Forum

Not Truth But Tolerance:
A (Much Belated) Response

to Atif Khalil

Sherman A. Jackson
I should like to begin this essay with a sincere apology. More than five years 
have passed since Professor Atif Khalil penned his scholarly critique of 
some of my suggestive ruminations on intra-Islamic theological ecumen-
ism in the introduction to my translation of Abū Ḥāmid al-Ghâzalî’s Fayṣal 
al-Tafriqa.1 While scholarly convention – not to mention etiquette ‒ would 
certainly demand a much more timely response than I have been able to 
manage, I am afraid that I can plead no better than to throw myself on the 
understanding of those who have insight into and appreciation for the vari-
ous ways in which the horrific attacks of September 11, 2001 have skewed 
the scholarly agenda of many an Islamicist. I sincerely hope that this delay 
will not be construed as some kind of veiled or surreptitiously snide dis-
missal of Professor Khalil’s thoughtful analysis. I also hope that it will not 
have gone so far as to suggest any inability on my part to respond to what 
I shall argue amounts to a clever but ultimately wrong-minded critique.

According to Professor Khalil, despite the ingenuity with which I 
approach the issue of Muslim theological diversity, many of my central 

________________________________________________________________________
Sherman A. Jackson is the King Faisal Chair of Islamic Thought and Culture, Professor of 
Religion and Professor of American Studies and Ethnicity at the University of Southern 
California.



Jackson: Not Truth But Tolerance   147

arguments are beset with “‘internal contradictions and incongruencies’ that 
might otherwise evade the casual reader” (85). At bottom, these can be 
summarized as: (1) the lack of an objective criterion for distinguishing “le-
gitimate [interpretive] traditions from illegitimate ones”; (2) an inescapable 
relativism, especially given my apparent assertion that the true meaning 
of historically transcendent scripture remains hopelessly closed to histori-
cally embedded, contingent human beings; (3) the notion that we can have 
‘aqīdah (sustained belief about God) independent of a systematic rational 
method to produce and sustain it; (4) the false dichotomy I assert between 
eisegesis and tafsīr (Qur’ānic exegesis); and (5) my failure to consider what 
the likes of the famed if controversial Muḥyī al-Dîn ībn ‘Arabī might con-
tribute to a project of theological ecumenism. There are a number of other 
minor issues and extrapolations tucked in the interstices of all of this. But 
what I have enumerated here constitutes the backbone of Khalil’s critique. 

I should like to begin my response by noting that, despite my recognition 
of the utility of al-Ghâzalî’s work for a contemporary program of theological 
ecumenism, the main point of my introduction was to empower the reader 
to arrive at a more informed understanding of the translated text. Through-
out Professor Khalil’s critique, however, it is not always clear whether his 
issue is with me or al-Ghâzalî, or both. In the end, it turns out to make little 
difference. For, as we shall see, Khalil’s critique is ultimately grounded in 
a fundamental and consistent misunderstanding of both al-Ghâzalî and me. 

Both al-Ghâzalî and I are clear and explicit about the objective of 
his (and by extension my) project ‒ to establish a criterion for theologi-
cal tolerance. The title of the book is On the Boundaries of Theological 
Tolerance in Islam. And the opening pages of the introduction explicitly 
state that al-Ghâzalî’s aim is, “not to establish who among the theologi-
cal schools is ‘right,’ but to demonstrate the folly and unfairness of the 
practice of condemning a doctrine as heresy simply because it goes against 
one’s own theology.”2 Khalil, on the other hand, seems bent on convert-
ing al-Ghâzalî’s (and my) aim into one of pursuing and policing theologi-
cal truth ‒ of establishing the means via which correct doctrine might be 
arrived at and of laying down a criterion on the basis of which true be-
liefs might be objectively differentiated from false ones. This is why, for 
example, he takes me to task for not spelling out an objective criterion 
for distinguishing “legitimate” from “illegitimate” interpretive backdrops 
against which scripture might be read. For on this omission, I (and here 
he explicitly targets me) am ultimately bound to accept as valid “all pos-
sible doctrine,” since a boundless array of equally legitimate interpretive 



The American Journal of Islamic Social Sciences 28:4148

backdrops must yield a boundless array of equally legitimate interpreta-
tions. As he put it, “This … opens the door to an acceptance of not only 
Process Theologian Hanbalites and Aristotelian-Neoplatonic Ash‘arites, 
but anyone who simply claims to speak on behalf of Islamic revelation, 
no matter how convoluted their logic might seem” (87). This lack of a 
criterion for separating true from false backdrops cum-doctrine is a par-
ticularly glaring oversight on my part, given that al-Ghâzalî, according 
to Professor Khalil, “suggests the exact opposite in Fayṣal al-Tafriqa, 
when he argues for the need of a common and agreed upon methodol-
ogy to eliminate flawed interpretations (pp. 93‒96)” (87, emphasis mine).

