Pakistan’s Problems: More Hypotheses

By Anjum Altaf

Christopher Hitchens had offered a hypothesis in Vanity Fair that Pakistan’s problems stemmed from deep-rooted sexual repression. The evidence for this was the occurrence of honor killings, and the consequence other morbid symptoms that transformed the country into one that was “completely humorless, paranoid, insecure, eager to take offense, and suffering from self-righteousness, self-pity, and self-hatred.”

Even if one were to accept the broad characterization as correct, it is difficult to take the hypothesis itself seriously. In my response, I had assumed that just a cursory consideration of the fact that honor killings occurred in India as well would have been enough to discredit the hypothesis because none of the morbid consequences are to be observed in India.

However, for various reasons, that did not turn out to the case and I had to spell out the hypothesis and how it could be tested in greater detail. I am reasonably sure that no support would be forthcoming for the hypothesis and we can move on to propose more plausible explanations for the problems that undoubtedly exist in Pakistan.

In doing so, I am leaving aside the issue of what might offer a better explanation for honor killings themselves. Readers interested in this phenomenon are invited to submit a guest post to the blog to continue the discussion (it can be sent to

In my response to Hitchens, I had quickly moved on to what I considered a considerably more plausible hypothesis, one similar to that offered by Lawrence Wright in his article in the New Yorker: that American aid undermined the strategic relationship between the US and Pakistan and “created monstrous contradictions within Pakistan itself” (this can be called the Wright hypothesis for purposes of this discussion).

It is interesting to note that after the sensational introduction of his sexual repression hypothesis, Hitchens himself abandoned it and devoted the bulk of his article excoriating the US and holding it co-responsible for the unacceptable outcomes of the US-Pakistan strategic relationship.

However, a careful reading would suggest a critical difference between the Hitchens and Wright hypotheses: In the Hitchens hypothesis, the US embraced a state that was already degraded because of sexual repression; in the Wright hypothesis, it was the US embrace that was responsible for the degradation. There is a crucial difference in causality and if the Wright hypothesis is correct, it was quite fortunate that the US did not embrace India instead and that the Indian leadership was wise enough to reject any such overtures if they were made.

Since we have already rejected the Hitchens hypothesis, we are left to evaluate the Wright hypothesis with more care. In my response to Hitchens I had suggested that the Wright hypothesis could only be part of the story. Since the US never embraced the people of Pakistan, only its ruling apparatus and often only a military dictator, its impact could well explain the perversities and deformities of the Pakistani state but could not explain the degrading conditions that include honor killings, gender violence, absolute poverty, malnutrition, children out of schools, etc. There has to be some other or additional explanation for these conditions.

Given that these conditions exist to varying degrees in India as well, and leaving aside the fact that they are being addressed with a great deal more seriousness, it should be clear that the American embrace alone cannot account for their presence in Pakistan. I had suggested two other explanations. First, I suggested that colonial interventions in Third World countries had interrupted the processes of indigenous industrialization and urbanization that were the critical underpinnings of social change in the ‘first movers’ that had themselves not been hampered by external interventions.

As a result, the social revolutions and the leveling of social hierarchies that marked the transition from monarchy to representative governance in Europe never occurred in India. The hierarchical social structures survived intact and it is now the democratic process that is working to achieve the effects that social revolutions did in Europe. But an evolutionary process is inevitably much, much slower than revolutionary change. Thus it is not surprising that social inequalities along with their manifestations continue to persist in Third World countries.

Second, almost since the independence of India and Pakistan, the world entered a neo-liberal global order in which the problems of the poor were accorded a much lower priority compared to maintaining a high rate of economic growth; the fruits of latter was assumed to trickle-down to the poor via the magic of the market. The Ayn Rand ideology of the free market in fact shifted, in practice if not in words, the blame for the problems of the poor on to the poor themselves.

These meta-hypotheses can go some way to explain the common existence of extreme social maladies in both India and Pakistan and their very slow redress while the American embrace can explain the additional deformities of the Pakistani state that are not shared with India. Of course, the deformities of the Pakistani state and its attempts to legitimize itself have a harmful feedback effect on the attempts to address social maladies which is why the direction of change is negative in Pakistan while it is positive in India and the trajectories of the two countries continue to diverge.

These hypotheses are tentative and offered merely to catalyze the discussion that is necessary to improve our understanding of the economic, social and political changes taking place in South Asia. Such an understanding is also needed to fully appreciate the very pressing needs of the majority of the citizens in the region.


  • Arun Pillai
    Posted at 21:52h, 01 July Reply


    I’m sorry to be persistent with my perspective on these matters. I hope you don’t mind.

    I think you have quoted only a small fragment of Wright’s two paragraphs which are pertinent in this context:

    “India has become the state that we tried to create in Pakistan. It is a rising economic star, militarily powerful and democratic, and it shares American interests. Pakistan, however, is one of the most anti-American countries in the world, and a covert sponsor of terrorism. Politically and economically, it verges on being a failed state. And, despite Pakistani avowals to the contrary, America’s worst enemy, Osama bin Laden, had been hiding there for years—in strikingly comfortable circumstances—before U.S. commandos finally tracked him down and killed him, on May 2nd.

