Q1

As I said, for Islamic philosophers God is an existent having no metaphysical structure; all real complexity is negated of Him. Now one kind of complexity is the composition between a bearer or possessor of an attribute (i.e., an essence, dhāt) and the attribute itself possessed. In the Islamic tradition, adherents of DDC allow this kind of complexity to characterize God: for they hold that ‘God’ does not refer to just the divine essence alone, nor the divine attributes alone, but rather the referent of ‘God’ is their combination or whole, that is, the divine essence together with the divine attributes.Footnote ¹²

For the Islamic philosophers, ruling out this type of composition is part of the argumentation for DDS-NT. According to them, the main issue with allowing this kind of composition in God is that it contravenes God’s intrinsic necessity of existence (something on which all parties to the debate agree).Footnote ¹³ Thus, given DDS-NT, and the Islamic philosophers’ acceptance of true statements of the form ‘God is F’, for example, God is a knower, God is powerful, and so on, such true descriptions – against proponents of DDC – cannot be taken as being grounded in properties or attributes distinct from, but subsistent by, the divine essence, and with which they metaphysically make up or compose God. But, given that the Islamic philosophers allow for substantive non-equivocal predications of God, how then are such attribute-ascriptions to be construed for them?

To understand the Avicennian account in particular, we need to begin with our attribution of ‘existence’ (wujūd) to God. For the Shaykh, the proposition that ‘God exists’ is demonstrable. Thus, on the Avicennian view, we can truly say ‘God exists’ or that ‘God exists of necessity’ – indeed, that ‘God is the intrinsically necessary being’ (wājib al-wujūd bi-dhātihi). But, it also turns out that God’s (intrinsically necessary) existence must be identical to Himself or His essence. That is, His existence cannot be something additional to His essence; rather, it just is Him. Now existence for the Shaykh is a positive notion; so, describing God as (necessarily) existing is describing Him in a positive way, that is, it is to characterize Him in affirmative terms.

But the Avicennian view also allows for a whole host of other true descriptions of God, for example, that God knows, that He is powerful, and so on. These sorts of predications, however, cannot be taken as being positive in status – that is to say, we shouldn’t understand them as referring to distinct positive realities characterizing God. The Avicennian commitment to DDS-NT forbids this; for, otherwise, divine simplicity would be violated, and consequently divine necessity of existence. Instead, for the Shaykh these other predications we truly make of God are to be understood as either negative or relational descriptions of His (necessity of) existence which, again, is something positive and identical to Him, that is, His essence.

So granting this, we can then say that the Avicennian allows for three kinds of true descriptions of God: positive, negative, and relational. But the positive only has one instance, and it serves as the basis of the negative and relational ones. The general rule or principle to be followed here, which is meant to preserve necessity of existence and unity, is this:

Divine knowledge and power

Let us then consider divine knowledge and power specifically. What do they return to on DDS-NT? According to the Shaykh, the former amounts to a negation, while the latter to a combination of a negation plus a relation. More precisely, divine knowledge returns to the divine essence’s necessity of existence considered with a certain negation, and divine power returns to that essence, but taken together with a negation and relation. Let me now unpack each in turn.

To understand the negational basis for divine knowledge, we first need to grasp what, in general, ‘knowledgeableness’ or the aptitude for knowledge involves on Avicennism.

For the Shaykh, knowledgeableness consists in enjoying a certain mode of existence – namely, an existence that is free of matter and its concomitants. Thus, according to the Shaykh, if an entity, x, exists in a way devoid of matter and its concomitants, then x is knowledge-able, that is, apt for being characterized with knowledge. Now, according to the Shaykh, anything that exists separately from matter and its concomitants exists in one of two ways: either it exists of itself (qā’im bi-nafsih) or not, in which case it exists in something else, y. Accordingly, if, then, x, in addition to existing apart from matter and its concomitants, also exists of itself, then x is not only knowledge-able but also as a matter of fact actually knowing of something – namely, itself. That is, x would be something that enjoys actual self-knowledge. If, on the other hand, x does not exist of itself, but rather exists in another, y, that itself exists of itself (and is separate from matter), then x is not only knowable to y, but is actually known by y. Likewise, y would not only be knowledge-able of x, but would actually know x. Thus, y would be something that has, in addition to actual self-knowledge, actual other-knowledge as well, that is, knowledge of x.

Given this, it is evident how knowledgeableness, on the Avicennian account, is something negative in consideration. For it is the negation of an association, on the part of an existent, with matter and its appurtenances. The Islamic philosophers generally call this ‘abstraction from matter’. Thus, in the divine case, divine knowledge would then be a negation in the following manner: the necessary existence of the divine essence, as unconnected to matter or anything bodily, is apt for truly being ascribed with knowledge – that is, it is, as such, fit for knowledgeableness or ‘knowledge-apt’, whether self-knowledge or other-knowledge. Thus, God, the Necessary Existent, insofar as He exists totally unmixed with matter, is a ‘knower’ or ‘knowing’ (ʿālim):

Hence, according to the above analysis, God’s being a knower does not involve Him possessing an ontologically positive item, that is, a knowledge-property, as additional to His essence. Rather, it consists in just the fact that He enjoys an (intrinsically necessary) existence that is free from matter and its concomitants. And this fact, being a negative item, does not really or positively compose anything with the divine essence, and so does no damage to the necessity and unity of the divine essence.