To be sure, both al-Ghâzalî and I have an obvious interest in truth cum-
correct theological doctrine.3 But, contrary to Professor Khalil’s reading, 
Fayṣal al-Tafriqa bayna al-Islam wa al-zandaqa is simply and emphati-
cally not about the pursuit of truth, at least not as a primary, first order con-
cern; nor is it, precisely for this reason, ultimately about accommodating 
multiple expressions of truth; nor is it even about distinguishing truth per 
se from falsehood per se. Rather Fayṣal al-Tafriqa is about sharpening and 
policing the definition of unbelief and affirming that not all untruth neces-
sarily amounts to kufr (formal unbelief)! In this light, Professor Khalil’s 
focus on what is substantively valid, correct, or legitimate simply misses 
the point. For the very crux of al-Ghāzalī’s argument is that any number of 
views that are substantively invalid ‒ indeed, demonstrably false, untrue 
or incorrect ‒ may be tolerated inasmuch as they do not constitute kufr! 
Again, al-Ghâzalî’s aim is to distinguish who is a Muslim – theologically 
speaking – and who is not and to confirm, much to the chagrin of those 
he characterizes as “extremists,” that one can be a Muslim despite one’s 
subscription to any number of substantively wrong theological views.4

As for Professor Khalil’s assertion that al-Ghâzalî aspired to lay 
down a “universal principle” to serve as “a common and agreed upon 
methodology to eliminate flawed interpretations,” this too reflects a ba-
sic misunderstanding of al-Ghâzalî’s project. For in Fayṣal, he is not 
at all interested in establishing correct or eliminating flawed doctrine 
per se. Rather, in Fayṣal, his concern lies with the tyranny of the uni-
versal and the exclusivist claims of those ‒ for example, ‘Abd al-Qāhir 
al-Baghdādī ‒ who believe themselves, rightly or wrongly, to be in pos-
session of correct theological doctrine and that the resulting “incorrect” 
doctrines of their theological adversaries necessarily amounts to unbelief.

Two examples from Fayṣal itself should help to throw my argument 
into relief. The first comes with al-Ghâzalî’s treatment of Twelver Shiism.5 
Speaking of the Imāmī doctrine of occultation (ghaybah), he states explicitly: 



Jackson: Not Truth But Tolerance   149

This is a false, clearly absurd and extremely abominable doctrine. But it 
poses no threat to religion. In fact, the only threat it poses is to the fool 
who believes in it…. The point here is that not everyone who embraces 
senseless hallucinations must be branded an Unbeliever, even if his 
doctrines are clearly absurd.6

Al-Ghâzalî clearly believes the Twelver Shiite doctrine of occultation 
to be neither valid nor legitimate, certainly not in the sense that Khalil uses 
these terms. And yet, it is precisely his aim in Fayṣal to refute and interdict 
such practices as branding Twelver Shiites as unbelievers simply because 
they hold this substantively incorrect belief. Indeed, Twelver Shiites, de-
spite their “wayward doctrine,” fall perfectly within the parameters of Is-
lam that al-Ghâzalî articulates and defends in Fayṣal. Again, pace Profes-
sor Khalil, al-Ghâzalî’s (and my) aim is simply not to rid the market of all 
wrong ideas and replace these with substantively correct ones. Rather, the 
only idea that al-Ghâzalî (or I) wants to rout from the field is the idea that 
all wrong theological views invariably take one outside the pale of the faith.

The second example speaks to Professor Khalil’s understanding of al-
Ghâzalî’s “common and agreed upon methodology” as being for the pur-
pose of determining which traditions are valid and which are not. Again, 
Khalil’s point here is that without an objective criterion for determining 
which traditions can be legitimately relied upon in interpreting scripture 
and which cannot, we have no way of distinguishing the true from the false 
among the interpretations generated thereby. On this inability, all inter-
pretations must be recognized as correct, since no tradition or interpretive 
backdrop can rightfully claim a status that it denies to all the rest. Again, 
however, a careful reading of al-Ghâzalî reveals that his preoccupation is 
not at all with determining which traditions are legitimate and which are 
not. Rather, his point of departure is the simple and undeniable fact that 
Muslim scriptural interpretations are informed by a variety of competing 
interpretive backdrops. To see him as focusing on (and to see me as having 
to focus on) which of these is legitimate and which is not is simply to miss 
his (and my) point.7 For even where a tradition is deemed “illegitimate,” 
this alone does not doom the status of the views it engenders. Speaking, for 
example, of the widely diffused tradition of rationalist kalām (speculative 
theology) as an interpretive backdrop and “methodology,” al-Ghâzalî states:

[W]ere we ourselves to put aside all pretensions of deference and 
decorum, we would declare outright that delving into speculative theology 
[kalām] is religiously forbidden (ḥarām)… (123). 

To avoid misunderstanding here, I should note this was a judg-
ment made by al-Ghâzalî in passing in Fayṣal, as any fair reading 



The American Journal of Islamic Social Sciences 28:4150

of the text will plainly bear out.8 Still, he explicitly negates kalām’s 
status as a, or the, normative backdrop or methodology against or 
via which to read or vindicate scripture. And yet, al-Ghâzalî clearly 
does not intend to proscribe all doctrines that recline upon kalām. 
This is because the legitimacy or validity of the interpretive tradition 
from which an interpretation draws its substance is for him largely 
a moot point. The operative issue is, rather, how the interpretation 
itself relates to his criterion for tolerance ‒ that is, not whether it is 
substantively right or wrong but whether or not it constitutes kufr. 