    American aid is hardly the only factor that led these two countries to such disparate outcomes. But, at this pivotal moment, it would be a mistake not to examine the degree to which U.S. dollars have undermined our strategic relationship with Pakistan—and created monstrous contradictions within Pakistan itself.”

    Wright fully acknowledges that American aid is “hardly the only factor … .” But you cite only the aid factor which you have already dealt with in some detail in your other articles.

    Would it not be more interesting to bring these other factors to light? And are they all simply factors involving the colonial powers? Are there no so-called internal factors, as I have referred to them earlier? Does Pakistan not bear any responsibility for the state it is in?

    • Anjum Altaf
      Posted at 03:18h, 02 July Reply

      Arun: Yes, go ahead and suggest some hypotheses for discussion.

  • Aakar Patel
    Posted at 03:04h, 02 July Reply

    My hypothesis: Lack of caste balance.
    Conversion of northern mercantile castes to Islam was limited. In West Punjab there are some khatris (Najam Sethi is one), including many from the area called Chiniot. They are exceptions.
    The dominant Punjabi mercantile castes were Hindu (Kapoor/Khanna/Gujral) though many Khatris also followed Nanak (Manmohan).
    So long as Punjab was united, its economic pattern was similar to that of all Indian states. Upper caste middle-class, including some but not many Muslims, and mercantile castes dominated the city and its economy. They peopled academies. The peasantry when it took to professions usually went to the army.
    In Pakistan, this balance – I’m not viewing it as Platonically perfect even in India, mind you – is absent because 1947 was a partition of castes. The merchants and traders left, and the peasantry dominated areas they never had before.
    My observation is that in places where caste balance remains (in Sindh where Karachi’s mercantile Gujaratis neatly took over from the Hindus), Pakistan is more normal.
    The problem is Punjab. It shows the problems of an incomplete culture. Its domination explains to me the tendency of the state to be immoderate, and interested in ventures that the peasant might find more appealing than the merchant.

    • Anjum Altaf
      Posted at 03:23h, 02 July Reply

      Aakar: Thanks. That is a very interesting hypothesis. It makes a lot more sense than Hitchens because it has room for regional variations. Just as dividing up an ecological unit can have damaging consequences so can dividing up a sociological unit, if this hypothesis holds.

      • Aakar Patel
        Posted at 04:30h, 02 July Reply

        Digression, but am curious.
        Do Pakistanis see those from mercantile castes for their caste?
        Are the qualities of Edhi/Hoodbhoy/Jinnah seen as emanating from their caste?
        I use these names because they are all Gujarati. In Gujarat, their moderation and quality would immediately be attributed, rightly or wrongly, to their Memon/Khoja backgrounds.

        • Anjum Altaf
          Posted at 05:30h, 02 July Reply

          Aakar: This is not an easy question to answer but let me offer some tentative thoughts. While caste consciousness is very much present in Pakistan it is not something that Pakistanis are comfortable with intellectually because it is not supposed to be present. Therefore, its understanding is fuzzy , more social than economic. There is always the mention of Chiniotis in the Punjab and of Memons in Karachi. It is just assumed there is something special about them without tying it back to any caste antecedents. There is a lot of social stratification in the Punjab within the peasant castes (transmuted into Biraderis) but not much work has been done on it. The last serious study I recall was Caste and Class in a Punjabi Village by Saghir Ahmad (Eqbal Ahmad’s brother) and that must have been in the 1960s.

          Some time back I had mentioned how I had been struck by the hypothesis of the late GM Mekhri about the dominance in Pakistan of the Gujarati mercantile castes that converted to Islam (Memon/Khoja/Bohri). He had tied it explicitly to the economic dimensions of caste and then added the hypothesis that these groups sustained their dominance over generations because they retained their old inheritance customs. But Mekhri’s work was done even earlier, in 1947. By the way, I have finally tracked the PhD dissertation courtesy of a retired professor of sociology at the University of Bombay. It is in the library and she has promised to read it on my behalf before we figure out how to obtain a copy (it is a not-for-loan manuscript in a condition not yet ascertained). For reference, here are the links to my posts about GM Mekhri:

          What you have suggested is a very powerful hypothesis and adding Mehkri’s ideas to it would make for a great PhD study on the socio-economic consequences of a systemic shock that entails an imbalanced exchange of castes. It could cover the Punjab, Sindh, NWFP and Bengal for understanding regional variations. Of course, there are purely accidental factors in how the consequences of the shock played out – the fact that Punjab happened to be the dominant province in Pakistan which incidentally is likely to be at the bottom of many other explanations.

          • Aakar Patel
            Posted at 05:39h, 02 July

            This is something we have discussed before, but will revisit your Mekhri links.
            I’m delighted your persistence has paid off.
            I’m just as keen to access the Mekhri dissertation.

          • Aakar Patel
            Posted at 05:55h, 02 July

            Since you mention the state, another thing that strikes me is that Bengal (which has no mercantile castes – Hindu or Muslim) is not particularly different from Bangladesh once British Calcutta and its elite are removed from the equation.