What about divine power? On the Avicennian view, power (qudra), generally, is the ability to do (something) on the basis of the agent’s wish or volition. So, if an action obtains or does not obtain from an agent because the agent wished, wanted, or willed it, then that agent is rightly said to be capable or powerful. Now divine want or volition, for the Shaykh, is understood as causal efficacy (ta’thīr) – the ability to existentiate something. Divine power, then, returns to divine knowledge in the following way: if it is true that God brings something into existence and fully knows it (claims established on independent grounds), then it is true that He acts voluntarily and so eo ipso it is also true that He is a powerful agent – irrespective of whether or not His volition is mutable (Avicenna Reference Hilāl1981, 28). The Shaykh states:

The qualification in the last sentence is important since, again on independent grounds, God’s knowledge and volition turn out to be immutable according to the Avicennian view. So on that view, the immutability of divine knowledge and volition does not preclude God from ‘power’ being truly predicable of Him. On the basis of the above, then, it is clear how divine power is to be conceptualized: it is the divine essence’s necessary existence considered with a negation and a relation. The negative aspect consists in the divine essence as knowing (in the sense clarified earlier), while the relative one consists in the divine essence as related to all that He voluntarily existentiates – more specifically, to all that He is satisfied to know Himself to bring about or have causal influence over

If then divine power on Avicennian DDS-NT is a negation-plus-relation, then divine power is not, as it is on the DDC view, a property possessed by the divine essence, and subsisting by It. Rather, like divine knowledge, it is just the divine essence, but considered a certain way, though a way different from how divine knowledge is considered. And this negative-relative composite item is nothing factually positive outside the mind considering the divine essence. Understood as such, divine power for the Shaykh, like divine knowledge, does not add some ontologically positive item to the divine entity – and thereby really or objectively, that is, mind-independently, compose something with that essence.

This much suffices for our purposes as far as DDS-NT goes. But here, one might raise a worry: given the above analysis of the divine essence-attributes relation in connection with DDS-NT, how does it make sense to identify the divine essence with negations, or relations, or with both? For if, on DDS-NT as the Avicennian understands it, divine knowledge is a relation and divine power is a negation-relation, then it seems incoherent to say, per DDS-PT, either that ‘His essence is identical to His knowledge’ or that ‘His knowledge is identical to His power’.

To resolve this, we turn to addressing Q2.

Q2

Let me begin by first getting an obstacle out of the way to a proper understanding of DDS-PT in an Avicennian framework. The impediment concerns two ways one might understand the identity claim between the divine essence and attributes, if one is coming from an analytic philosophy background. For, at first blush, one might take DDS-PT to be committed a set of statements like:

1. (i) There exists a property or attribute F,

2. (ii) The divine essence instantiates F, and

3. (iii) The divine essence is identical to F

Or, alternatively, by DDS-PT one might take the Avicennians to say something like:

1. (i) There exists a property or attribute F

2. (ii) F is possessed by the divine essence, and

3. (iii) The divine essence is identical to F

On both views, we have cases of concrete identity. On the first construal, DDS-PT ends up saying that the divine essence both exemplifies, for example, ‘knowledge’ and is the very same as it; and the second construal of DDS-PT ends up saying the divine essence is the very same thing as a property it has or possesses.

But this is decidedly not what the Avicennian philosophers intend by DDS-PT. So both accounts of DDS-PT should be rejected. The first view because that sort of account, aside from the fact that it seems incoherent, presupposes a kind of Platonism about properties, that is, that they exist as abstract objects to which concrete entities bear a certain relation. So ‘God is knowledgeable’ would mean something like: there exists an abstract entity ‘Knowledge’ to which the divine essence bears a relation, that is, that of exemplification or participation or what have you. That’s just a non-starter for the Islamic philosophers; for, outside the mind, they only recognize the existence of non-repeatable, that is, concrete, things. As for the second view, it is equally incoherent, insofar as it identifies two categorically different beings: an inherent feature (a concrete property) with a non-inherent entity (the concrete possessor of it). Again, for the Avicennian philosophers, that’s just a wrong-headed way to think about DDS-PT from the get go. So we should forget any such accounts.

To properly understand the DDS-PT thesis of the Avicennian philosophers, it is best to start with the contrast case, that is, the relation between an attribute, such as knowledge, and an essence, in creatures – specifically, human souls.

Previously, we learned what, on the Avicennian view, suitability for truly being described as ‘knowing’ consists in – namely, a thing’s existing as unmixed with matter and its concomitants. We also saw how, as the basis of our true description, this was something negative or privative. However, on the side of the entity described as existing in that way, there’s something factually positive obtaining for it insofar as it exists in that way, that is, as separate from matter. But, again, this positive fact is not something distinct from, and on the same ontological plane, as it were, as that entity – such as a property subsisting by it. Rather, that positive fact is its immateriality understood as its intellectuality. For, according to the Islamic philosophers, being intellectual is what being immaterial positively consists in. And an aspect of this intellectuality is a certain other positive fact, which is phenomenological in nature, that is, a something it-is-like to be an intellectual entity. And that positive phenomenological fact is that immaterial/intellectual entity’s being in a certain ‘revelational state’ (ḥāla inkishāfiyya), where that state’s what-it-is-like consists in a disclosure or a ‘representation’ (tamaththul), that is, of something, happening to or obtaining for the entity ‘experiencing’ it. For the Shaykh, this revelational state actually characterizes anything that is knowledge-apt, that is, that exists apart from matter. Indeed, it is just what knowledge is, phenomenological speaking. Let us call knowledge so understood ‘knowledge^P’. Thus, on the Avicennian view, if entity, x, exists immaterially, then x is necessarily apt for enjoying knowledge^P, that is, for being in such a phenomenological state – a state in which something is being revealed to x. So the fact that x exists without matter explains why knowledge^P can obtain for x at all.