In many ways, and with a number of obvious qualifications, Fayṣal 
al-Tafriqa might be more profitably understood not as a theological 
tract but as a political one ‒ far more akin in spirit and intent with, say, 
John Locke’s A Letter Concerning Toleration than with Augustine’s On 
Christian Doctrine.9 Its aim is neither to affirm, establish, or police the 
concrete substance of theological truth but to referee and impose a sem-
blance of discipline upon competing pronouncements of theological in-
fidelity. Understanding this subtlety is key to understanding the role and 
function of reason in Fayṣal. And understanding al-Ghâzalî’s deployment 
of reason in this text is the key to understanding that ‒ pace Professor 
Khalil, neither al-Ghâzalî nor I are advocates of theological relativism.

For al-Ghâzalî, reason is not the primary – and certainly not the exclu-
sive ‒ means to religious truth; indeed, for him ‒ pace modern liberalism 
‒ reason is emphatically not the only means of knowing. Rather, truth and 
knowing for al-Ghâzalî are often far more subtle and subjective indulgences. 
This epistemological insight is clearly reflected in his critique of rationalist 
kalām, in which he scoffs at the claim that speculative theology (deemed 
by many at the time to be reason par excellence) is the only way to truth:

 [A]nyone who believes that the way to faith (îmân) is speculative 
theology, abstract proofs, and systematic categorization is himself 
guilty of unsanctioned innovation. For faith in God comes rather of 
a light which God    casts into  the hearts of His servants, as a gift 
and gratuity from Him.  Sometimes this comes in the form of a proof 
that appears to one internally but which one cannot explain to others; 
sometimes it comes through visions in one’s sleep; sometimes it comes 
by witnessing the ways of a religious man whose light is transferred to 
one upon befriending and spending time with him; and sometimes it 
comes by way of circumstantial considerations…. (121)10

Now, while systematic reason may not be the means via which one 
arrives at all of one’s religious convictions, reason remains, ceteris pa-



Jackson: Not Truth But Tolerance   151

ribus, the most likely if not only ostensibly objective11 medium through 
which such views can be publicly negotiated cum-validated. The aim of 
Fayṣal is to enlist reason as a public medium for negotiation by laying 
down the “rules of engagement.” These rules, however, are neither cali-
brated nor bound to preempt the emergence of any and all substantively 
wrong views. Rather, the whole point of al-Ghâzalî’s Fayṣal, literally, 
“decisive criterion”; it is to determine how wrong a recognizably wrong 
view must be if its proponent is to be legitimately branded an infidel. 

Professor Khalil seems to think that this kind of accommodation 
amounts to theological relativism. But this, again, bespeaks a serious mis-
understanding of al-Ghâzalî’s (and my) project, as well as the very nature 
and pluralistic sensibilities of Islam as a “religion.” In a word, “tolerated” 
is simply not the same as “correct,” “endorsed,” or “agreed with.” Nor does 
the fact that I tolerate a wrong view in others mean that I have no interest 
– or even a reduced interest ‒ in ensuring that my own views are substan-
tively correct.12 Since, however, the true ground of so much theological be-
lief is demonstrably subjective and closed to the dictates13 of reason, I must 
accept the fact that I may not be able to communicate my beliefs faithfully 
to others or enlist their assent. Again, however, such recognition should not 
be mistaken for a lack of conviction regarding my own views. Rather, for 
me, absolute truth exists ‒ and I know what it is! I also know, however, that 
you “know” too and that I may not be able to convince you of my truth.14 
At its essence, al-Ghâzalî’s (and my) project is about allowing each of us to 
look upon our own truth as absolute without having to look upon what we 
deem to be the falsehoods of our coreligionists as absolute in the sense of 
constituting kufr. This is hardly an exercise in theological relativism. For, 
assuming due-diligence, it neither entails nor necessitates the slightest hes-
itation in judging one’s own or others’ views to be absolutely true or false.15 

If I had to guess, I would hazard that Professor Khalil’s reading of al-
Ghâzalî (and of me) owes much to the intellectual liberalism enshrined by 
the Western academy, where closed, subjective conduits of knowledge ‒ the 
heart (qalb), the self/soul (nafs), the primordial disposition (fiṭrah), guid-
ance from God (hudā) ‒ have no epistemological value at all. This in turn 
prompts him to superimpose upon al-Ghâzalî (and me) an Enlightenment 
understanding of the relationship between knowledge, truth, and reason. 
What we know, we know only through reason; and our claims to knowledge/
truth can be validated only by rational arguments.16 On this understanding, 
what al-Ghâzalî must be seeking to negotiate through his patently rational 
approach in Fayṣal  is the substantive content of truth. For not only is truth 
the business of reason; it is the latter’s inevitable and inextricable result. 

http://en.wikipedia.org/wiki/Fitrah


The American Journal of Islamic Social Sciences 28:4152

Again, however, a careful reading of al-Ghâzalî reveals that not only is he 
opposed to this kind of rationalist fundamentalism,17 Fayṣal is dedicated to 
negotiating not truth but tolerance. The role and function of reason in this 
text is to promote not the religious truth but the secular peace! Whether or 
not this or that Muslim group or individual ultimately ends up in possession 
of the actual religious truth is a matter to be left to God in the life to come.18 
In the here and now, the issue is how competing claims to truth and judg-
ments of falsehood can be accommodated – not endorsed! – without threat-
ening or undermining the integrity of Islam and the Muslim community. 