          • Anjum Altaf
            Posted at 06:11h, 02 July

            Aakar: That’s an interesting point about Bengal. There was discussion on the blog some time back you might find useful. Just track the next few comments beginning here:


            This, in particular, could help us pursue your suggested hypothesis further:

            “I have found the following breakdown of population by religion (percentages) for Punjab in the paper Demography of the Punjab by Gopal Krishan (page 83):

            Census Year 1881: Hindu (44) Muslim (48) Sikh (8) Christian (-) Other (-)
            Census Year 1941: Hindu (29) Muslim (53) Sikh (15) Christian (2) Other (1)

            The following explanation is given for the change: “A big erosion in the percentage share of Hindus was caused by the conversion of many of them to Islam, Sikhism and Christianity. Such a change of religion was much more typical of lower castes among the Hindus, such as chuhras, chamars, jhiwars, and malis. Conversion was negligible from the higher castes, such as Brahmins, Aroras, Khatris, and Aggarwals.”

            The important point to keep in mind is that this heavy conversion was well into the period of the British Raj and after the end of the Mughal dynasty in 1857.”

          • Aakar Patel
            Posted at 06:36h, 02 July

            This is splendid information. Will pursue it and look to you for guidance.
            My guess is that the big numbers actually came from the conversion of an unmentioned “middle” caste, the Jat peasantry. I have somewhere a study on the Punjab army of that time, by Tan Tai Yong.
            Should I find it, will take up this thread again. I suspect the conversions have something to do with recruitment.
            That ask-a-question link on conversions is most fascinating. Will spend some time on that too.
            My own view is that conversions in South Asia were secular and not religious in motivation.

  • Arun Pillai
    Posted at 03:24h, 02 July Reply

    I do not know Pakistan well enough to generate hypotheses. All I know is that people on the left typically blame colonialism for all the world’s ills and people on the right blame the colonized. I believe an insightful analysis ought to be able to transcend these knee-jerk responses.

  • Haris
    Posted at 10:56h, 02 July Reply

    The disempowerment of nascent capitalists (essentially mercantile castes) and the privileging of peasant-soldiers was underway in Punjab before the demographic change associated with the partition. The Land Alienation Act of 1900 divided society, in revenue terms, between agricultural and non-agricultural tribes. Imran Ali (1989, Punjab Under Imperialism 1885-1947) has shown that the British colonial government made concessions to the mostly Muslim (but also Sikh and Hindu) Jats, Rajputs and others who feared capitalist expropriation at least partly in the expectation that the yeoman-farmer was a solid support base for British rule. In fact, it might be argued that the Land Alienation Act and the class-caste politics around it forms an important backdrop to the partition of Punjab – at least as important as the communal factor.

    I have tried to draw this connection in (section 2 of) a speculative essay on the history of land and reform. This essay has been published in:

    Panos South Asia (2011), Leveling the Playing Field – A Survey of Pakistan’s Land Reforms,

    and available at:

    Rough caste analysis of partition demographics is possible using the 1931 census. We are doing something like that, primarily on the ‘lower’ castes (see Table 1.2: , and could do this for mercantile castes too.

  • neel123
    Posted at 17:04h, 02 July Reply

    @ Anjum Altaf and the knowledgeable contributors,

    There was a well calculated push by the collaborators ( read the Anglo-American powers) of the Pakistani ruling establishment ( read the military ) to steer Pakistan towards the Saudi version of (extreme) Islam, that would drive Pakistan in a direction opposite to that of India, knowing fully well that Indian culture for thousands of years is based on tolerance and accommodation of all faiths, races and culture.

    The ultimate objective was to prevent formation of a powerful entity and leave a permanent divide in order to weaken and destroy the potential.

    For the Pakistani ruling establishment, the carrots were many fold, (1) constant flow of Economic and military aid
    (2) nuclear weapons
    (3) the most important of all is the assurance of legitimizing Pakistani leadership over the Sunni world. The recent Pakistani military intervention in Bahrain to preserve American interests is a glaring example of this unspoken big deal.

    It is win-win all the way for the Anglo-Americans and needless to say, they have succeeded in their evil designs with Pakistan. Now they have their eyes on India, arming India in order to make it comparable to the Chinese in terms of destructive capabilities.

    All they expect to see now is destructive wars between the nations in the region (of thousands of years of civilizations) , that would eliminate any future challenge to the hegemony of the Anglo-American powers in Asia.

    Is it not time for China-Pakistan-India to be able to see through ….. ?

    • Anjum Altaf
      Posted at 08:05h, 05 July Reply

      neel123: I wouldn’t go as far as to say that there was a conspiracy to steer Pakistan towards the Saudi version of Islam. There might well be but it is not something that can be asserted without more evidence. I feel after the 1979 Islamic Revolution in Iran, the Saudis had incentive enough to shore up their ideological defenses. The coincidental Soviet invasion of Afghanistan in the same year and the American response to promote Islamic jihad complicated the picture considerably and it became hard to unravel the strands. On the Indian side, the inability to see and counteract what was happening in Pakistan, for whatever reasons, can be considered a political failure given its own stake in a peaceful region. There was no conspiracy that forced India to ignore its long-term interests in this manner.