Now, the obtainment of knowledge^P for x can take place either in virtue of the very essence of x or not. In the latter case, for x to be truly characterized by knowledge^P, something additional to x’s essence is needed. The Avicennians call this additional item a ‘form’ (ṣūra), F. And this item, F, on their view, is also said to be ‘knowledge’, but in a different sense from the phenomenological one. This new sense is a metaphysical sense of knowledge. Let me explain this further.

Consider some human knower, K. According to the Mashshā’īs, when we truly ascribe knowledge to K in, for example, saying ‘K knows’, in one sense we mean knowledge^P, that is, that K knows in the phenomenological sense. Here, we are saying that there’s a certain state of revelation that obtains for K, that is, some revealing is going on for K. As noted, this is a state in which a disclosure (of something) obtains (for K). In this respect, knowledge is the fact of K’s being in a state where something is represented (tamaththul) to K. The Islamic philosophers and their DDC opponents accept knowledge^P; it is non-controversial, a basic datum of experience among them. But, where there’s disagreement is about the underlying nature of this phenomenological fact. And this is where other, more metaphysical, senses of knowledge are distinguished. For the Avicennians, there are two such senses.

Under one metaphysical sense, knowledge^P’s underlying reality consists in a certain feature, F, characterizing K as an attribute. From this point of view, F is an externally existing property of K (on par with any other such property that K might have (e.g., K’s height)), inhering in K as in a subject, and falling under the Aristotelian category of ‘Quality’ specifically. This is the respect in which one says: ‘K has or possesses (an item of) knowledge’, for example, K has mathematical knowledge. Call this metaphysical sense of knowledge ‘knowledge^M1’. The second, metaphysical sense of knowledge, is this F but considered as ‘mental content’. From this perspective, F is not an item on a par, metaphysically speaking, with any other externally existing property of K; rather, it is an item that enjoys a special ontological status in K, that is, a mode of being the Shaykh calls ‘mental existence’ (wujūd dhihnī). This is the respect in which one says: ‘K knows’ that is, some object of knowledge, like some proposition, for example, K knows that a² + b² = c². Call this ‘knowledge^M2’.

With the above distinctions in place, consider Zayd and some mathematical proposition p (e.g., the Pythagorean Theorem). We can ask the following question: what does it take for Zayd to gain knowledge^P of p? That is, what does it take for p to be revealed or represented to Zayd? In particular, does Zayd’s essence suffice for the obtainment of that state of disclosure (of p) or no? Evidently, it does not suffice (never mind for now the reasons for thinking why it does not). Therefore, something additional to Zayd’s essence is needed for p to be revealed to him. What is this additional item?

For the Mashshā’ī philosophers the answer turns out to be: a form, F. Thus, they hold that once this F comes to exist for Zayd, it comes to characterize him as a property or an attribute would (=knowledge^M1), and this property in turn represents or reveals (= knowledge^P) some conceptual content to him, that is, that of p (=knowledge^M2). So the revelational state (=knowledge^P), that is, that Zayd enjoys, is a trace or feature of the property or attribute (=knowledge^M1) that comes to inhere in him. Thus, insofar as Zayd is ‘propertied’ by F, we have knowledge^M1; insofar as there’s a representation of p occurring to Zayd in virtue of F, we have knowledge^P; and insofar as some determinate conceptual content is being revealed to him by F, we have knowledge^M2. The upshot of this is that Zayd is something ontologically ‘needy’ or ‘poor’, that is, dependent on something else, namely F (=knowledge^M1), for actually knowing^P p (=knowledge^M2). Put another way, for phenomenological access to p, Zayd’s essence needs to be metaphysically supplemented with some ontological item, that is, some qualitative property/attribute trope (=knowledge^M1).

However, for the Islamic philosophers, this requirement for accessing p is not due to anything about knowledge^P per se. Rather, it is due to something about the knower, K, that is, Zayd in our case – ultimately, it is the fact that Zayd happens to be a certain kind of existent, that is, a contingent being. Thus, the Islamic philosophers leave it an open question whether there may be K that is not in need of an item like F for p’s disclosure to it. And in fact, for them it turns out that there is such a K, and necessarily so – namely, the Divine knower or God. On the Avicennian view in particular, in God’s case, based on independent considerations about divine necessity of existence, a given object of knowledge – whether, say, Himself or something other than Himself – is represented to God not only (i) without dependence on an external cause but also (ii) without dependence on something that, though not external, is metaphysically non-identical to Him – namely, an attribute or property or a form, F, additional to His essence, subsisting by that essence. In other words, the divine essence, unlike Zayd’s essence, suffices for God’s being knowing, that is, of Himself and other things. To use the example mentioned above, the divine essence suffices for God’s having, and so for our truly ascribing to Him, the trace or feature of F in the normal case, that is, knowledge^P (of p).

Hence, given the above distinctions between the three senses of ‘knowledge’, the DDS-PT thesis, that is, that Divine knowledge is identical to the divine essence can accordingly be taken in three ways:

1. 1. Divine knowledge^P is identical to the divine essence

2. 2. Divine knowledge^M1 is identical to the divine essence

3. 3. Divine knowledge^M2 is identical to the divine essence

From these three, I say that the Avicennian view of DSS-PT must be understood only in way 2. As for ways 1 and 3, to read Avicennian DDS-PT in those senses wouldn’t even be wrong; it would just be straight up incoherent. For if we take it in way 3, we’d be saying that God is identical to (some) mental existent (represented to Him). That makes no sense. And if we take it in way 1, we’d be saying that God is identical to a state of representation (to Him of some mental existent). That makes no sense either. So, the account must be taken in way 2; the reason is because knowledge in that sense, that is, as knowledge^M1, you’ll recall, refers to a metaphysical item, F, of which knowledge^P is a feature. What DDS-PT, interpreted in sense 2, then says is that the divine essence is identical to this metaphysical item, which is to say that, in the divine case, God is identical to that which is the source or ground of knowledge^P. The upshot of this interpretation: the divine essence, on the Avicennian view, turns out to be self-subsistent knowledge^M1.