This is the actual ideational context and utilitarian intent of Fayṣal. In 
its light, not only does the bulk of Professor Khalil’s major criticisms be-
gin to fade, his more minor criticisms are also revealed to be problematic. 
Take, for example, the claim that I argue that “revelation qua revelation, 
insofar as its source is ahistoric, remains forever inaccessible to the histori-
cally contingent theologian (of any school) and that the most people can 
do is engage in limited, fallible attempts to interpret and understand the 
divine intention behind scripture…” (87). Again, having misdiagnosed the 
main objective of Fayṣal, Professor Khalil goes on to confuse the ques-
tion of what we can know with that of what we can prove. I never said 
(or implied) that because humans are embedded in history they can never 
know the intended meaning of scripture. How could I, when Islam’s very 
creation narrative includes transcendent God teaching contingent Adam 
the names of all things? And for the post-creation period of secular his-
tory, the Qur’ān repeatedly refers to God’s gifting humans – that is, not 
just prophets – direct knowledge of truth. Pharaoh, for example, and his 
people, are said to have received God’s signs but to have, “rejected them, 
out of impudence and arrogance, despite the certainty of their truth that 
had settled in their souls” (27:14). And if “signs” here is too ambiguous to 
dispose of the matter, we need only note the case of a segment of the Chil-
dren of Israel who are explicitly credited with having an accurate reading 
of scripture: “They hear the speech of God and then knowingly distort its 
meaning after they had clearly understood it” (2:75). In none of these cases 
is there any question but that historically embedded humans can know the 
intent of scripture. What they cannot do – and this was the point that Pro-
fessor Khalil seems to have misconstrued – is impose their historically 
embedded understandings on others, who do not share their interpretive 
background, as self-authenticating, unassailable, universally valid truth.

Another minor criticism leveled by Professor Khalil is his apparent 
attribution to me of a rather curious strain of antirationalism, according 
to which reason is deemed incapable of knowing anything. I never said 



Jackson: Not Truth But Tolerance   153

or implied this (nor did al-Ghāzalī).19 What I suggested, rather, following 
the lead of al-Ghâzalî, was that systematic reason has its epistemological 
limits. Of course, reason can know, certainly from the perspective of the 
reasoner him- or herself. But we should be clear here that theology reclines 
not simply upon the natural faculty of reason but upon specific regimes 
or systems of reason. And while these regimes or systems can produce 
metacognitive, internally self-authenticating knowledge or truth, what they 
cannot do is sustain universal claims to this truth and knowledge across 
boundaries that separate them from other regimes or systems of reasoning.20 

Again, Professor Khalil finds this to be highly problematic. For, for him, 
this denies the possibility of arriving at universal truth that can be indepen-
dently validated as such. As he put it, “if no school can claim comprehensive 
[read, “universal”] doctrinal truth, then are we not forced to accept that all 
schools can, in the absence of an agreed upon method, claim it at the relative 
level?” (87). My response to this would be, “No, all schools can claim – and 
actually believe! ‒ it at the absolute level!” What al-Ghâzalî wants to do is 
simply keep these mutually competing and contradictory claims to absolute 
truth and falsehood from leading to a national pastime spent at the gallows. 

Related to this point is Professor Khalil’s criticism of my claim that 
one can have ‘aqīdah without relying on theology as a systematic means 
of arriving at and validating it. Here, however, he appears to stare reality 
in the face but to misapprehend it by dint of the rationalist fundamental-
ism he seems to have embraced. He notes explicitly (though he deems it 
misleading for me to note it) that the Prophet (ṢAAS) could have “direct 
access to transcendent truth by virtue of his prophethood,” and that “He 
had no need for a theological method to understand revelation because he 
was in a state of constant communion with God” (89).The reality, however, 
is that the Prophet had no need for theology (or fiqh, for that matter) not 
because he was in communion with God (after all, Muslims believe that 
God can inspire (yulhim) or guide (yahdī) any of us). The Prophet had no 
need for theology because his authority and status as Prophet relieved him 
of any and all necessity of validating his views to others. In other words, 
he could claim for his theological views ‒ by sheer, undemocratic, pro-
phetic fiat ‒ an authority that no one else could either claim or challenge. 

