« Prop 54 | Gene Expression Front Page | The reverential agnostic »
October 07, 2003

Transition to liberalism

The father who killed his daughter got a life sentence (see related post). One thing I find curious though-where are the mentions of honor kills in Gulf states? After all, Jordan (even Pakistan) are relatively liberal Muslim countries, yet the story mentions both as hot-beds of honor killings (Jordan especially seems to crop up many a time). What about Saudi Arabia? Kuwait? etc. Perhaps the stories aren't coming out of those nations because of their closed societies, but I wonder if the situation might be modelled as an energetic barrier:


A) conservative society + no freedoms for women =>
B) conservative society + some freedoms for women =>
C) non-conservative + freedoms for women

In society A, there is no question of "honor" because women are never in the position to dishonor, in society C "honor" loses its value (relatively), while society B sits atop the energy hill, tottering back & forth between the two extremes, with the volatile mixture of conservative values but enough freedoms to transgress social expectations[1]. I believe today we are seeing many societies moving over that energetic barrier and the tensions & stresses to the social frabic are byproducts of the transition....

fn1. From what little I know, much of it revolves around the appearance of impropriety-a situation more plausible in a society where women have legal freedoms to act as independent agents (to some extent) and possibly enage in "dishonor." Note the the attacks upon great Roman women by historians, from Livia to Agrippina the Younger, a direct consequence of the conflict between a conservative society, but one liberal enough to allow women a place in public life.

Posted by razib at 10:54 PM




"Honour" (izzat, pukthunwalli) is a tribal cultural trait because of inter-tribal competition, "face" culture and marriage seen as necessary to cement (or break) alliances. When there are competing clans then honour and love tend to clash (Romeo and Juliet).

Islam's ethos was compatible with tribal societies and was assimilated rapidly by them because of the overlap. Hence Islam's distribution from Morroco to Pakistan and Uighurstan but the sharp "breaks" with Spain, India (Rajasthan, E. Punjab and Gujarat), Sinic China and northern agricultural Russia (all of whom were settled and continuous civilisations).

The unity of the Islamic lands is not so much because of Islam because their fundamental tribal matrix. Indeed it would not be a leap to claim that Muslims (at least Sunnis) are historically a coalition of tribes as evidenced that a mere century ago Iran's tribal population hovered around 25% and Pakistan was overwhelming populated by pastoralists*.

This explains the preponderance of tribal traits such as honour killings, cousin marriage and extended family kinships amongst Muslims. I would believe this trend is indicative of correlation not causation.

*Riparian West Punjab has no rain and was only suitable, prior to British irrigation works, for grazing. Incidentally there is a "barani" (rain fed) region around Lahore, which is where Partition was most contested and murderous because it was where the boundaries blurred.

Posted by: Zachary Latif at October 8, 2003 02:02 AM


My feeling about Saudi Arabia is that honor killing there is routine and nobody's business but the family and so nobody ever hears about it. When I was a writing teacher I had a Saudi student who wrote English as well as any ESL student I ever had (and better than almost all the American students, but then I taught mistly remedial). He was politically aware (just a modernizer and moderate Muslim, really) with a public health degree and was glum about returning. I suggested that he do investigative journalism and he said that that would be totally impossible. He didn't seem to have any doubts at all about that.

In countries with a secret police and no free media, people just disappear and even the family pretends it didn't happen. (China also). This is true of political prisoners but would be just as true of defiant or unchaste daughters. I can sort of imagine the euphemisms -- "Their daughter Fatima is having serious problems. It's really terribly sad".

Posted by: Zizka at October 8, 2003 07:47 AM


zack, you might note that hindu jatis have many of the traits of "tribes." and the adivasis of central india were not islamicized despite 1,000 years of proximity to muslim states of the deccan...i find your thesis interesting, but i'm not convinced.

Posted by: razib at October 8, 2003 12:08 PM


I nominate this as the longest comment:)

Islam's spread in Africa circumvented Amharic-speaking Ethiopia (granted Ethiopia was a Christian polity) but proliferated through tribal regions, which had exposure to civilisation.

I add the further caveat that Islam, defined by an urban-tribal dichotomy, was absorbed by tribes hovering near the pale of civilisation and exposed to the settled culture hence the gradient of Islamic expansion in Africa.

Islam's transmission would have been the conquest of imperial capitals and "elite dominance". The peripheral and subjugated tribal entities would have been converted whereas the surrounded rural population would retain their original way of life.

The agrarian population, in an Islamic entity, would have eventually converted to an adulterated Islam because of the "top to bottom" hierarchical patronage systems that were characteristic of these ancient despotic centres of civilisation (Egypt, Sindh, Iran, Levant, Iraq).

The Islamicised tribes would then have continued their conquests in neighbouring lands and assimilating as the elite, convert the peripheral tribes whilst ignoring & tax the rural population etc. This analogy is particularly apt in Egypt & North Africa, Indian Sub-continent (I believe Bangladesh was outside the scope of Brahmanical civilisation with the original inhabitants being tribal indigenes), Middle East and Iran & Central Asia.