  • Aakar Patel
    Posted at 04:59h, 03 July Reply

    This is valuable, thank you for it.
    Will respond after I have had a chance to go through your two papers.

  • Anil Kala
    Posted at 06:02h, 03 July Reply

    Neel: I suppose all significant political moves are well calculated but your theory about Anglo-American push in Pakistan to opt for Arab kind of Islam seems too far fetched. I don’t think Pakistan needed to be driven anywhere to be anti Indian. India and Pakistan were born to be natural adversaries, if any push was needed it was to set up a forced equilibrium, to make them friends.

    The second part is also bogus. When India was economically weak it still followed an independent foreign policy, now that it is much better off economically, it will be toeing somebody else’s line is really laughable.

    Personally I am very happy that the only superpower is US, not Russia, China or Saudi Arabia.

  • neel123
    Posted at 13:54h, 03 July Reply

    @ Anil Kala,

    Clearly you think with your “ghutna”, have no knowledge of the history of British evil in the Indian subcontinent, and have no idea about how the Anglo- American policies towards the region in general and Pakistan in particular have undermined India’s security interests in the past five decades, cleverly designed to put all the nations in the region at risk.

    • Anil Kala
      Posted at 15:12h, 03 July Reply

      Neel: Tell me where I said Anglo-American policies on subcontinent did not undermined India’s security. I merely refuted your assertion about them driving Pakistan to Arab brand of Islam.

  • Mubbashir Rizvi
    Posted at 20:55h, 03 July Reply

    Re: Punjab Castes and Pakistan Troubles Hypothesis

    I will like to add to this very insightful exchange about caste-class politics in Punjab by stating a few observations.

    Firstly, it is important to note that large swath of Punjab up till 1880’s was made up of a very different pastoral economy with different social structures, and trade links with Afghanistan and Central Asia (Neeladri Bhattacharya’s work is important here). For a large part of this region the competitive caste-class politics as we know it started with the large irrigation projects. Hence it will important to think about the role of technologies in remaking the environment and rural society.

    The canal colony settlements were based along caste and biradari ties to keep social cohesion, which meant that entire villages were settled by a peasant biradari given its aptitude for dairy farming, vegetable produce, etc, The caste equation in canal colonies differed from Eastern Punjab which had a substantial Urban Muslim presence in Amritsar, Jullundar, which included a large population of Kashmiri Muslim merchant families. Whereas the new urban centers in Sahiwal, Lyallpur, Okara, Sargodha were dominated by Hindu and Sikh merchant classes. This economic transformation also had major impact on older cities like Lahore which was the center of administration, press, and higher education

    In Western Punjab the rural regions were predominantly made up of mostly Muslim, Sikh peasant but there was a large community of Hindu Jats who were a major source of headache for Arya Samaj

    Secondly, going back to Aakar’s point about the flight of Hindu merchant communities, from my own observations in Okara district it seems that much of the merchant network was replaced by Muslim Rajputs (like the Rao’s), old landholding families like the Wattu’s, and other merchant families who resettled here from eastern Punjab. Whereas, merchants from Western UP settled in the urban centers of Southern Punjab -seraiki areas..

    As we can infer from Imran Ali’s work, the class-caste equation in this region is tied to notions of development, regional modernity and a very tight relationship with a paternalistic bureaucracy. The recruitment of peasant soldiers was another crucial element as the canal colony land was used to offer incentives for recruitment but also the large infrastructure of new cantonments, services needed, dairy farms, ghori pal schemes created a large economy servicing British Indian Army (Mazumdar discusses this in a book on Punjab and the British Army).

    During this time Punjab became the biggest agricultural production, the largest number of peasant recruits in the Army and relatively less upsurge of nationalist sentiment. However, it also became the site of intensive of competition when it comes to positing the interests of rival communities.

    Hence, ordinary agriculture became the most politicized in some ways but not in the ways we recognize it in terms of nationalist struggle but in terms of competition along the fault-lines of caste-class lines that took on communal overtones.

    I think Pakistan Punjab’s continued faith (but now faltering) in state institutions has to do with this unique history of what one can call regional modernity that intensified the links between colonial state and rural population but it also increased the competition within caste-class communities as they vied for patronage, reform, state support with disastrous consequences in 1947.

    Since independence the Punjab has been one province that has bought into the state narrative, the need for administrative solutions and the rejection of popular politics in support of the Army. These things might be changing now but there is still a certain faith in state institutions that might go back to the points I discussed above (1) the rise of a new kind of peasant settler society which has a very different relationship to local environment, and history (2) administrative elective governance organized around caste-class politics (3) rise of new families of industrialists (packages group, nishat group, heraj group) (4) proximity to state institutions like the Army

    I just wanted to broaden out what people have mentioned already.

    Neeladri Bhattacharya discusses pastoral economies in an anthology edited by Claude Markovtis

    R. Mazumdar discusses the role of the Army in the making of Punjab as we know it

    Kessinger’s Vilayatpur and Talbot’s book on Amritsar discusses the large presence of Muslim in urban centers of Eastern Punjab
    like Amritsar

  • Arun Pillai
    Posted at 23:38h, 03 July Reply

    I find all this detail very interesting but I am still mystified by what the argument for the role of caste in Pakistan’s current problems is. Aakar Patel, as I read his post, seemed to have a clear hypothesis: that the absence of mercantile castes led to a lack of capitalist development. Is this reading correct?