At the bottom, however, the real issue here is, again, the difference 
between “arriving at” and “validating” a belief. As a matter of private, indi-
vidual belief, anyone could give assent to his or her understanding of what 
the Prophet handed down and hold that as their theological belief, with no 
effort or even ability to validate this. The moment, however, they ventured 
into the public and claimed (or implied) that this belief was normative for 
others; they, unlike the Prophet, would have to point to something beyond 



The American Journal of Islamic Social Sciences 28:4154

that belief itself ‒ for example, rational argument, historical precedent, 
their recognized status as a person of gnosis, or supersensory knowledge ‒ 
in order to endow it with enough authority to validate it. Clearly, however, 
this process of validation would be separate and distinct from the actual 
instantiation of the belief itself. Professor Khalil’s understanding of the 
role of reason, however ‒ that is, as the necessary basis of both the instan-
tiation and the validation of beliefs ‒ seems to blind him to this distinction. 
But, unless we are talking about correct doctrine (as determined, that is, 
by some specific criterion), it seems to me merely to state the obvious to 
say that one can have all kinds of beliefs about God without systematic 
ways of arriving at or validating them. Professor Khalil’s obsession, how-
ever, with doctrinal correctness ‒ in tandem with the almost mechanically 
causal relationship he seems to posit between reason and truth ‒ leads him 
to impose this obsession on al-Ghâzalî (and me). Reason and only rea-
son can produce and judge truth. Otherwise, “All we could attain in re-
gards to … [our] truths would be a kind of mindless assent to a very small 
and specific set of assertions about God that are explicitly spelled out in 
revelation” (89). Now, even if we leave aside the question of what, then, 
judges reason, and even if we ignore the implications of acknowledging 
that some things can be known directly by virtue of how explicitly they 
are laid out in revelation, we are inexorably brought back to the fact that 
measuring the concrete correctness of competing doctrines is simply not 
al-Ghâzalî’s program, at least not in Fayṣal. On the contrary, al-Ghâza-
lî’s concern lies, again, not with doctrinal correctness but with tolerance. 

This misapprehension of the distinction between truth or doctrinal cor-
rectness and tolerance also informs Professor Khalil’s critique of my refer-
ence to process theologian Charles Hartshorne. Khalil suggests that I am 
disingenuous when I invoke Hartshorne’s logic but do not subscribe to it. 
Again, however, this is based on a misapprehension of the meta-context 
of my and al-Ghâzalî’s project and the role that these references to Harts-
horne play in it. Professor Khalil starts out by noting correctly that I adduce 
Hartshorne to back the argument that Traditionalist Hanbalites could make 
an equal claim to being rational, since, on Hartshorne’s logic, “settling” on 
the throne, for example, could not be deemed irrational.21 But then he goes 
on to claim that by using Hartshorne to argue for equal “legitimacy” be-
tween Traditionalists and Rationalists, I run into the problem of not having 
a criterion for distinguishing “legitimate traditions from illegitimate ones” 
(87). I have already dealt with this above. Here, however, I would simply 
add that my reference to Hartshorne was not at all an effort to identify an 
alternative backdrop (or logic) that might lead to correct doctrine; it was 



Jackson: Not Truth But Tolerance   155

simply to argue for the existence, pace the Muslim Rationalists, of multiple 
regimes of reason (‘aql), and to point up the fact that if the criterion for 
acceptance of a doctrine was simply that it be reasonable, Traditionalist 
doctrine should pass muster. Again, this was an effort to enlist Hartshorne 
into the cause of al-Ghâzalî’s campaign for tolerance, not, as Professor 
Khalil seems to see the matter, of co-opting him into the cause of truth. 

There are two final points that Professor Khalil raises that are actually, 
in my view, rather marginal to the main argument but still deserve perhaps 
some comment. The first of these is his critique of the distinction I draw 
between eisegesis and tafsīr. Here, let me state openly that if I were to 
rewrite this section ten years later, I would word it more carefully. I would 
not, however, change the basic substance of what I said. What I said was 
that as a discipline, tafsīr is supposed to amount to a simple exercise in 
exegesis ‒ that is, of extracting the meanings of words in simple diction-
ary fashion. Khalil challenges this and argues that in point of fact tafsīr 
routinely if not necessarily entails some level of eisegesis. Now, in the 
main, I agree with Khalil. But I think he rather exaggerates (and in so do-
ing distorts) my point. I was not arguing that tafsīr never entails eisegesis; 
in fact, the example I gave of Imām Ahmad clearly demonstrated that it 
does.22 My point was simply that because tafsīr is supposed to be a simple 
matter of extracting meaning in simple dictionary fashion ‒ that is, with no 
formal, ideological presuppositions informing this process ‒ every theo-
logian would want to pretend that he or she was involved precisely and 
only in this enterprise when doing theology. But theology, I argued, being 
grounded as it is in the attempt to make sense of revelation in the context 
of some formal or quasi-formal regime of reason, cannot dispense with 
eisegesis, even if it is not always willing to admit this.23 Now, Khalil points 
to the fact that al-Ṭabarī contradicts the definition of tafsīr I attributed to 
him (91). In a sense, however, this actually confirms the point that I was 
trying to make: eisegesis routinely invades the domain of exegesis. But 
Khalil takes al-Ṭabarī’s contradiction of the definition of tafsīr I adduced 
and attributed to him to be a contradiction of the definition itself. Yet, 
when we consult such authoritative classical lexicons as Lisân al-‘Arab, 
we find precisely the definition of tafsīr I cited: “fsr … and tafsîr is to 
uncover the meaning of a difficult expression (wa al-tafsîr kashf al-murâd 
‘an al-ḷafz al-mushkil).”24 Again, my point was not that tafsīr never en-
tailed eisegesis. My point was rather that, properly speaking, it is not sup-
posed to ‒ even if, especially in the service of theology, it routinely does.

