāb 9, faṣl 1: Fī khawāṣṣ al-insān). See Mabāḥith 2:421-4 (= 2:409-13). This is another example of a passage based on Rāzī but with significant differences. 64 Asfār 8

some people argue that the soul is the complexion.
Six arguments show the falsity of their claim.
[Several] arguments show the falsity of their claim. competing with one another to get disengaged. That which makes ( j-r-r) them That which obliges ( j-b-r) them to be a mix (imtizāj) be a mix (imtizāj) and to produce the complexion (mizāj) is a power other than what follows their being a mix. And after is not identical with before.
is a power other than them.
Whether we say that the elements remain in their specific form, as is 67 The parallel between the two texts continues beyond the portion to that, to be joined and to produce the mix, and there is no and it is not the case that there is something that forces them to cohere or keeper to keep them, a keeper to keep them or to keep the mix. that would be the soul.

A.2
Text 2 for the quiddity is only through its parts, like the genus and the differentia, and the definition according to the act of being is not possible through the parts, for the act of being has no part. It is thus either through the agent and the end, if it is complete, or through the action, if it is deficient. This because of what we alluded to earlier, i.e. that the definition and the demonstration of the act of being are one and the same thing, and the demonstration by the 'why' is stronger than that by the 'that' . This being established, we say: due to its simplicity, each faculty is an individual instance of being. The faculties can be explained through their actions. Thus, the nutritive faculty is explained by that which comes from it, We mentioned already that the nutritive and it is said that it A

Mabāḥith 2:271-3
Asfār 8:93-7 There is a form and a matter for this change, and a place; and for the agent, there is a goal in its action.
There is a form and a matter for this change; and for the agent, there is a goal in its actions. Nutrition and growth reach their completion through three things.
(1) That parts similar in quiddity to that which is fed and to that which grows are obtained.
(1) That nutrients similar in quiddity in potency to that which is fed and to that which grows are obtained.
(2) That [these parts] are attached to them.
( 2)  And that which the Shaykh mentions in the Shifāʾ, and is approved by one of the fuḍalāʾ, that the power of growth separates the junction of the member and introduces in these gaps the nutritive parts. And the same goes for nutrition.
that the power of growth separates the junction of the member and introduces in these gaps the nutritive portions and that the same goes for nutrition, is not satisfactory for us.