One might take issue with me here, noting that my rejection of readings 1 and 3 is surprising given that for the Shaykh, in the case of God’s self-knowledge, the intellect/er (i.e., the knower, K), the intellection, and the intelligible are identical. My response is that nothing I’ve said is inconsistent with that identity thesis. For we need to keep in mind that there’s a distinction between the concrete object (that’s known or intelligible) and the mental item representing it. Likewise for the concrete act of intellection and the phenomenology of that act. On the Avicennian view, in self-knowledge – whether human or divine – the intellect/er, that is, K, is identical to the concrete object that’s intelligible, that is, K itself, and to the concrete activity or actuality of the intellection, that is, the being of K itself. But K is not identical to the mental item (= knowledge^M2) represented to K, nor to the phenomenological state, that is, the feature of being revealed/represented (i.e., knowledge^P) to K.Footnote ¹⁴

So if then the divine essence = self-subsistent knowledge^M1, this entails that the divine essence, as self-subsistent knowledge^M1, suffices for having knowledge^P. That is to say, with respect to knowledge^P, God’s essence alone, qua self-subsistent knowledge^M1, plays the metaphysical role that, in the creaturely case, an essence and an attribute additional to it play together. Thus, when Avicennian philosophers say:

This means: the divine essence, as self-subsistent knowledge^M1, all by itself, that is, without the consideration of anything else, achieves that which, in the creaturely case, would only be realized with the conjunction of an essence plus some attribute F, for example, a knowledge-property or form, that’s distinct from that essence. One more time: God, by His essence alone, secures or guarantees knowledge^P of p, that is, the representational or revelational state wherein some mental content (conceptual or propositional) is disclosed (to God). In the creaturely case, on the other hand, the obtainment of knowledge^P of p for Zayd relies on an essence and attribute, F, in combo. Neither Zayd’s essence alone, nor the distinct knowledge-property by itself, that is, without a subject to characterize, can result in p’s disclosure (to Zayd). So, for any p knowable by Zayd, both essence and an attribute, F, must come together for p to be representationally accessed, that is, by Zayd. But not so with God; He, being self-subsistent knowledge^M1 – insofar as He is an intrinsically necessary, immaterial self-subsistent being – knows^P anything He can know^P (Himself and others) purely in virtue of Himself. This is the import of the ‘li-dhātīhi locution in the following two texts. One is about divine self-knowledge specifically, wherein the Shaykh states:

The other text is divine perfection and knowledge generally. It runs as follows:

In light of the preceding, then, if we suppose F (=knoweldge^M1) could exist on its own, then the Avicennian view urges that God’s self-sufficiency for knowledge^P is what F, in the familiar case, would be like if it had aseity as opposed to existing dependently, that is, by means of a subject such as Zayd. For, as the Shaykh says, if a knower like Zayd enjoys an occurrent or actual episode of knowing^P some p only because of knowledge^M1, that is, only when a metaphysical item or form, F, comes to exist in him, then if F could exist on its own, F itself would be more worthy of, and more suitable for, having actual knowledge^P of p than Zayd himself:

On Avicennian DDS-PT, therefore, the divine essence, lacking all the imperfections and limitations (present in Zayd and F) that come with contingency as well as association with matter, is a fortiori or all the more truly said to be ‘knowing’ than them or anything else. In this respect, it is, basically, like a Platonic Form of knowledge.Footnote ¹⁵

The same considerations, mutatis mutandis, apply with respect to the attribute of ‘power’ (qudra). For, on the one hand, ‘power’ can be taken as a metaphysical item (call it power^M), characterizing something, and in virtue of which that thing can exercise causal influence or bring about effects (ta’thīr). On the other hand, ‘power’ can be taken as the very exercise of causal influence itself (call it power^P). Now in the creaturely case, power^P is truly had by something because that thing possesses power^M, that is, as a property or attribute inhering in it. That is, a creature enjoys causal influence (=power^P), which is the trace or effect or feature of power^M, by means of an essence and an attribute-property G (metaphysically) working in tandem. For example, consider the burning (=causal influence) of some cotton. This phenomenon takes place in virtue of (among other things) the fire (an essence) and its property of heat (an attribute) together. Or take the moving of a rock; that phenomenon takes place in virtue of Zayd’s essence-plus-attribute complex (e.g., Zayd with his motive power).

But in the divine case, again on independent considerations establishing DDS-NT, the divine essence by itself suffices for having the causal influence It has, that is, for having power^P – an influence that is only achieved, in the non-divine case, by a distinct attribute of power^M along with the essence it characterizes. Thus, with respect to causal influence, the divine is like what, as the last sentence of T6 above indicates, the power^M to heat would be like if it could subsist on its own. Hence, the divine essence, just as it was self-subsisting knowledge^M1, is likewise self-subsisting power^M. That is, the divine essence is as it were the Platonic Form of power. As such, there’s no reliance, in God’s case, on a property distinct from, and subsisting by, His essence for achieving what power^M achieves, that is, causal influence.