Finally, there is the suggestion that I was remiss in not mentioning the 
famed Sufi Muḥyī al-Dîn Ibn ‘Arabī in the context of explicating my (and 



The American Journal of Islamic Social Sciences 28:4156

al-Ghâzalî’s) project. There is much to be said about this, but let me limit 
myself here to the following. It seems to me that Ibn ‘Arabī in particular 
would be a problematic figure to integrate into a discussion of al-Ghâzalî’s 
project, inasmuch as the entirety of Ibn ‘Arabī’s thought, as I understand him, 
proceeds on the basis of a cosmology fundamentally at odds with that of al-
Ghâzalî. For al-Ghâzalî – certainly in the context of the theological universe 
and interlocutors he assumes in Fayṣal 25 ‒ the Creator-created dichotomy 
and distinction is both vertical (that is, hierarchical) and absolute. For Ibn 
‘Arabī, on the other hand, and his waḥdat al-wujūd (unity of existence), 
this distinction might be said to be neither vertical nor entirely absolute. As 
one of my teachers (a master and proponent of Ibn ‘Arabī’s thought) used 
to explain it, one can say, “God is this chair,” even if one cannot say, “This 
chair is God.” Now, I am not sure about how all of this might or might not 
be reconciled with al-Ghâzalî. But I fail to see how caution and prudence in 
this regard can be turned into a charge of gratuitous, unwarranted omission.

There are numerous lesser issues in Professor Khalil’s critique on 
which one might wish to comment.26 In closing, however, I would like 
to take the opportunity to address a broader problem informing the study 
of Islam in the West of which I think Professor Khalil’s analysis might 
be a reflection. Given the power and prestige of the Western academy 
– including the rise and incursion into Islamic studies of the social sci-
ences, according to whose approach the foundational texts and tradition 
of Islam are often marginal to a determination of the normative param-
eters of the religion ‒ it seems that Muslims who write out of a classical 
idiom are increasingly less understood on their own terms. Instead, the 
interpretive prisms of the Western academy and its intellectual liberalism 
routinely impose meanings and implications that distort if not subvert the 
integrity of classical and neo-classical Islamic thought. Beyond their sub-
stance, moreover, these interpretive interventions routinely go on to im-
ply a finality that all but raises them beyond critique. This is not always 
the case; but it seems to be increasingly so. We see it, for example, in 
the superimposition of fundamentalist/literalist interpretations onto the 
Muslim legal tradition by scholars who avoid any serious engagement of 
Islamic law itself. On this approach, Muslims cannot honestly extract a 
pluralistic political theory from the Qur’ān, Sunnah or Shari‘ah tradition, 
because those writing in the powerful Western academy cannot.27 We see 
it as well in the tendency to graft the legal monism of the Western nation-
state onto Islam, in which light Muslims must take it to be their duty to 
impose a uniform body of rules indiscriminately across all segments of 
society, because this is what modern Western states purport to do. Shari‘ah 



Jackson: Not Truth But Tolerance   157

in this context, whether it really is or not, comes to be seen as a mortal 
threat to all non-Muslims and even Muslims who might dare to dissent.28 

And now, with Professor Khalil, one wonders if we are witnessing 
yet another incursion of Western liberalism in the form of imposing the 
pursuit, if not imposition, of truth as the only real and legitimate concern 
of religion, to the exclusion ‒ in our case, of tolerance, but more generally 
of order, efficiency, and a host of other secular concerns. Viewed through 
this prism, Islam can only commit to tolerance as an apology, a pale substi-
tute for its failure to live up to its normative ideal. This ultimately derives 
from and perpetuates the myth that only modern, secular orders can ac-
commodate falsehood and that tolerance and critique must remain modern, 
secular monopolies. Islam being neither modern nor secular, we should 
not even seriously expect it to rise to such a challenge. To be fair, I sin-
cerely doubt that this is Professor Khalil’s consciously held position. But 
if a religiously grounded program as explicit in its commitment to toler-
ance as is al-Ghâzalî’s Fayṣal al-Tafriqa bayna al-Islam wa al-zandaqa 
can be so thoroughly misapprehended and seamlessly appropriated to the 
lone cause of pursuing and policing truth, one can only wonder how con-
scious one’s commitments to liberalism have to be for the storied prisms 
and historically driven presuppositions of the modern West to inform ‒ and 
inform thoroughly – one’s scholarly perspectives and interpretive thrusts. 

Endnotes

1.  Atif Khalil, “Is an Intra-Islamic Theological Ecumenism Possible? A 
Response to Sherman Jackson,” The American Journal of Islamic So-
cial Sciences 22:4 (2005): 84‒95.

2.  Sherman A. Jackson, On the Boundaries of Theological Tolerance in 
Islam: Abū Ḥāmid al-Ghāzalīʼs Fayṣal al-Tafriqa bayna al-Islam wa 
al-zandaqa. (Karachi, India: Oxford University Press, 2002), 5. Empha-
sis added here.

3.  Nowhere is this perhaps more vividly demonstrated than in his moving 
autobiographical testimony in al-Munqidh min al-Ḍālal (Deliverance 
from Error).

