Al-Ghaza¯lı¯’s Philosophical

in the outside world (   f ı¯ l-a  ¶ya¯n ) removes an important difference: whether pos-

sibility and necessity exist in things outside of our mind, or whether they are 

simply predicates of our judgment. Al-Ghaza¯lı¯’s critique of Avicenna’s under-

standing of the modalities was anticipated by al-Juwaynı¯’s notion of necessity 

and possibility in his proof of God’s existence in the  Creed for Niz.a¯m al-Mulk . 

Al-Juwaynı¯ begins his argument there with an explanation of the modalities. 

Every sound thinking person fi nds within himself “the knowledge about the 

possibility of what is possible, the necessity of what is necessary, and the im-

possibility of what is impossible.” 

134

  We know this distinction without having 

to study or make further inquiry into the world; it is an impulse ( badı¯ha ) of our 

rational judgment (  ¶aql ). 

 The impulsive possibility that the intellect rushes to apprehend 

without [any] consideration, thinking, or inquiry is what becomes 

evident to the intelligent person when he sees a building. This [ scil.  

the building] is [simply] a possibility that comes into being ( min jawa¯z 

h.udu¯thihi ). He knows decisively and offhand that the actual state 

( h.udu¯th ) of that building is from among its possible states (   ja¯ 7iza¯t )  and 

that it is not impossible in the intellect that it had not been built. 

135

  

 The intelligent person ( al- ¶a¯qil , here meaning a person with full rational capac-

ity) realizes that all of the features of the building—its height, its length, its 

form, and so forth—are actualized possibilities that could be different from 

what they are. The same possibilities apply to the time when the building is 

built. We immediately realize, al-Juwaynı¯ says, that there is a synchronic alter-

native state to the actual building. This is what we call, contingency ( imka¯n ). 

Realizing that there is such an alternative is an important part of our under-

standing: “The intelligent person cannot realize in his mind anything about the 

states of the building other than through a comparison with what is contingent 

like it ( imka¯n mithlihi ) or what is different from it ( khila¯fi hi ).” 

136

  

 Knowledge about the modalities is “on an impulsive rank” ( bi-l-martaba al-

badı¯ha ), meaning it is  a priori : it cannot be derived from any other prior knowl-

edge. 

137

  This statement is limited to the modalities when they are considered 

by themselves. Al-Juwaynı¯ realizes that God’s creative activity makes all the 

unrealized possibilities impossible. If considererd on its own, the actual move-

ment of the celestial spheres ( afl a¯k ) from east to west could be imagined dif-

ferently. The intellect can imagine that the spheres could move in the opposite 

direction. Studying the movements in heaven, however, leads to the realization 

that this possibility is not actualized. 

 Al-Juwaynı¯ understands possibility as synchronic alternative states to what 

actually exists. This is different from Avicenna’s understanding of possibility 

1 7 2   a l - gh a z a

¯ l 1

¯ ’ s   ph ilosoph ic a l   t h e olo g y

and necessity as modes of actualized beings. It also shifts the perspective of 

the modalities away from what exists in actuality toward what is considered al-

ternative states in the human mind. Al-Ghaza¯lı¯’s critique of Avicenna’s modal 

theory is in no way haphazard but is an outcome of long-standing considera-

tion of modalities developed in Ash ¶arite  kala¯m . 

138

  

 What Does al-Ghaza¯lı¯ Mean When He Claims That 

Causal Connections Are Not Necessary? 

 Once Avicenna’s and al-Ghaza¯lı¯’s differing understandings of the modalities 

are applied to the initial statement of the seventeenth discussion, they change 

the established meaning of this passage. When al-Ghaza¯lı¯ says that “accord-

ing to us (  ¶

indana¯ ),” the connection between any given effi cient cause and its 

effect is not necessary, 

139

  he aims to point out that the connection  could   be  dif-

ferent, even if it never will be different from what it is today. For Avicenna, the 

fact that the connection never was different and never will be different implies 

that the connection is necessary. Not so for al-Ghaza¯lı¯. His understanding of 

modal judgments does not require that any given causal connection was dif-

ferent or will be different in order to be considered possible and not necessary. 

The possible is that for which the human mind can perceive an alternative state 

of affairs. For al-Ghaza¯lı¯, the connection between a cause and its effect is pos-

sible—or, to be more precise: contingent ( mumkin )—because an alternative to 

it is conceivable in our minds. We can imagine a world in which fi re does not 

cause cotton to combust. Or, to quote the second sentence of the initial state-

ment of the seventeenth discussion: 

 It is within divine power to create satiety without eating, to create 

death without a deep cut ( h.azz ) in the neck, to continue life after 

 having received a deep cut in the neck, and so on to all connected 

things. The  fala¯sifa  deny the possibility of [this] and claim it to be 

impossible. 

