University of Virginia Library

Search this document 
Dictionary of the History of Ideas

Studies of Selected Pivotal Ideas
expand section 
expand section 

expand sectionVI. 
expand sectionV. 
expand sectionVI. 
expand sectionI. 
expand sectionVI. 
expand sectionV. 
expand sectionIII. 
expand sectionIII. 
expand sectionVI. 
expand sectionVI. 
expand sectionV. 
expand sectionV. 
expand sectionIII. 
expand sectionVII. 
expand sectionVI. 
expand sectionVI. 
expand sectionIII. 
expand sectionIII. 
expand sectionII. 
expand sectionI. 
expand sectionI. 
expand sectionI. 
expand sectionV. 
expand sectionVII. 
expand sectionVI. 
expand sectionV. 
expand sectionIII. 
expand sectionIII. 
expand sectionIII. 
expand sectionII. 
expand sectionI. 
expand sectionI. 
expand sectionI. 
expand sectionVI. 
expand sectionVII. 
expand sectionIII. 
expand sectionVII. 
expand sectionVII. 
expand sectionVII. 
expand sectionV. 
expand sectionVI. 
expand sectionVI. 
expand sectionVI. 
expand sectionVI. 
expand sectionVI. 
expand sectionVII. 
expand sectionIII. 
expand sectionIV. 
expand sectionVI. 
expand sectionVI. 
expand sectionVI. 
expand sectionV. 
expand sectionV. 
expand sectionV. 
expand sectionIII. 
expand sectionIII. 
expand sectionVII. 
expand sectionIII. 
expand sectionI. 
expand sectionV. 
expand sectionV. 
expand sectionVII. 
expand sectionVI. 
expand sectionI. 
expand sectionI. 
expand sectionI. 
expand sectionI. 
expand sectionVI. 
expand sectionIII. 
expand sectionIV. 
expand sectionIII. 
expand sectionIV. 
expand sectionIV. 
expand sectionIV. 
expand sectionVI. 
expand sectionVI. 
expand sectionVI. 
expand sectionV. 
expand sectionIII. 
expand sectionVI. 


In an endeavor to safeguard what it regarded as the
Qur'anic concept of divine omnipotence, the dominant
school of Islamic theology (kalām), founded by al-
Ash'arī (d. 935), adopted the occasionalist doctrine that
causal efficacy resides exclusively with the divine will.
The Ash'arites denied the concept of “natural” causa-
tion, that is, that action proceeds from an existent's
very nature or essence. They thus rejected the Aris-
totelian concept of natural efficient causality, subject-
ing it to criticism on logical and empirical grounds.
They also rejected Aristotle's theory of eternal matter,
advocating a metaphysics of contingent atoms and
accidents that are created ex nihilo, combined to form
bodies, and sustained in temporally finite spans of
existence by direct divine action. Accordingly, the
orderly flow of these events has no inherent necessity,
being no more than a habit ('āda), decreed arbitrarily
by the divine will. Hence when God creates a miracle,
that is, when He disrupts the habitual course of nature,
no contradiction obtains. As for human volitions, acts,
and cognitions, the Ash'arites regarded these also as
temporal events (ḥawādith), the direct creation of God.

This doctrine, held with individual modifications,
became widely accepted and represents the most dis-
tinctively Islamic causal concept. The history of its
development reflects two main phases: (1) an early
phase where it is primarily concerned with doctrinal
questions within kalām; (2) a later phase, initiated by
Ghazali (al-Ghazālī; d. 1111), where it directs itself


more explicitly against the necessitarian metaphysics
of the Islamic Neo-Platonists, Alfarabi (al-Fārābī; d.
950) and Avicenna (Ibn Sīnā; d. 1037).


Ash'arite causal doctrine was at first adopted largely
in opposition to fundamental tenets of the most impor-
tant of the earlier schools of kalām, the Mu'tazilite.
This school attained the height of its influence in the
first half of the ninth century, a period caught in the
throes of the first wave of the transmission of Greek
philosophy and science to the Arabs. Although greatly
influenced by Greek thought, the Mu'tazilites remained
essentially dogmatists, selecting and reformulating
philosophical ideas to serve theological ends. Now the
Islamic version of atomism that underlies Ash'arite
causal doctrine had its origin with the Mu'tazilite
school, many of whose members adopted varying oc-
casionalist views. There was, however, no consensus
of opinion among the Mu'tazilites on causal questions,
their differences betraying the difficulties they encoun-
tered in reconciling the concept of divine omnipotence
with the two cardinal principles of their theology,
those of divine unity and justice. Ash'arite causal doc-
trine meant the rejection of these principles.