4.  See, incidentally, Jackson, On the Boundaries of Theological Tolerance 
in Islam, 54, where none other than the reputedly puritanical Tradition-
alist Ibn Taymîya expresses a similar ecumenical attitude: “[A]nyone 
who considers this carefully will know that many of the people of un-
disciplined passions and unsanctioned innovation (bida‘/s. bid‘a) may 
be ignorant, wrong-minded Believers who have simply veered away 
from some aspect of what the Prophet brought (mu’min mukhti’ dâll ‘an 



The American Journal of Islamic Social Sciences 28:4158

ba‘d mâ jâ’a bihi al-rasûl) just as they may be Hypocrites and Crypto-
infidels who pretend to be other than what they are.” 

5.  I must beg the indulgence of contemporary Twelver Shiites here, as my 
aim is neither to endorse nor advertise al-Ghâzalîʼs invective but simply 
to highlight his deeper and more important point in the context of the 
present discussion. 

6.  Jackson, On the Boundaries of Theological Tolerance in Islam, 119‒29.
7.  In a modern context, in other words, Fayṣal would not see it as its task 

to make a blanket judgment on whether or not modern science is a “le-
gitimate” tradition or interpretive backdrop. Nor would its primary in-
terest be in pronouncing which of the interpretations spawned by mod-
ern science was substantively correct. Rather, Fayṣal’s focus would be 
on which interpretations generated by modern science is substantively 
wrong enough to take its proponent outside the fold of Islam. And in 
this context, any number of wrong scientifically based views might be 
tolerated even if not endorsed. 

8. See Jackson, On the Boundaries of Theological Tolerance in Islam, 123. 
I do not wish to put too sharp a point on al-Ghâzalî’scritique of kalām. 
After all, he makes exceptions for those who need it to clear up doubt 
and for those who wish to learn it either to treat those beset with doubts 
or to beat back the attempts of adversaries who seek to undermine the 
doctrinal integrity of the faith. Still, he sees kalām as far from the ideal, 
normative or necessary backdrop or methodology for a proper let alone 
exclusively true reading of scripture. 

9.  Of course, notable differences in the historical, political and institution-
al context in which they wrote underwrite obvious differences between 
the outlook, concerns, and approach of al-Ghāzalī and Locke. 

10.  Shockingly, Khalil states that the second part of my introduction merely 
deals with “the historical background of the work,” and that, as such, 
“This article is only a response to the first introduction” (94n3). Given 
that any number of issues directly germane to his critique are dealt with 
in the second part of the Introduction (for example, al-Ghâzalî’s critique 
of kalām, his view on the role and function of reason and its relation-
ship to truth, the fact that theologians routinely disguise their eisegesis, 
etc.). I will leave it to the reader to determine the cost of this omission 
on Khalil’s part.

11.  By “objective,” I mean simply equally available and accessible to all.
12.  Speaking inter-religiously, this insight might go some way in explain-

ing, at least in part, why Islam did not produce a Treaty of Westphalia 
and all that came in its train. Whereas European kings, magistrates and 
prelates may have been less willing or able to recognize any difference 
between tolerating and endorsing heresy or infidelity, this distinction 



Jackson: Not Truth But Tolerance   159

was woven into the very warp and woof of Islam, from the (now much 
maligned) accommodation of dhimmīs, to al-Ghāzalī’s’s Fayṣal, to the 
Ottoman Capitulations.

13.  Which is not to say that belief is beyond the analytical function of rea-
son. Reason, in other words, can look back on belief, but it cannot get 
fully behind it in order to push it into place. On this point, see, for 
example, R. M. Unger, Knowledge and Politics (New York: The Free 
Press, 1975), esp., 29‒62. 

14.  This is a paraphrase of Stanley Fish, “Postmodern Welfare: The Igno-
rance of Our Warrior Intellectuals” Harper’s Magazine, July 2002.

15.  As we saw, for example, with his depiction of the Shiite doctrine of oc-
cultation. Again, for al-Ghâzalî, this doctrine was not relatively wrong; 
it was absolutely wrong, as he put it, “ a false, clearly absurd and ex-
tremely abominable doctrine.” Yet, it did not constitute kufr and was 
thus not a justifiable basis for excluding Shiites from the pale of the 
faith.

16.  This is what routinely dogs religious students in the secular Western 
academy. There they are taught, directly or indirectly, that they can only 
know what they can prove; and the only acceptable proof of anything 
is rational proof! Compare, this, however, with the myriad sources of 
valid knowledge and belief adumbrated by al-Ghâzalî above (and else-
where in Fayṣal). Of course, even al-Ghâzalî would agree that reason 
is the only (or most likely means) of negotiating the validity of one’s 
beliefs in the public square. But a more honest and sympathetic position 
for the academy to assume would be to acknowledge more explicitly 
the difference between what we can know (or justifiably believe) and 
what we can prove rationally ‒ and that one’s inability to prove the ex-
istence of X on rational grounds is not the same as proving that X does 
not exist. 