Consequently, when the Islamic philosophers say

They mean:

That is, that the divine essence by itself or without the consideration of anything else, insofar as it is self-subsisting power^M, accomplishes that which, in the creaturely case, would only be realized with the conjunction of an essence and some attribute G, where G is a power-property.

On the basis of the above considerations, then, we can in a general way put the Avicennian DDS-PT thesis, that is, that

like so:

This in turn entails that the divine essence suffices for our truly predicating F’s perfections or traces of It. Then, where F = knowledge^M1, DDS-PT would say:

which, in turn, would mean (where p is among the perfections or traces of F, that is, p is the representational content of F)

In sum, we’d be saying something like: for any p that is knowable^P, that is, representable or revealable, to God, God’s essence suffices for knowing^P p.

And likewise with the attribute of power. On DDS-PT, the identity of God’s essence and power means:

which is to say:

Let it be known that this view does not entail that the divine essence in itself is not literally characterized by knowledge and power, but only metaphorically – as if It just brings them about without the need for anything additional (as in creatures) but in itself it is devoid of them.

This is a wahm (mistaken understanding). To repel it, go back again to the example the Shaykh gives T5–6: that of the self-subsistent heat or intelligible form. Suppose there’s a heat-trope, h, ontologically independent of a substrate. It would be true to say of h – in the strictest and most literal sense in fact – both that, for example, ‘h is hot’ and that ‘h heats (other things)’. Same thing with the intelligible form that self-subsists (were there such a thing): we’d be asserting a truth if we say of it ‘it knows or intellects’, that is, the content it is taken to reveal to the subject in which it normally inheres. In fact, it would be more truly knowing of that content than that subject. Likewise in the divine case, that is, with the divine as self-subsistent knowledge and power. In this respect, God would be a ‘knower’ and ‘powerful’ in the most literal and paradigmatic sense and truly said to be a ‘knower’ and ‘powerful’ in the strictest sense.

One of the later scholars, Mīr Sayyid Jurjānī, likely taking into account passages like T5–8, crisply summarizes the general doctrine of the Islamic philosophers we parsed in some detail above by elucidating their intent behind DDS-PT as follows:

In sum, with regards to both knowledge^P-M2 and power^P, God’s essence alone, that is, without a knowledge^M1 and a power^M superadded to it, does for God what a superadded knowledge^M1 and power^M do for creatures.Footnote ¹⁹

The identity among divine attributes themselves is given the same treatment on the Avicennian account. That is, their identity is to be understood in terms of a reduction; God’s power^M returns to, or just is, God’s knowledge^M1, which itself returns to, or just is, God’s essence. Hence, divine power^M returns to, or just is, the divine essence. As result, on Avicennian DDS-PT, the identity of divine knowledge and power entails that:

Or more specifically, that:

In other words, x’s being represented or revealed to God as subject to His causal influence guarantees God’s actual causal influence over x, that is, His actually bringing it into being.

What DDS-PT does not require is that the divine attributes be identical in the sense that the aspects in which the divine essence can be considered by our minds are identical, so that, for example, the negative aspect qua which there’s a disclosure, that is, to the divine essence, is identical to the relational aspect qua which there’s an enactment, that is, by the divine essence. For this would be to identify considerations of the mind, which is nonsensical. Rather, since the divine essence is full and pure disclosure and full and pure causal influence (in the way explained earlier), then, in the divine case, in a real sense, that is, metaphysically, divine knowledge^M1 (= the divine essence) is the same as divine power^M (= the divine essence). Consequently, in God’s case, an epistemic state (= knowledge^P) is or guarantees a causal state (= power^P); for, as T8 says, the divine essence’s cognizing is the explanation of the divine essence’s agency. Thus we have here in the divine case a single metaphysical entity that is identically (maximally) cognitive and (maximally) capable:

What obtains in the divine case shouldn’t be so incredible to us (though it is quite extraordinary). For we can relate to it in some sense. For consider a given piece of practical knowledge one might have, for example, of some craft or art. This item is both a piece of knowledge and a power or disposition, that is, to effect or influence something. It is said to be ‘know-how’ precisely in that respect; it is both an epistemic state and a capacity to act (in a given way) or execute. In the divine case, this is also true, but of course without any of the usual limitations. Among these is the following: in the creaturely case, merely knowing^P something (and so merely having knowledge^M1-M2) is not enough for bringing that thing into existence in actuality. For we depend on a whole host of things for realizing the (causal) influence required for existentiation, among them being a distinct power^M, that is, attribute-property inhering in us. The Shaykh explains the contrast between Creator and creature in this regard like so:

According to the Islamic philosophers, the above account of DDS-NT and DDS-PT has the consequence that in conceiving God, that is, as the Necessary Existent, the notions of being knowing or being powerful (and other attributes) are extractable from, and truly predicable of, Him in virtue of His essence alone. For the divine essence, as the Necessary being, has no positively objective composition; it is a monolithic, metaphysically structure-less being. But this entity can be beheld in various ways by us, where these ways amount to distinct conceptual considerations (of the divine essence) that serve as the basis for our making true distinct predications of that entity. And these distinct beholdings arise from our mentally comparing or relating this Necessary Existent to something else, either by way of negation, affirmation, or both. So for example, when we see that the simple divine essence exists in a way unmixed with anything material, we negate a relation to matter from that essence, and this serves as the basis for our extracting a notion from that divine essence (which notion we call ‘knowledge’) and for then (truly) predicating that notion of It. And when we relate the simple divine essence to some other things, along with a consideration of Its knowledge and volition (notions we’ve already extracted), we extract another notion from It (which notion we call ‘power’) and for then (truly) predicating that notion of It. And so on.