    Could someone clarify for the benefit of those who do not know these social nuances – like me – what the actual argument for Pakistan’s problems is. What problems are being analyzed and what is the role of caste that have led to these problems?

    Otherwise, it all remains rather descriptive.

    Incidentally, I have met Neeladri and am broadly aware of such work done by historians and others. But I am interested for the moment in the broad picture. I would be most grateful.

  • Arun Pillai
    Posted at 23:58h, 03 July Reply

    Just to clarify how I see Aakar’s argument:

    1. Caste imbalance, flight of mercantile castes (historical fact, which I accept)

    2. In the context of South Asia, mercantile castes are required for capitalist development (theoretical hypothesis)

    3. Therefore, relative lack of capitalist development in Pakistan (conclusion from #1 and #2)

    The other posts I see as descriptive elaborations of #1 above. I have two questions:

    a. Have I read the basic argument correctly? If not, what is the argument being made?

    b. Is #2 above true? Can one give a convincing argument for it instead of simply assuming it?

    • Haris
      Posted at 04:59h, 04 July Reply

      I read Aakar Patel’s argument slightly differently:

      1. Sociological India includes most parts of most South Asian states, hence a comparative perspective is helpful.
      2. The social material for the class of nascent capitalists in sociological India is provided by traditional mercantile castes.
      3. A mostly unnoticed consequence of the communal partition of sociological India was the eviction and disempowering of the traditional mercantile castes in Pakistan.
      4. This explains an imbalance in Pakistan (taking sociological India as benchmark), which in turn explains why the constituency for capitalism and democracy remains weak.

      I agree with #1, and am very sympathetic to #2 (there is empirical regularity), though it is useful to acknowledge that caste identities change over time too. I added to #3 that the politics of disempowering mercantile castes in western Punjab (and possibly Sindh) predates the communal partition of sociological India, and can be traced back to early 20th C Punjabi class politics and British political strategy. This means that perhaps #3 did not cause the imbalance, but was itself caused by a prior imbalance. British imperial strategy required the over-militarisation of western Punjab, and Punjabi politics took care of the rest.

      This reading is optimistic. Because while we cannot turn the demographic wheel, we can do something about the over-militarisation of western Punjab, or indeed the disproportionate power a militarised Punjab enjoys within Pakistan. The politics of doing precisely this is moving ahead nicely.

      • Aakar Patel
        Posted at 06:34h, 04 July Reply

        Yes, that’s how I see it.
        To demonstrate the power of caste in Indian capitalism, here is a link which shows that 26 of the 55 Indians on Forbes’ list of global billionaires are from the small (under 1% of population) twice-born vaish caste. Many of these 26 are self-made, first generation businessmen with no more access to capital than the rest of us. Of the two Muslims on the list, one (Azim Premji) is Khoja, and many of the others are also from mercantile groups.

        • Anjum Altaf
          Posted at 12:21h, 04 July Reply

          Aakar: In this context there is something we can add to the discussion. I recall an earlier op-ed you wrote regarding the growing divergence in the caste compositions of the corporate and political elites in India. I had written a note about Pakistan in 1982 when the Galtung Hypothesis had been newly proposed:

          To describe this process formally we can make use of Galtung’s (1980) concept of ‘rank disequilibrium’ as a precursor of social change. In Galtung’s framework groups or classes are ranked on three scales representing economic, political and social status. Only when the major groups or classes are ranked similarly on all three scales is societal equilibrium stable. Otherwise a transitional period, usually marked by conflict, results during which groups contend with each other to restore a new and stable equilibrium.

          The paper itself can be useful in our discussion: the large scale migration to the middle east of what we might characterize as the peasant castes and their consequent enrichment gave dominance to a certain set of values. The way I had looked at it then, I felt this would strengthen a right-wing ideology in the country which is part of the problem we are dealing with now.

          The original paper was published here:
          I put it on the blog to provide a space for discussion:

  • Arun Pillai
    Posted at 11:48h, 04 July Reply

    Thank you both for your lucid responses. I have several further queries if you don’t mind.

    1. I don’t see much material difference between Haris’s formulation of the argument and mine. His #1 simply sets the context and plays no direct role in the argument. His #2 is similar to my #2 although there are certain differences that I point out below. His #3 and first part of #4 is a more accurate restatement of my summary #1. And the conclusions are roughly the same. I point this out just to clarify that the logic is similar.

    2. The key to the argument is #2. Both Haris and Aakar seem to see this as an empirical fact. But this empirical fact by itself is not enough; it needs to be explained.

    3. Haris’s formulation of #2 suggests that he sees the presence of mercantile castes as a sufficient condition for capitalist development whereas my #2 suggests that it is a necessary condition. The former fits Aakar’s analysis of the caste origins of Indian billionaires but if one looks at the remaining thirty billionaires who were non-baniyas, then the conclusion might be that capitalism can develop without baniyas as well. In other words, mercantile castes are not necessary but may be sufficient.