The first principle denied the distinction between
God's eternal attributes and His essence. This raised
a question concerning the concept of divine will in
relation to the doctrine of the world's temporal crea-
tion. Most of the Mu'tazilites rejected Aristotle's theory
of the potentially infinite divisibility of substance,
adopting atomism as the only view consistent with the
Qur'anic statement that God knows the determinate
number of all things. Moreover, they transformed
Greek atomism into a doctrine of transient atoms and
accidents, created ex nihilo, constituting a world cre-
ated in time. The doctrine of the world's eternity,
they maintained, deprived God of will. It meant the
simultaneity of cause and effect which only obtains,
as in natural causes, when the effect is necessitated by
the agent's nature or essence. Here, however, their
principle of divine unity faced a major difficulty: if
the divine will is conceived as an eternal attribute and
hence not distinct from the divine essence, God's acts
become in reality essential, not voluntary. This led
many Mu'tazilites to argue that the divine will itself
is created—a doctrine vulnerable to the Ash'arite criti-
cism that such a will requires another created will to
create it and so on ad infinitum.

The principle of divine justice posed further diffi-
culties. Based on the premiss that some acts have in-
trinsic moral value, it asserted (a) that God performs
only the just act and (b) that He performs it because
it is in itself good. Al-Naẓẓām (d. ca. 848) argued that
it is impossible for God to act unjustly. Other Mu'ta-
zilites, rejecting this view as a denial of divine omni-
potence, held that God has the power to perform unjust
acts, but in fact never does. How then were they to
account for evil without implicating God? In one at-
tempt to solve this problem, Mu'ammar (d. ca. 825)
maintained that God creates only bodies, imprinting
on them specific natures: accidents (bad or good) are
either the necessitated effects of these natures or the
effects of voluntary agents like man. This latter point
entailed the doctrine of man's freedom of the will,
universally endorsed by the Mu'tazilites who argued
that a just God can only judge men for acts they
themselves initiate. They agreed that man “creates”
his acts, but disagreed as to whether the human will
is effective in the outer world. Is man the real author
of “the generated acts” (al-af'āl al-mutawallida), the
events in the outer world normally regarded as the
effects of his action? Some affirmed this either wholly
or with qualifications; others denied it. Among the
latter, some held that the generated acts are necessi-
tated by the causal properties in nature; others, de-
nying natural causes, attributed them to direct divine
action. A curious position was attributed by Muslim
writers to Thumāma Ibn al-Ashras (d. 828) who was
alleged to have argued as follows: since it is possible
for a person to will an act and then die before the
consequent generated acts, and since action is not
attributable to the dead, generated acts cannot be
attributed to man. Nor can they be attributed to God,
since they may be evil. Hence, generated acts have
no author.

These inconsistencies, the Ash'arites argued, arise
from the erroneous Mu'tazilite principles of unity and
justice. The eternal attributes, they argued, are “addi-
tional” to God's essence and include His will. It is thus
that God's acts are voluntary, not necessitated by His
essence. As for the second principle, all acts, the
Ash'arites argued, are in themselves morally neutral.
An act is just simply because God performs or com-
mands it; unjust, because He prohibits it. Divine deci-
sions are not conditioned by any intrinsic values in acts
and the question of whether God has the power to
act unjustly is redundant. Men do not create their acts.
Human acts, like all other events, are created directly
by God. To avoid identification with the Islamic deter-
minists (al-mujabbira) who held that human acts are
“compelled” by God, the Ash'arites resorted to the
proverbially enigmatic doctrine of “acquisition” ( al-
). They distinguished between a compulsory act,
like a spasmodic bodily movement, and a deliberate
act. While both kinds of acts are created in man by
God, the second is created together with the “power”
(al-qūwa), which exists only with the act, not before


it. They also seem to have held that whatever is nor-
mally regarded as the effect of man's deliberate act
is also created by God simultaneously with the power,
so that both “power” and “effect” are acquired by man
from God. In other words, the relation between man's
“created power” and the “created effect,” as in all
sequences in the world, is that of mere concomitance.
(But this point has been interpreted differently; see


Initial Ash'arite repudiation of natural causation, as
we have seen, arose out of doctrinal disputes within
kalām. The arguments, as far as the available sources
indicate, are mainly theological and metaphysical. A
second phase in the history of the Ash'arite causal
theory is marked by a more explicit attack on the
Aristotelian theory of necessary efficient causality as
adapted to the emanative metaphysics of Alfarabi and
Avicenna. Here we encounter two developments: (1)
an emphasis on the purely epistemological argument
that necessary causal connection is provable neither
logically nor empirically; (2) an attempt to uphold the
Aristotelian method of scientific demonstration and its
claims of attaining certainty, on occasionalist, non-
Aristotelian, metaphysical grounds. Both these devel-
opments are largely due to Ghazali. Elements of the
epistemological argument are found in earlier Ash'arite
writings, but it was Ghazali who gave this argument
its most forceful expression and who first attempted
an occasionalist reinterpretation of Aristotelian dem-
onstrative logic.