Thus, as a consequence, when we truly predicate knowledge or power of the divine essence the preceding analysis requires that what makes such statements true cannot be, as on alternative views like DDC, the something other than the divine essence taken together with It – like an intrinsic attribute or property, F, additional to, or subsisting by, It. Rather, what grounds the truth of those judgments of ours would just be a single entity, that is, the divine essence alone, albeit beheld in various ways – either relationally, negationally, or both. In other words, God’s essence by itself, metaphysically speaking, though objectively considered a certain way, mentally speaking, serves as the truth-maker of such predications. This is unlike in the creaturely case, where, again, what makes it true that ‘Zayd is knowledgeable’ or ‘Zayd is powerful’ is not Zayd himself (i.e., his essence alone) but rather Zayd together with something else metaphysically different from his essence, that is, a qualitative property (we call ‘knowledge’ or ‘power’) inhering in him.Footnote ²¹

Conceptual identity of divine attributes

The Avicennian account goes further. For in addition to the extensional identity of the divine attributes just discussed, that account is also committed to their conceptual sameness. That is, not only do the notions of knowledge and power, when predicated of God, have a single referent and truth-maker (miṣdāq), as we just saw, but such notions, when ascribed to God, also have a single understood content (mafhūm). Thus, according to the Shaykh, even the determinate objective content (maʿnā) one conceives from ‘divine knowledge’ is the same as the maʿnā one conceives from ‘divine power’ (and likewise with other attributes of perfection). He states:

How are we to understand this conceptual sameness thesis regarding the divine attributes?

Note first that the Shaykh does not say ‘power’ as such and ‘knowledge’ as such are the same in their formal accounts (maʿānī). Rather, he says it is power and knowledge as they pertain to the divine essence – hence divine power and knowledge – that have the same maʿnā. Thus, the identification is between the natures of power and knowledge as qualified, that is, by a relation to the divine essence, not between the two absolutely taken. So yes, knowledge and power considered in the abstract, that is, apart from the concrete entity they characterize, do not have the same maʿnā. But as descriptors of the divine essence, that is, with respect to the divine essence as what they describe (mawṣūf), they cannot, given DDS-NT, have distinct maʿānī. For according to the Shaykh, distinct understood contents (mafhūmāt) in the mind entail distinct extra-mental realities to which those notions correspond. Thus, if the understood content of F is different from that of G, then the extra-mental counterpart of F is different from that of G (Avicenna Reference Bidarfar1992, 112). Hence, if the divine essence, as the arguments for DDS-NT show, is absolutely simple, which is to say lacking all real composition or multiplicity, then different notions, that is, revealing really distinct realities, aren’t extractable by the intellect from that essence alone (cf., Zabihi Reference Zabihi2005, 41). Otherwise, there would be, as a matter of fact, some layer of positively real or objective composition embedded within the divine essence prior to the intellect’s reflection on it, serving as the basis of its extractions. And in that case, we don’t really have DDS-NT anymore.

The intellect’s different beholdings of the divine essence, however, is not inconsistent with the objective simplicity of that essence. That is because, for the Shaykh, there’s a distinction between:

1. (i) F and G are different in meaning (maʿnā)

and

1. (ii) F and G are different in consideration (iʿtibār)

On the Avicennian view, a maʿnā is just some semantic content that is represented by a concept. As such, it is not necessarily mind-dependent, a mental item.Footnote ²² An iʿtibār, on the other hand, is a mental act. As such, it is necessarily mind-dependent.

Thesis (i) holds when F and G can be understood independently of each other. That is, if you can understand F without understanding G, then F and G are different in their maʿnā. For if they can be understood one without the other, this requires that they have distinct conceptual content, which in turn correspond to distinct realities. Otherwise, that is, if the two can’t be so understood, then F and G are different only in consideration. In this case, their positive conceptual content is shared – the difference between them may be that we get F when that single positive content is taken with say a relation, and G with a negation. Further conceptual analysis might then reveal certain, mind-independent priority-posteriority relations between those conceptual considerations (iʿtibārāt). Thus, with respect to the issue at hand, for the Shaykh to say that the divine attributes are also the same conceptually is to say that they have a single, positive conceptual content in common. And that happens to be: (intrinsic) necessary existence. In other words, if we were to prescind from the negative/relational features of F and G (as they’re predicated of God), we would see that they’re identical in maʿnā, that is, having one, positive conceptual content in common.Footnote ²³

Given the preceding points, what is disallowed by the Avicennian in the divine case then is (i), but not (ii). This is because, for the reasons mentioned, the former entails objective, positive composition, but not the latter. Accordingly, on the Shaykh’s version of DDS, the divine essence can be looked at in different ways (e.g., now as knowing, now as powerful, etc.), but this need not be due to the derivation of distinct maʿānī from it. What it involves instead is beholding different priority and posteriority relations within a single understood content (maʿnā) – namely, the divine essence’s being a necessary existent. One first beholds it as a (intrinsically) necessary existent, one then views that as knowing, then as powerful, and so on. These considerations have an objective order and, as noted in T1 by the Shaykh, they all amount to that necessity of existence taken with either some relation, or a negation, or both. So understood, none of them impugns the necessity or simplicity (in sense DDS-NT) of the divine essence. For such beholdings of the divine essence, that is, as intrinsic necessary existence, are not, as noted, reflective of different maʿānī.Footnote ²⁴

So, given DDS-NT, the divine essence, qua the intellectual revelation of something to Itself – whether that thing is its very self or something else – is said to be ‘knowing’: ‘[T14] [His knows] in that He intellectually grasps His essence and what His essence necessitates of the modality of the being of the good in the all, …’. (Avicenna Reference Nurani1984, 20).