    4. To make the argument for weak capitalist development in Pakistan on the basis of the relative absence of mercantile castes, one needs my formulation of necessity; Haris’s formulation of sufficiency is too weak because it leaves open the possibility that capitalist development could have occurred in other ways. If this is correct, then #3 above suggests the argument is partly flawed because there are a substantial number of non-baniyas on the Indian billionaires list.

    5. It is pertinent to quote Dr. Robotka’s argument here: “My hypothesis is that the colonial type of capitalism which was brought into India by British has created much of the trouble that the subcontinent is facing today. Because of the top-down mode it took roots only in a very small part of Indian society and left large parts unchanged in its pre-capitalist, pre-modern state of affairs. While Indian capitalism in the Indian Union today is slowly making up for the deficits that the colonial introductions had created, still only a part (if a growing one) of Indian society is capitalist; still large parts of Indian society are out of its scope.

    Now, Pakistan comprises of the areas which were constituting the fringes of British colonial India where the introduction of capitalism was weak (Sindh) or even absent (Balochistan, KP, GB). Those areas were even hardly under British rule like Kalat/Balochistan or the tribal areas, Swat, Dir and the whole North of Pakistan and consisted of territory which was accessible only with difficulty and agriculture or industry were too difficult or impossible there. That is why capitalism is weakest in Pakistan, much weaker than in India. That means pre-capitalist and pre-modern influences in society remain dominant.”

    Her analysis is different as should be plain; she does not seem to attribute the relative lack of capitalist development to the absence of mercantile castes but to other factors involving the mode of production. Perhaps the two arguments can be combined to yield a more robust argument.

    6. It seems to me that for capitalist development to occur, a large number of factors is required: the entire environment has to be right (e.g. the Marxist legal theorist Pashukanis of the early 20th century argued persuasively that capitalism takes root when there is a sphere of law homologous to the sphere of commodities; the liberalization of the Indian economy in 1991 led to rapid economic development; the Chinese reforms of the late 1970s; the Meiji restoration of the mid-19th century; and perhaps many other such examples; even the analysis of factors like Mekhri’s which involve various constraints that operate in a culture, the institutions of science etc.). To my mind, if one or more of these environmental factors was missing in Pakistan, it could have inhibited capitalist development. This is quite likely and if this observation is correct, then again the proffered argument based on the absence of mercantile castes is weakened.

    7. Throughout this argument, the presumption has been that we were looking at the relative lack of capitalist development in Pakistan. But Anjum’s post is titled “Pakistan’s Problems”. I took this in a broader sense than just economic development. Is it right to reduce its current problems to just economic factors?

    8. Finally, I think the argument based on mercantile castes may apply to a lesser degree today because I have assumed our interest is in the period after 1947 and more particularly the twenty-first century. As Haris points out himself, castes do evolve and whether what happened at the turn of the last century and in 1947 is still as influential is moot. I personally still think a lot more argument is needed to make the case for the importance of caste for capitalist development in South Asia. A few statistical observations, however striking, may not be enough.

    • Aakar Patel
      Posted at 13:04h, 04 July Reply

      Re (3), actually even the non-Baniya component of that list is overwhelmingly mercantile. I hadn’t broken it down further for want of space. The numerically major castes are more or less absent.
      That said, I do not think caste is a necessary condition for capitalism in India. There are exceptions which show this, even if they are few.

      Re (7), you’re quite right. This isn’t purely about economic development. Anjum and I have had exchanges before and this took off from a position where we understood each other’s views.
      My hypothesis addresses the question: Why is Pakistan the way it is (a national security state with a martial mindset and a popular army)? Does it have a culture that is particular to it? If so, what defines it? My hypothesis was that it is an imbalance in its society.

      • Anjum Altaf
        Posted at 18:10h, 05 July Reply

        Aakar: Yes, clearly the issue is not about capitalist economic development but capitalist economic values and the acceptance of those values in the way you mentioned for the mercantile culture of Gujarat. It is no surprise that ‘bania’ still has a pejorative connotation in Pakistani discourse.

        We would benefit from turning to the work of Dierdre McCloskey whose 2010 book Bourgeois Dignity: Why Economics Can’t Explain the Modern World explores this crucial difference. Her forthcoming book Bourgeois Revaluation goes into more details.

        The point she makes (see the draft preface to the new book) is that it is only when bourgeois values become respectable that the old world changes. Capitalist accumulation had taken place many times earlier but it was an activity looked down upon.

        Reading her ideas made me construct a thought experiment. Imagine if the Dutch bourgeoisie at the time she is describing had all been shipped off to Poland and replaced with an equal number of peasants. What would have been the outcome? Well, we can’t say what might have happened because these things depend on many other variables. But surely we can say that there would have been some difference in the evolutionary trajectory of the society and it would have been driven by the differences in values. This impact would not have had a direct one-to-one correspondence with every event today, but should we expect such a correspondence?