Ghazali's criticism of necessary efficient causation
pervades his Tahāfut al-Falāsifa (The Incoherence of
the Philosophers
), directed mainly at Avicenna's
emanative metaphysics. Avicenna described the effi-
cient cause as that “which brings about an existence
other than itself.” In natural science, he held, this
existence represents motion in one of its forms. For
the metaphysician, however, and here we note his
Neo-Platonism, the efficient cause is also productive
of existence as such. In Avicenna's cosmogony, the
world emanates eternally from God as a chain of ne-
cessitated and necessitating existents, terminating in
the world of generation and corruption. In this sub-
lunar world, for the effect to follow from the efficient
cause, the material, formal, and final causes must also
exist. The efficient natural cause must be the proximate
cause and there must be no impediment. Agency, in
a natural cause, is “an essential attribute,” hence pro-
ductive of one specific kind of act. Action is also
determined by the specific nature of its recipient.
When such causal conditions obtain, the effect follows
by necessity.

Ghazali attacked Avicenna's concept of divine cau-
sality as a negation of the divine attributes of life, will,
and power. Only inanimate beings, Ghazali argued, are
said to act by necessity. By definition, a necessitated
act is not a voluntary act. Ghazali also opposed
Avicenna's scheme on the grounds that it does not
allow God to act directly in the world of men, but
only through the mediation of other causes. Since the
chain of existents proceeding from God is necessarily
connected, there can be no disruption of its order.
Miracles, defined by the Ash'arites as the disruption
of nature's habitual order, are thus impossible and a
prophet proclaiming their occurrence becomes a de-
ceiver. It is in arguing for the possibility of miracles
that Ghazali levelled his epistemological argument
against the concept of necessary causal connection:

The connection between what is habitually believed to
be the cause and what is habitually believed to be the effect
is not necessary for us. But in the case of two things, neither
of which is the other and where neither the affirmation nor
the negation of the one entails the affirmation or the nega-
tion of the other, the existence or non-existence of the one
does not necessitate the existence or non-existence of the
other; for example, the quenching of thirst and drinking,
satiety and eating, burning and contact with fire, light and
the rising of the sun, death and decapitation.... On the
contrary, it is within God's power to create satiety without
eating, death without decapitation, to prolong life after
decapitation and so on in the case of all concomitant things.

[Trans. by the author.]

He then argued that necessary causal connection is
never observable in nature: when, for example, cotton
is brought in contact with fire, all that is seen is the
occurrence of burning “with” (ma') the contact, not
the burning of the cotton “by” or “through” (bi) the
fire. The one who enacts the burning, he asserted, is

Ghazali did not deny that events in the world are
ordered in sequences of priority and posteriority, tem-
poral and ontological, ordinarily regarded as causes and
effects and on the basis of which scientific inferences
about nature can be drawn. He denied, however, that
these latter are real causes and that their order is
inherently necessary. He endorsed Aristotle's method
of scientific demonstration, but sought its interpretation
in occasionalist terms. He thus used Avicenna's argu-
ment to justify the principle of nature's uniformity, but
drew from it a different conclusion. Avicenna (basing
himself on Aristotle, Physics ii. 5. 196b 10-15) had
argued that mere observation of past uniformities does
not suffice to give us the certainly of their future
continuance; in addition, there is the “hidden syl-
logism” to the effect that if these had been accidental
or coincidental, “they would not have continued al-


ways or for the most part.” Ghazali endorsed the argu-
ment to this point, but unlike Avicenna, who concluded
that the uniformity derives from the natural necessary
connection between things, Ghazali maintained it de-
rives from God's arbitrary decree. For Ghazali, God
creates in man knowledge that the world is orderly,
but also that its order is contingent and disruptible.
When a miracle occurs, God refrains from creating in
man the expectation of the uniform event, creating
instead knowledge of the miracle. Ghazali did not
elaborate on this latter point, leaving unanswered seri-
ous questions arising from it.