17.  Speaking against the pretensions of such rationalist fundamentalism, 
al-Ghâzalî insists in another context: “You have erred in stating that rea-
son (al-‘aql) is a motivator (dâ‘in). Nay reason is only a guide (hâdin), 
while impulses and motives (al-bawâ‘ith wa al-dawâ‘î) issue from the 
self (al-nafs) based on information provided by reason.” Al-Mustasfâ 
min ‘ilm al-usûl , 2 vols. (Cairo, Egypt: al-Amîrîyah Press, 1322/1904), 
1: 61. For a fuller treatment of this point, see my, “The Alchemy of 
Domination? Some Ash‘arite Responses to Mu‘tazilite Ethics,” Inter-
national Journal of Middle East Studies, no. 31 (1999): 185‒201. 

18.  Indeed, among the oft-repeated themes in the Qur’ān is that reflected in 
such verses as, “Then you will be returned to the Knower of the seen 
and the unseen, whereupon He will inform you concerning that about 
which you differed” (9:94). See also: 3:55, 5:48, 6:164, 16: 92, and pas-
sim.



The American Journal of Islamic Social Sciences 28:4160

19.  Nor did I say, or imply, pace Khalil, that, just because a doctrine is 
grounded in human thought and is thus not historically transcendent, it 
can never be true. 

20.  Again, Khalil thoroughly misses the point when he argues that, “mind-
less assent” “may have been the position of a small number of Muslims 
in history, the ‘we-believe-without-asking-how’ group, but this is not 
al-Ghâzalî’s position as he articulates it, for example, while defending 
the theological enterprise in the second section of Qawâ‘id al-‘Aqâ’id 
(The Principles of the Creed), the second book of his Iḥyâ’ ‘Ulûm al-
Dîn” (89). First, the issue is not at all one of assent but of the degree 
to which such assent can be imposed on others as a normative duty. 
Second, al-Ghazâlî’s Iḥyâ’ is a very different book from his Fayṣal. 
Richard M. Frank in Al-Ghazālī and the Ash‘arite School (Durham, 
NC: Duke University Press, 1994) and others have noted the palpable 
evolution in al-Ghâzalî’s thought, which should warn us against assum-
ing that all of his works are tied together by a common aim. Third, in 
Fayṣal, al-Ghâzalî directly – at least in part ‒ contradicts this view of 
Khalil: “[I]t had to occur to them that there were groups of uncivilized 
Arabs during the time of the Prophet* and the Companions, may God 
be pleased with them, who were steeped in idol-worship and who de-
voted no attention at all to systematic proofs (‘ilm al-daîl) – and even 
if they had devoted attention to this they would have not understood it 
– who in the end were adjudged (by the Prophet and the Companions) 
to be Muslims.” Jackson, On the Boundaries of Theological Tolerance 
in Islam , 120‒21. See also p. 122 for a similar statement.

21.  I never said, incidentally, pace Khalil, that the Hanbalites were “Process 
Theologians” (87).

22.  Jackson, On the Boundaries of Theological Tolerance in Islam, 12‒13.
23.  See, for example, the conclusion in the second part of the introduction 

to On the Boundaries of Theological Tolerance in Islam, 66‒68 for a 
confirmation of this.

24.  Ibn Manẓūr, Lisân al-‘arab, 6 vols. (Cairo, Egypt: Dâr al-Ma‘ârif, 
n.d.), vol. 5, 3412‒13.

25.  And we might recall that the falāsifah (Muslim Neoplatonic philoso-
phers) are the only ones he explicitly places outside the boundaries of 
tolerance that he defines in this text.

26.  Most notably perhaps the assertion that: “Some of the dilemmas he 
[Jackson] encounters in his attempt to formulate a coherent theory of 
theological diversity seem to lie in his use of diverse and conflicting 
strains of thought, which range from al-Ghazâlî and African studies to 
Christian theology and postmodernism. It was perhaps only inevitable 
that a project that synthesized ideas from such a wide range of disparate 
sources would encounter its own incoherencies – interestingly, the same 



Jackson: Not Truth But Tolerance   161

dilemma al-Ghazâlî’s own writings face” (92). Let me just add here 
that it seems strange then that Khalil would criticize me for not adding 
yet another “strain of thought” in the person no less of Muḥyī al-Dīn 
Ibn ‘Arabī! On another note, I would also address, inter alia, Khalil’s 
tendency toward what appears to be an unstoried, sola scriptura of sorts 
that all but ignores the spontaneous understandings of the Prophet’s 
generation and their (real or purported) ideological heirs ‒ leaving rea-
son as the sole determinant of the kinds of normative meanings that can 
be legitimately derived from or read into scripture. 

27.  In this regard, see, for example, my “Jihad: Between Law, Fact and Ori-
entalism,” Mélanges de l’Université Saint-Joseph 62 (2009): 133‒48, 
where I discuss the views of Patricia Crone on jihad as a legal institu-
tion.

28.  On this point see my, “The Beginning of History? Between Islamic Law 
and the Nation-State,” Oxford Handbook of Islam and Politics, ed. J. 
Esposito and E. Shahin (New York: Oxford University Press, forthcom-
ing), where I discuss the views of Abdullahi an-Na‘īm on Islam and the 
secular state; see also my “Legal Pluralism Between Islam and the Na-
tion-State: Between Medieval Romanticism and Modern Pragmatism,” 
Fordham International Law Journal, vol. 24 (2007), 101‒109.