        Dierdre McCloskey’s home page is here and well worth exploring:

  • Arun Pillai
    Posted at 16:52h, 04 July Reply

    The past several posts have been about Pakistan: Osama, Hitchens, Pakistan’s problems, and so on. If one takes a wider view of these issues, how does caste imbalance explain something like the following article from today’s NYT:

  • neel123
    Posted at 14:15h, 05 July Reply

    @ Anjum Altaf,

    ” On the Indian side, the inability to see and counteract what was happening in Pakistan, for whatever reasons, can be considered a political failure given its own stake in a peaceful region. There was no conspiracy that forced India to ignore its long-term interests in this manner.”

    – Would you please care to explain what you think India could have done better in its own long term interest ?

    • Anjum Altaf
      Posted at 16:19h, 05 July Reply

      neel123: I am in the process of writing this up. I started with an earlier post where I set out my initial thoughts. I suggest you read it so you get a sense of the directions in which I am thinking. Obviously there is room for differences but if we explore them in an academic fashion we can have a win-win interaction. The key to the whole argument is of course the definition of India’s long term interest. Depending on how it is identified, very different policies would make sense.

  • Arun Pillai
    Posted at 19:19h, 05 July Reply


    Your post about pro and anti capitalist values raises the natural question where these values come from. Are the anti-capitalist values prevalent in Pakistan in current times – say, post 2000 – sourced from the caste backgrounds of the relevant groups there, especially since, as you have said, caste itself – let alone baniya values – is not openly acknowledged in Pakistan?

    It also suggests that the incentive structures in the environment, notably those created by the state (like the reforms in China and India), must play some role in engendering pro-capitalist values. Are such incentives missing in Pakistan and why?

  • neel123
    Posted at 21:12h, 05 July Reply

    Dear Mr Anjum Altaf,

    In your article ( ” India’s Pakistan policy” link above ) you have had a good discussion with a number of contributors.

    My personal opinion is as follows :

    1. The Indian political leaders in the past 60 years have not been known to be strategic thinkers with a sound understanding of the dynamics of geo-politics, or any vision for the nation. Systematic institutional strategic analysis and thinking that can lead to policies, is currently in infancy in India, limited to a few individuals. It is ad-hoc ism all the way.

    2. India has always been on the defensive vis a vis Pakistan, a Pakistan backed diplomatically, financially and militarily by the mighty Anglo-American powers on one side and China on the other. Therefore reactive defence ( tit-for-tat, in your words) seems to be the only visible policy.

    As of now, India’s lack of Pakistan policy is not important, because it is the Pakistan policy of the big and influential players like the US, Britain China, Iran and Saudi Arabia, that would keep the ball rolling for Pakistan, and by extension for India.

    If the trends continue, two or three decades down the road India will be in a much better position, with its new found economic clout, to have a visible Pakistan policy.

    • Anjum Altaf
      Posted at 05:24h, 06 July Reply

      neel123: My response would be the following:

      1. There is no dearth of strategic thinkers in India. After all, the policy of non-alignment was a very strategic and successful one; it kept India from the pitfalls that hurt Pakistan. But this strategic thinking was never applied systematically to Pakistan. One can speculate on the reasons for this phenomenon.

      2. You could be right. Perhaps India felt on the defensive about Pakistan. I am not fully convinced yet. If something is so critical, it makes all the more sense to have a policy. India has an Afghanistan policy even though all the same powers are involved. Why couldn’t it have a Pakistan policy? Of course, one must keep open the possibility that there was Pakistan policy all along; it is just that we don’t know about it.

      3. I disagree that as of now India’s lack of a Pakistan policy is not important. Two or three decades down the road might well be too late.

  • neel123
    Posted at 13:19h, 06 July Reply

    @ Anjum Altaf,

    ” Two or three decades down the road might well be too late.”

    – there is a clear threat in the above statement.

    1. If it is a threat for Pakistan, India could not care less.

    2. If it is a threat for India, well we Indians are used to such threats and blackmails from the Pakistani establishments and media personalities, where use of nukes is implicit. India’s stance would be to strengthen its own defense rather than respond proactively to Pakistani provocations, while prepare for the worst case scenario.

    • Anjum Altaf
      Posted at 13:48h, 06 July Reply

      neel123: It was not intended as a threat just as a statement of fact. If India waits for two decades to formulate a Pakistan policy, Pakistan might have collapsed by that time. And a collapsed Pakistan would be a threat to the whole region. It was a reiteration of the common understanding that prevention is often less expensive than cure. My point in the post about India’s Pakistan policy was precisely to argue that the attitude ‘If it is a threat for Pakistan, India could not care less’ is shortsighted and contrary to India’s own interests. But as I mentioned the entire argument turns on our understanding of what is India’s long term interest and how best is it to be served. On that opinions can differ a great deal.

  • neel123
    Posted at 15:08h, 06 July Reply

    Dear Mr Anjum Altaf,

    Pakistan, a million strong army and a 160 billion dollar economy, will never collapse unless the Pakistani ruling establishment does something really stupid and as long as the Anglo-Americans are with Pakistan.

    Indications are, the Americans are too scared of the Pakistani nukes and will never leave Pakistan alone. And as far as the Pakistani establishment is concerned, they are playing a game of brinkmanship, a game of high stake poker, testing the resolve of the adversaries as well as the collaborators to the limit.