In Dalālat al-Ḥā'irīn (The Guide of the Perplexed),
of which a Latin version, based on a Hebrew transla-
tion, was known to Christian scholastics early in the
thirteenth century, Moses Maimonides (d. 1204) dis-
cussed the occasionalist atomism of kalām, criticizing
it mainly on the metaphysical level. Averroës (Ibn
Rushd; d. 1198), in his Tahāfut al-Tahāfut (The In-
coherence of “The Incoherence”
), answered Ghazali's
Tahāfut, quoting almost all of it; a Latin translation
of Averroës' work was first made in the fourteenth
century. These translations have raised the question of
a possible Islamic influence on parallel criticisms of
causation in Europe, particularly that of Nicolaus of
Autrecourt (d. 1350) whose writings suggest acquaint-
ance with Maimonides' account of Islamic atomism.
For the history of the concept of causation in Islam,
however, Averroës' Tahāfut is of special interest. In
this and other shorter works Averroës was attempting
to check the spread of Ash'arism, particularly in North
Africa and Muslim Spain. The attempt, however, was
abortive, and Aristotelian causal theory, though it con-
tinued to be held in Islam, remained on the defensive.


Pertinent medieval Arabic sources in translation include
al-Ash'arī, Kitāb al-Luma', trans. R. J. McCarthy (Beirut,
1953); Averroës, Tahāfut al-Tahāfut, trans. S. Van Den Bergh
(London, 1953), particularly the 3rd and 17th discussions;
Avicenna, La Métaphysique du Shifā', French translation
by M. M. Anawati, mimeographed edition (Quebec, 1952),
particularly, Book IV, Ch. 1; al-Ghazālī, Tahāfut al-Fal
trans. S. A. Kamali (Lahore, 1958); Ibn Khaldūn, The
trans. F. Rosenthal, 3 vols. (New York, 1958;
2nd ed., 1967), Vol. III, Ch. IV, Sec. 14; al-Khayyāt, Kitāb
reprint of M. Nyberg's Arabic edition with a
French translation by A. N. Nader (Beirut, 1957); Maimon-
ides, The Guide of the Perplexed, trans. S. Pines (Chicago,
1966), Chs. 73-76.

A basic historical account of Islamic occasionalism with
a philosophical discussion defending a Thomistic approach
to causality is M. Fakhry's Islamic Occasionalism and its
Critique by Averroës and Aquinas
(London, 1958). For an
interpretation of Ibn Khaldun's discussion of causality, see
H. A. Wolfson, “Ibn Khaldun on Attributes and Predes-
tination,” Speculum, 34 (October, 1959), 585-97, reprinted
in H. A. Wolfson, Religious Philosophy: A Group of Essays
(Cambridge, Mass., 1961). For a detailed discussion of
Ghazali's attempt at interpreting Aristotelian demonstrative
logic in occasionalist terms, see M. E. Marmura, “Ghazali
and Demonstrative Science,” Journal of the History of Phi-
3 (October, 1965), 183-204. On kalām theories of
free will and ethical value, see M. Fakhry, “The Mu'tazilite
View of Free Will,” The Muslim World, 42, 2 (April, 1953),
95-109; G. F. Hourani, “Two Theories of Value in Medieval
Islam,” The Muslim World, 50, 4 (October, 1960), 269-376;
A. N. Nader, Le Système Philosophique des Mu'tazila
(Beirut, 1956); W. M. Watt, Free Will and Predestination
in Early Islam
(London, 1948). For an interpretation of the
doctrine of kasb differing from ours that allows a measure
of genuine efficacy in deliberate human action, see R. M.
Frank, “The Structure of Created Causality according to
al-Ash'arī,” Studia Islamica, 25 (1966), 13-75. On Islamic
atomism, see S. Pines, Beiträge zur Islamischen Atomenlehre
(Berlin, 1936). For the question of a possible Islamic influ-
ence on Nicolaus of Autrecourt, see J. R. Weinberg, Nicolaus
of Autrecourt
(Princeton, 1948). For general accounts of
Islamic theology, see L. Gardet and M. M. Anawati, Intro-
duction à la Théologie Musulmane
(Paris, 1951); D. B.
Macdonald, Development of Muslim Theology, Jurispru-
dence and Constitutional Theory
(New York, 1903), outdated
but still pertinent; W. M. Watt, Islamic Philosophy and
(Edinburgh, 1962).


[See also Atomism; Causation; Causation in Law; God;
Islamic Conception;
Necessity; Neo-Platonism.]