140

  

 Of course, a world in which fi re does not cause combustion in cotton would be 

radically different from the one in which we live. A change in a single causal 

connection would likely imply that many others would also change. Still, such a 

world can be conceived in our minds, which means it is a possible world. God, 

however, did not choose to create such an alternative possible world. He chose 

to create this world among alternatives. 

 In the initial sentence of the seventeenth discussion, al-Ghaza¯lı¯ argues 

against two types of adversaries. First, he argues against those who hold that a 

causal connection is necessary by itself. This group includes people who claim 

that any given proximate effi cient cause is an independent effi cient cause (   fa¯  ¶il ) 

of its effect. This group also includes some natural philosophers who reject sec-

ondary causality as well as the Mu ¶tazilites, who argue that humans create their 

actions and the immediate effects of them. Al-Ghaza¯lı¯, however, makes a clear 

 

t h e   s e v e n t e e n t h   disc us sion   of  

THE

 

INCOHERENCE

  1 7 3

distinction between the teachings of the Mu  ¶tazilite and those of the Avicennan 

 fala¯sifa . 

141

  The Avicennan  fala¯sifa  are the second group of adversaries in the 

seventeenth discussion. Although al-Ghaza¯lı¯ does not argue against the idea 

of secondary causality in Avicenna, he does reject Avicenna’s teaching that the 

connection cannot be any different from what it is. Being contingent by itself, 

according to Avicenna, the connection between cause and effect is necessary 

on account of something else, namely, God’s nature. God’s nature cannot be 

conceived any differently from what it is. For Avicenna, there can be no world 

alternative to the one that exists. 

 In the initial statement of the seventeenth discussion, al-Ghaza¯lı¯ also 

claims that “the connection [between cause and effect] is due to the prior de-

cision ( taqdı¯r ) of God.” 

142

  When he objects to Avicenna and states that these 

connections are not necessary, al-Ghaza¯lı¯ wishes to express that God could 

have chosen to create an alternative world in which the causal connections 

are different from those of this world. Al-Ghaza¯lı¯ upholds the contingency of 

the world against the necessitarianism of Avicenna. For al-Ghaza¯lı¯, this world 

is the contingent effect of God’s free will and His deliberate choice between 

alternative worlds. 

 While rejecting this necessitarian element in Avicenna’s cosmology, al-

Ghaza¯lı¯ does not object to the philosopher’s concept of secondary causality. 

Of the two pillars in Avicenna’s cosmology—secondary causality and neces-

sitarianism—al-Ghaza¯lı¯ rejects only the latter. In the First Position of the sev-

enteenth discussion, al-Ghaza¯lı¯ uses secondary causality to refute the view that 

proximate causes are independent effi cient causes. In the Second Position, he 

offers two alternative explanations (“approaches”) of prophetical miracles, the 

fi rst based on occasionalism, the second, on secondary causality and the exist-

ence of natures ( t.aba¯  7i ¶  ). In all this discussion, al-Ghaza¯lı¯ says nothing about 

whether God actually breaks his habit, meaning the existent laws of nature, 

when creating the prophetical miracle. For al-Ghaza¯lı¯, the connection between 

the cause and its effect is contingent even if God never changes His habits. 

The sole possibility of His breaking His habit—that we could conceive of God 

breaking His habit—or just the possibility that He could have arranged the 

laws of nature differently means that any individual connection between two 

of His creations is not necessary. Although it is conceivable and therefore pos-

sible that God would break his habit or intervene in the assigned function of 

the secondary causes, an actual break in God’s habit is not required for the con-

nections to be contingent. 

This page intentionally left blank 

 

 7 

 Knowledge of Causal 

Connection Is Necessary 

 In the seventeenth discussion of the  Incoherence , is there a consis-

tent line of argument with regard to causality? After proposing his 

most radical epistemological criticism in the First Position—that 

sense perception does not lead to necessary judgments—al-Ghaza¯lı¯ 

presents in the Second and the Third Positions two alternatives to the 

Avicennan model of metaphysics and physics. In the First Approach 

of the Second Position, occasionalism is contrasted with the deter-

ministic cosmology of his opponents. Al-Ghaza¯lı¯ aims to show that a 

congruent occasionalist model can be a viable alternative to Avicen-

nan metaphysics. He implicitly claims that the  fala¯sifa  can accept this 

model and still continue to pursue the natural sciences. The “laws 

of nature” that, according to the  fala¯sifa,  govern God’s creation may 

be understood as habitual courses of action subject to suspension, 

at least in principle. Our human experience, however, has shown us 

that God does not frivolously break His habit. This insight allows us 

to equate God’s habit with the laws of nature, for all practical pur-

poses. In the natural sciences, we study God’s actions and reformu-

late their habitual course into laws that we justifi ably consider, if not 

necessary, at least stable, unchanging, and permanent. 