    • Anjum Altaf
      Posted at 16:35h, 06 July Reply

      neel123: I consider it unsafe to make categorical statements that something will ‘never’ happen. The Soviet Union had a bigger army, a bigger economy and many more nukes – it collapsed. Pakistan may or may not collapse – time will tell. Its ruling establishment has already done some incredibly stupid things and more cannot be ruled out with any confidence. In high stake poker, the possibility of over-playing one’s hand is always present.

  • Anjum Altaf
    Posted at 14:02h, 07 July Reply

    Two articles provide a nuanced analysis of the problems in which Pakistan is enmeshed. The first provides details of the divisions within Pakistan and helps to move beyond generalizations that treat the country as a person or unitary actor:

    Pakistan’s Army: Divided it Stands by Pervez Hoodbhoy

    The second details the historic trajectory of Islamism and helps overcome the notion that the problem is religious and not political:

    America’s secret romance with Islamism by Praveen Swamy

    Readers can disagree with the specific conclusions that the authors have derived from their analyses or their recommendations without negating the need for an approach that is historical and sufficiently disaggregated to be useful.

  • anon4cec
    Posted at 03:43h, 16 July Reply

    Aatish Taseer’s article in the WSJ hits the nail on its head:

    This is the root cause of all problems of Pakistan.

    • Anjum Altaf
      Posted at 17:51h, 16 July Reply

      anon4cec: What is the root cause of all problems of Pakistan?

      I remain unconvinced by accounts that say things like “Pakistan turned its back on all that had been common between Muslims and non-Muslims in the era before partition.” Who in Pakistan did that? Salman Taseer represents a type in Pakistan but there are other types too. Can one generalize from Salman Taseer to all of Pakistan? And even Salman Taseeer married a non-Muslim Indian; so there is more to the story than a simple rejection. I tried to explain this in a post some time back:

      I can’t quite understand why American papers give so much space to novelists to write about history and politics. See this review of a recent book by Philip Oldenburg and note how a competent professional sees through the superficial anthropomorphic accounts:

      It is not that Pakistan doesn’t have problems but they are not the ones Atish Taseer identifies as the cause of everything.

      • anon4cec
        Posted at 14:59h, 17 July Reply

        So, by your logic, VS Naipaul, Arundhati Roy or Amitav Ghosh have nothing useful to talk about India’s history and politics? Sorry, I find it ridiculous to think that nobody other than the dwellers of academic ivory towers armed with their PhDs are qualified to speak about a society’s history or politics. If anything, their insights are far more powerful and revealing than the supposedly disinterested and unbiased academic inquiries that far too often are neither.

        • Kabir
          Posted at 16:11h, 17 July Reply


          I don’t think that Anjum is saying that novelists have nothing useful to say about politics or history. Often times novelists bring insights that academics do not. However, novelists have not been trained in the rigorous methodology of research and analysis that social scientists use. Essentially someone like Arundhati Roy (whom I have great respect for) can only analyze issues from her own understanding and experience which, while valuable, is still just anecdotal evidence. I think that novelists/public intellectuals and academic analysts both serve different and useful functions.

        • Anjum Altaf
          Posted at 17:41h, 17 July Reply

          anon4cec: I guess one should assess everyone’s writings on its merits. The novelists you have mentioned are no doubt brilliant novelists and their writings contain very insightful observations on the human condition. But their non-fiction does not become privileged just because they are great novelists. The general consensus is that Naipaul’s and Roy’s non-fiction does not command the same respect as their fiction and will not get them any prizes. Nor do all PhDs have useful things to say, especially when they are talking about subjects outside the domain of their specializations. There they have to be judged like any other non-specialist on the merit of their arguments.

  • SouthAsian
    Posted at 07:46h, 03 January Reply

    One strand of the commentary on this post comprised a discussion of the caste composition of the Punjab before 1947 and the impact on it of the cataclysmic transfer of populations caused by the partition. Aakar Patel put forward the hypothesis that selective migration of specific caste groups had a negative impact on capitalist development and thereby on politics in Pakistan.

    This remains an important line to pursue. Aakar Patel has raised it again in a recent op-ed which provides an opportunity to pick up the discussion again.

    • Vikram
      Posted at 20:39h, 04 January Reply

      Some statistical evidence from the Joshua Project (I am not sure how reliable their estimates are, but they seem to make sense to me),

      Out of a total population of about 180 million, about 70 million of Pakistan’s people belong to peasant/warrior castes, thats about 40 % of the total population. These castes make up more than half the population of Pakistani Punjab, a bulk of the population in Khyber. In Sindh and especially Karachi they make up a much smaller part of the population.

      • Aakar Patel
        Posted at 06:32h, 05 January Reply

        We haven’t done caste census in India since the 1930s, but the Mandal report says over half of India is OBC. This is the peasantry. No reason to believe this ratio is different in Pakistan, including Sindh.
        Some of the identity will be lost over generations because family myths will claim other origins. However the culture will remain.
        NWFP and Balochistan are tribal and don’t have caste in the way that South Asians do, so far as I understand this.

Post A Comment