And this ‘knowledge’-consideration is negational, that is, it is based on taking the divine essence, as the necessary existent, without any connection to matter and its concomitants. As for divine power, that is the divine essence but qua what is revealed to It as admitting Its causal influence, that is, being existentiable by It: ‘[T15] The power that He enjoys is His essence being an intellecter of the all, an intellection that is the [causal] source of the all, not derived from the all, and a [causal] source in virtue of itself, not dependent on the existence of anything else, …’. (Avicenna Reference Nurani1984, 21).

This ‘power’-consideration is a negation-relation composite, that is, it is based on taking the divine essence as associated with that over which It knows Itself to exercise causal influence. The same reasoning, mutatis mutandis, apply to the rest of the divine attributes. In every case, the conceptual content remains the same, that is, the divine essence as the necessary existent; what differs is the respect in which that content is beheld by the mind – now in a way termed ‘knowing’, now in a way termed ‘powerful’, and so on. The Shaykh concludes the account:

Conceptually, that is, in terms of maʿnā, knowledge and power, as said of the Necessary Existent, are, on independent grounds, the same. Under this consideration, we’re talking about knowledge and power as they extra-mentally exist, that is, as descriptions of some subject. But, knowledge and power generally taken, that is, apart from the specific subject to which they’re applied, are not conceptually the same. Under this consideration, though, we’re talking about knowledge and power as they exist only in the mind, since general things are mind-dependent. Given certain facts about the nature of the subject we apply them to, our application of them to that subject will have to be qualified. For such terms are ambiguous (mushakki/ak), not univocal or equivocal. Hence the difference between the divine case and the creaturely case is in our respective predications.

Alright that much should suffice by way of clarifying DDS on the model of the Islamic philosophers in general, and the Avicennian model in particular. Let me now turn to briefly addressing some objections leveled against that account by proponents of DDC.

Objection 1

One of them goes like this: if the divine attributes of knowledge and power were the same as the divine essence, the two attributes would be identical with each other. And if that were so, then whatever one is true of so is the other. In that case, the divine essence would be subject to the divine power; for the divine essence knows itself. This is problematic; for it would mean the divine power has a causal influence on the divine essence, that is, that it brings the divine essence into being. But clearly, the divine essence, being intrinsically necessary, isn’t subject to divine power. Hence, the range of the two attributes is different. But if so, then the two are not identical. And thus, contra DDS, the two also aren’t identical to the divine essence either.

This argument comes from the Asharite theologian ʿAbd al-Qāhir al-Baghdadi in his book on the principles of religion. He writes:

Some contemporary opponents of DDS who are moved by this argument formulate it as follows (Muhtaroglu Reference Muhtaroğlu2020, 4–5):

1. 1. God’s essence is identical with divine power and knowledge. (divine simplicity thesis)

2. 2. Divine power is identical with divine knowledge. (transitivity of identity)

3. 3. The logical scope of divine power is identical to the logical scope of divine knowledge. (corollary of 2)

4. 4. God Himself or His essence is included within the scope of divine knowledge. (by definition of divine knowledge)

5. 5. God Himself or His essence is included within the scope of divine power. (3, 4 the principle of the indiscernibility of identicals)

6. 6. God Himself or His essence cannot be included within the scope of divine power (otherwise it goes against divine eternity and perfection)

7. 7. Divine simplicity thesis is false. (5, 6 contradiction)

Response

In light of the distinctions drawn in the preceding, it should first be noted that the identity claim in 1 is to be understood as ‘sufficient for’ in the way explained earlier. Accordingly, premise 1 should be parsed in one of two ways. Either saying:

Or as

So if we take 1 in sense 1*, and maintain the same senses of knowledge and power throughout the rest of the argument, it would read:

1. 1*. God’s essence is identical with divine power^M and knowledge^M1 (divine simplicity thesis)

2. 2*. Divine power^M is identical divine knowledge^M1. (transitivity of identity)

3. 3*. The logical scope of divine power^M is identical to the logical scope of divine knowledge^M1. (corollary of 2)

4. 4*. God Himself or His essence is included within the scope of divine knowledge^M1. (by definition of divine knowledge)

5. 5*. God Himself or His essence is included within the scope of divine power^M. (3, 4 the principle of the indiscernibility of identicals)

6. 6*. God Himself or His essence cannot be included within the scope of divine power^M (otherwise it goes against divine eternity and perfection)

7. 7*. Divine simplicity thesis is false. (5, 6 contradiction)

On this reading, premise 2* of the argument in 1*–7* is true. For what it states is something like: that in virtue of which God is causally efficacious is the same as that in virtue of which God is revealed to, that is, the divine essence (= God). However, as for premises 3*–6*, they make no sense, and so aren’t even truth-apt (so as to be considered false). For, on this reading divine power^M and knowledge^M1 is a concrete entity, that is, the divine essence. And for the Islamic philosophers, ‘logical scope’ is a property of concepts (mental beings). Thus, it is a category error to ascribe logical scope to a concrete entity. Hence, the argument against DDS in 1*–7* fails.