Islam caught on like wildfire with the tribes and its present day contiguous dispersion, from Morocco to Pakistan, is characteristic of a tribal swathe with a huge geographic spread but a fairly limited population concentration.

However I should be more precise in my terminology with "tribes" and "tribal". Tribals entities are itinerant clans that exhibits aggressive, highly migratory patterns and extensive interaction with civilisation.

Under this definition the adivasis are akin to aboriginals, Veddas of Sri Lanka, as exhibited by their historic enclosure, passivity (Munda kings did not annex Bihar) and limited interaction with sedentary civilisation (it was to protect the sanctity of their environment that the British enacted laws preventing the flood of migrants into the adivasis regions) and are termed as tribes for administrative convenience. If you're referring to the Santalis (or Munda???) in the Chotanagpur Plateau then their exposure to Indic civilisation has always been limited and the Muslims of Jharkhand are immigrants from Bihar.

If there had been pre-existing cultural contact then the adivasis would have had a Hindu sub-strate rather than retaining animist beliefs and for instance their Munda kin in the alluvial Gangetic plans were absorbed into the Brahmanical caste system (Hindi word for plough is Munda in origin). Their isolation is analogous to the Empire of Trebizond, which was sustained by a network of alliances and the hazardous nature of its terrain.

Furthermore the historic aversion of the Mughals and the Shi'ite Deccan states in encouraging conversions would have had an inhibitory effect. For instanceOrissa was relatively unscathed because of its isolated geography though it was a bastion of civilisation. Consequently isolated aboriginals eking out an existence in the highlands would not have been magnets for conquest, conversion and Islamicisation (the highlands are rich for minerals but could not bring in the taxation possible in dense Hindu belts). Muslim Bengalis anchor north-east India however ultimately the aboriginals of the highlands did not Islamicise. Indeed in neighbouring Burma, which shares cultural affinity and topographic landscape, the Muslim population consists of coastal Malays and immigrant Indians.

The Jats present a further corollary in their patterns of Islamicisation. For instance the Hindu and Sikh population of Western Punjab had always been confined to a predominantly urban and mercantile minority. The non-Islamic “cultivator castes” settled en masse by the British in the canal colonies of the West Punjab contributed to a large non-Muslim population at Partition (post Partition some cultivator castes merely resettled their villages in East Punjab). Prior to the irrigation works there was the sharpest break between East and West Punjab (the ancient language “Lahnda” predominating in the latter), which explains the religious distribution. The Indo-Gangetic plain was not cultivable from Derajat to Lahore and settled agriculture started from East Punjab. For instance the Gujars (the tribe that gave it’s name to Gujarat, Gujrat and Gujranwallah) upon converting to Hindusim would assimilate in the plains whereas Muslim Gujars preserve their tribal traditions and livelihood. The Hindu and Sikh Jats practise exogamy, retain the caste system (persistence of the village helot\slave in Central and East Punjab) and were settled agriculturists. For instance during partition the differences were highlighted subtly amidst the bloodshed whereas violated Muslim girls were not ostracised as their Sikh or Hindu counterparts who often committed suicide to avoid ritual impurification.

Those Jats that had settled in the Eastern Punjab had thoroughly assimilated in settled Hindu culture (especially considering their proximity to the pale of Hindu civilisation which was but a region away), and to an extent lost their tribal ways, thus their relatively apathy towards Islam. However one could argue that Sikhism evolved in Central Punjab as reflecting the tensions between the tribal antecedents of the Jats and their settled Hindu culture. Sikhism may be an outgrowth of Hinduism (strands of Bhakti and Yogi Hinduism) but evolved to incorporate many tribal characteristics such as an egalitarian impulse (Jats never had much use for Brahmins) and their martial spirit.

Posted by: Zachary Latif at October 9, 2003 03:17 AM


I am thankful for the excellent posts by Zachary.
I have certain fundamental philosophical questions, unrelated to this particular topic but which have been stimulated by the posts here.

1) Is it true that in the tribal stage society is largely egalitarian? I would point out the alpha-male primate-like strucuture among African tribes today, or the general proclivity of tribes which subsequently became civilised to form elaborate aristocracies as a counter example.

2) What are the conditions that cause certain tribes to progress through stages of stratification, division of labour and formation of administrative and economic aristocracies, whereas other tribes continue in their original state. What is the role of agriculture, technology, hunting-gathering and military environment in stimulating or depressing this progression.

3) If Islam is indeed a theological doctrine that enables coagulation and solidification of tribal patterns of life, how will the essence survive industrialisation? Can modern industrial civilisation sustain a hierarchical structure or is it neccessarily equalising? I tend to believe that the modern western liberal egalitarian condition is in no way an inevitable consequence of industrialisation but the result of judaeo-christian morality guiding the evolution of society by the exploitation of industrial changes.

-Timurlane

Posted by: Timurlane at October 9, 2003 02:12 PM