 In the Third Position, al-Ghaza¯lı¯ puts up a far less radical 

alternative to Avicennan metaphysics and natural sciences. Although 

not clearly explicated, this theory appears to be a slightly altered ver-

sion of Aristotelian physics. This physical theory postulates that in ad-

dition to the rules of logic, God cannot violate laws of nature that rely 

on the relationships of implications. Such implications are usually 

formulated in defi nitions. Will is defi ned as existing in a being that 

has knowledge, for instance, and knowledge is defi ned as existing in 

a being that has life. God therefore cannot create will in a being that 

1 7 6   a l - gh a z a

¯ l 1

¯ ’ s   ph ilosoph ic a l   t h e olo g y

is lifeless. Equally, God cannot “change the genera” ( qalb al-ajna¯s ),  meaning 

that He cannot transform a material body into an immaterial being and vice 

versa. Al-Ghaza¯lı¯ was certainly aware that these three conditions limit God’s 

omnipotence signifi cantly. He here lists what can be viewed as the unchange-

able essence of God’s creation. And although the laws of nature from among 

this core group cannot be altered once creation unfolds, God reserves the power 

to alter others of His habits, such as making water fl ow uphill or creating life in 

any given material object, such as a stick. 

 These two alternative theories to Avicenna’s cosmology frame a passage of 

roughly two pages, which, to the Avicennan, forms the most persuasive part 

of the seventeenth discussion. In addition to these two alternative cosmologi-

cal theories (alternative to Avicenna’s cosmology), al-Ghaza¯lı¯ defends a slightly 

modifi ed Avicennan explanation of causal connections in the Second Approach 

of the Second Position. Here, al-Ghaza¯lı¯ is willing to accept that chains of sec-

ondary causes connect every event in creation with the creative activity of the 

creator. In this part of the seventeenth discussion he clearly accepts the exist-

ence of “natures” ( t.aba¯ 7i ¶  ). He requires the Avicennan simply to acknowledge 

that we lack exhaustive knowledge of the full possibilities of these natures. They 

might allow causal connections that we have not yet witnessed. The miracles 

reported in revelation have causes unknown to us. They are not true miracles 

but mere marvels. 

 In the  Incoherence,   al-Ghaza¯lı¯ presents what might be called a nominalist 

criticism of the modalities, in some sense a criticism of human judgments as 

a whole. Using the parlance of Avicenna, al-Ghaza¯lı¯ implicitly asks whether we 

can know that any given object that we witness in the outside world is possible 

by itself ( mumkin bi-dha¯tihi ) and at the same time is necessitated by something 

else ( wa¯jib bi-ghayrihi ).  Al-Ghaza¯lı¯ rejects Avicenna’s assumption that modali-

ties exist in the outside world. This rejection goes to the heart of the Avicen-

nan ontology that regards potentiality as a paradigm that strives to actualize 

itself. Like Avicenna, al-Ghaza¯lı¯ views human knowledge as a conglomerate 

of judgments. 

1

  He agrees with Avicenna that true knowledge is congruent to 

the outside world and describes it as such. For Avicenna, however, there can be 

only one true explanation of any given phenomenon in the world. True human 

knowledge describes the necessary and only way the world is constructed. Dem-

onstration ( burha¯n ) is the best means to achieve such correct knowledge about 

the world. Where demonstration is not available, humans choose less perfect 

means of acquiring knowledge. Al-Ghaza¯lı¯ agrees with Avicenna on the imper-

fect nature of these means. He realizes, however, that where demonstration can-

not be achieved, multiple explanations are compossible, that is one explanation 

may coexist with another without needing to decide which applies. The inability 

to demonstrate the unchanging nature of the connection between cause and 

effect creates a situation in which more than one explanation of causal connec-

tions is viable. Only a nominalist position toward human knowledge allows the 

assumption of two different explanations of a given process as compossible. 

 Al-Ghaza¯lı¯’s nominalist critique of Avicenna is an important element in 

the understanding of his cosmology. We must point out that al-Ghaza¯lı¯ was not 

 

k now led ge   of   c a usa l   c on ne c t ion   is   ne ce s sa ry  

1 7 7

a nominalist in the sense of his contemporary Roscelin (d.  c.  1120) or William 

of Ockham (d. 1347) in the Latin West. 

2

  These nominalists outspokenly denied 

any ontological coherence between things and their formal (and universal) rep-

resentations in our minds. In the Latin dispute about the status of universals—

a dispute that lasted from the late thirteenth to the end of the fourteenth 

 centuries—the nominalist criticism was directed against the Aristotelian claim 

of an eternal and invariant formal level of being that shapes both the individual 

things in the outside world as well as our knowledge of them. This position, 

which is known as epistemological realism, essentially maintains that individ-

ual things are what they are because of real existing universals. The consistency 

of our knowledge with the outside world is due to the ontological coherence 

between the two. Human souls have access to these universals, and their ap-

prehension constitutes our knowledge. In the Latin West, Avicenna was one of 

the most important proponents of the realist position. 