Alternatively, we can take 1 in sense 1**. Then, again keeping the same senses of knowledge and power throughout, we get:

1. 1**. God’s essence is identical with divine power^P and knowledge^P (divine simplicity thesis)

2. 2**. Divine power^P is identical divine knowledge^P. (transitivity of identity)

3. 3**. The logical scope of divine power^P is identical to the logical scope of divine knowledge^P. (corollary of 2)

4. 4**. God Himself or His essence is included within the scope of divine knowledge^P. (by definition of divine knowledge)

5. 5**. God Himself or His essence is included within the scope of divine power^P. (3, 4 the principle of the indiscernibility of identicals)

6. 6**. God Himself or His essence cannot be included within the scope of divine power^P (otherwise it goes against divine eternity and perfection)

7. 7**. Divine simplicity thesis is false. (5, 6 contradiction)

But on this reading, 1** and 2** are unintelligible. If so, 3** can’t follow as a corollary from 2**. Thus, this argument, too, fails.

One might balk at the specifications I’ve made to ‘identity’ and ‘power’ and ‘knowledge’ when explicating the ḥukamā’s version of DDS in dealing with the above objection. Why is that an issue though? I mean, suppose someone opposes the theologians (mutakallimīn) by urging the following piece of reasoning:

The minor premise of this argument is based on Quran 6:19, while its major comes from Quran 11:4. The conclusion says that God falls within the scope of divine power. But, they would agree, that is inconsistent with God’s perfection.

How would the Asharite handle this argument? Plausibly, he may deal with it by specifying the middle term, that is, ‘thing,’ by saying something like: ‘in the major it means “the incipient thing” (ḥādith) or “the contingent thing” (mumkin), but in the minor it means something else’. Then he’ll specify divine power and its scope accordingly.

But if so, the DDS proponent is then well within his epistemic rights to do the same, that is, make relevant specifications, that is, by, for example, distinguishing various senses of (divine) knowledge and knowledge (in dealing with Baghdadi’s argument in T1). Nothing about doing that per se is excluded by a commitment to DDS, and especially not when DDS is properly understood. And what the preceding analysis has really shown is that the objection fails to even address the DDS Islamic philosophers defend, since it fails to properly understand it insofar as its premises fail to accurately reflect their distinctive theory.

One might push back again, this time rephrasing the initial objection in terms the Avicennian just employed in clarifying his own account, as follows:

1. i*. God’s essence is identical with divine power^M and knowledge^M1 (DDS)

2. ii*. Divine power^M is identical divine knowledge^M1. (transitivity of identity)

3. iii* God’s knowledge^P is sufficient for God’s power^P (the entailment of 1* and 2* as the author states)

4. iv*. God’s knowing^P that He is powerful^P over x suffices for God’s causing x to exist (=power^P) (the entailment of 3* as the author states)

5. v*. God Himself or His essence is included within his knowledge.

6. vi*. God’s knowledge of himself is sufficient for God’s power^P over himself. (3*, 5*)

7. vii* God’s knowing^P that He is powerful^P over himself suffices for God’s causing himself to exist (=powerP) (5*, 6*)

8. viii*. God causes himself to exist.

9. x*. 1* is false.

Response

First, I didn’t say iv* is an entailment of iii*. I said it is a specification of iii*. That is, iv* is a precisification of what iii* states. Second, if that is so, then v* should be read in light of that specification. And when read in its light, v* would really say something like:

And what this means, given what I said earlier, is that God’s essence/God Himself is revealed to God as subject to His causal influence. Thus, God’s essence or Himself being (as v* has it) ‘included within His knowledge^P’ would mean, given iv*, that God knows^P Himself/His essence as subject to His power^P, that is, as subject to His causal influence. But if so, I don’t grant v** at all; for it is decidedly false. To be sure, God of course (I maintain) knows^P Himself, but definitely not as subject to His causal influence. The argument that is laid out in i*–ix* is, therefore, unsound.Footnote ²⁶

Objection 2

The other objection comes from Taftāzānī in his Sharḥ al-ʿAqā’id. There, we get the following, terse ilzāmī argument against DDSFootnote ²⁷: ‘And it follows upon youFootnote ²⁸ … that the Necessary is not self-subsistent, … (Taftāzānī Reference Taftazani2020, 168).

When unpacked, the line of reasoning in it is said to go like this (Muhtaroglu Reference Muhtaroğlu2020, 5–6):

1. (1) Divine attributes are identical with God’s essence. (divine simplicity thesis)

2. (2) Divine attributes are dependent beings. (since all attributes are dependent)

3. (3) God is a dependent being. (1, 2 the principle of the indiscernibility of identicals)

4. (4) God is independent of anything. (by definition of God)

5. (5) Divine simplicity thesis is false. (3, 4 contradiction)

Response

There are several ways to address this argument. The one we opt for here, in light of the previous considerations, goes as follows: the DDS proponent would reject (2) as false. For the divine attributes, if DDS is true, are not dependent beings – they are just the divine essence, in the sense specified preciously. Thus, given DDS, they would be independent beings but identical to God. Hence, the argument in (1)–(5) simply begs the question with that premise. This is because that premise already presupposes that attributes (whether divine or not) are additional to the subject they characterize – insofar as they are dependent on it. But, again, that’s precisely what DDS denies in the divine case. For the Islamic philosophers, the additionality or identity of an (divine) attribute is no more part of its definition as is its being originated or eternal. And were the DDS proponent to grant (2), he’d simply be giving up DDS thereby (and there’d be no need to go through the rest of the argument). Hence, it seems to me the only way (3) would follow is if the DDS proponent grants (2), which is to say the only way (3) would be forced upon him is if he holds, inconsistently, that the attributes are somehow both identical (i.e., per (1)) and also additional (i.e., presupposed by (2)). But, no proponent of DDS holds that, nor would they do so. Hence, I conclude that Taftāzānī’s argument is either unsound or question-begging.