 In the Muslim East, the parameters of the dispute on the status of univer-

sals were different. Here, the nominalist criticism of Avicenna developed from 

Ash ¶arite occasionalism, as in the case of al-Ghaza¯lı¯. Yet nominalist positions 

were not unknown within the discourse of  falsafa  in the East. Justifying his 

position that the modalities exist only in minds and not in the outside world, al-

Ghaza¯lı¯ cites a moderate nominalist view toward human knowledge that were 

current among the  fala¯sifa . He tries to persuade his philosophical readers to ac-

cept his position on the modalities by comparing them to universals. According 

to views held by the  fala¯sifa  themselves, al-Ghaza¯lı¯ continues, the universals are 

just concepts in the mind without referring objects ( ma ¶lu¯ma¯t ) in the outside 

world. The universals do not exist in the outside world: 

 What exists in the outside world (  f ı¯ l-a ¶ya¯n ) are individual  particulars 

that we perceive with our senses and not in our mind. But they are 

(only) the cause; because the mind abstracts from them  intellectual 

judgments that are empty of matter. Therefore being a color 

( lawniyya ) is a single judgment ( qad.iya ) in the mind (  ¶aql ) similar to 

 blackness or whiteness. One cannot conceive that there exists a color 

that is neither black nor white nor any other of the colors. In the 

mind there exists the form of “being a color” without any details; and 

one says it is a form and it exists in the minds and not in the outside 

world. 

3

  

 The position referred to here needs not be that of a nominalist. Avicenna him-

self taught that the perception of individual objects cannot lead to universal 

judgments. 

4

  Although admitting that universals have no existence in matter, the 

Avicennan opponent still holds that they exist in a real and immaterial way in 

the active intellect, outside of the human mind. Al-Ghaza¯lı¯ uses this argument, 

however, to advance a distinctly nominalist critique of the position that modali-

ties exist outside of the human mind. We will later see how al-Ghaza¯lı¯ made 

productive use of some nominalist tendencies within Avicenna’s œuvre. 

5

  

 In the methodological introduction to  The Highest Goal in Explaining the 

Beautiful Names of God,   al-Ghaza¯lı¯ develops a distinctly nominalist theory of 

1 7 8   a l - gh a z a

¯ l 1

¯ ’ s   ph ilosoph ic a l   t h e olo g y

semantic relations that combines Ash ¶arite notions with philosophical distinc-

tions. 

6

  It is also apparent, however, that the infl uence of Avicenna’s realist epis-

temology on him was so strong that he often applies to his own writings a realist 

concept of the universals. 

7

  What distinguishes al-Ghaza¯lı¯ from Avicenna, as we 

will see in the course of this study, is that he remained ontologically uncom-

mitted to the existence of the universals outside of individual human minds. 

Although the universals may exist as entities in the active intellect, such an 

existence cannot be demonstrated. The realist understanding of the universals 

may or may not be true. In the Second Approach of the seventeenth discussion, 

he counters the realist position with the occasionalist position that human cog-

nitions are the immediate creations of God and are only congruent with the 

outside world if God wills it. 

 Some of al-Ghaza¯lı¯’s criticism in his  Incoherence of the Philosophers   cent-

ers on questioning the ontological connection between the formal structure of 

the world and the formal structure of our knowledge. Averroes (d. 595/1198), 

for instance, who shared Avicenna’s realist epistemology, was surprised by 

al-Ghaza¯lı¯’s effort to defend an occasionalist position with the argument that 

human knowledge may become disconnected from the world it aims to describe. 

That cannot be the case, Averroes says, “because the knowledge created in us 

is always in conformity with the nature of the real thing, since the defi nition of 

truth is that a thing is believed to be such as it is in reality.” 

8

  Yet this conform-

ity ( taba ¶  ) is precisely what al-Ghaza¯lı¯ argues against. Since there is no proof of 

the necessity of the connection between a cause and its effect, there is also no 

proof of the necessary conformity of our knowledge with the world. The mere 

possibility of a disconnect between the two proves that there is no formal—and 

thus necessary—coherence between the world and our knowledge of it. 

 In a later passage of the  Incoherence ,  al-Ghaza¯lı¯ comments on what he does 

in the seventeenth discussion. This comment appears in the twentieth discus-

sion of the book, on the subject of corporeal resurrection in the afterlife. The 

 fala¯sifa  argue that a resurrection of bodies is impossible, as it necessitates the 

impossible feat of transformation of substances, such as iron transforming 

into a garment. In his response, al-Ghaza¯lı¯ refers his readers back to the Sec-