Cultural Geography Archives

Geographical Patterns and the Ethnicity of Russia’s War Deaths, 2022-2026

Article-Grid, Ethnicity, Geopolitics, Russia, Ukraine, and Caucasus / Martin W. Lewis

A widely circulated map of Russian war deaths from February 2022 to August 2026, based on information from the Russia 200 project, reveals some stark differences among Russia’s federal subjects (divisions of the country), as well as some interesting regional patterns. Moscow sits at the low end of the spectrum, with 28 deaths per 100,000 people, while Tuva in southern Siberia occupies the top spot, with 650. The lowest figures cluster in two highly dissimilar regions: the western half of European Russia, the country’s core, and the north Caucasus, a mostly Muslim and distinctly peripheral region. The highest mortality figures cluster in south-central Siberia, far eastern Siberia, and northern European Russia. The first two of these regions are far indeed from Russia’s political, economic, and demographic core.

The low level of deaths in Moscow and nearby regions is often said to reflect a desire to minimize the impact of the war on Russia’s core, thus reducing the threat of popular opposition. In this view, distant regions are a lesser concern and can thus be forced to endure much higher casualty rates. The North Caucasus, however, defies this pattern, as is distinctly peripheral but has a death count similar to that of Moscow and its environs. As the north Caucasus has a restive population and history of rebellion, it would be dangerous to antagonize it. Chechnya, which sought independence in a bloody struggle after the collapse of the Soviet Union, has the second lowest death date in the country (35 per 100,000). It is perhaps not coincidently that the only part of the North Caucasus with an average death rate is North Ossetia-Alania, a mostly Christian republic situated in a heavily Sunni Muslim region.

Given the concentration of high mortality in other peripheral regions of Russia, the question arises as to whether casualties are also concentrated among non-ethnic-Russian populations. A pairing of the mortality map with one showing “Federal Subjects by Percentage of Ethnic Russians” suggests as much. As can be seen, areas dominated by ethnic Russians tend to have low to average death rates, while those with high percentages of non-Russian generally have higher death rates (except the North Caucasus).

Russian War Deaths and Russian Ethnicity maps

This correlation, however, is not as strong as it might seem at first glance. In the map posted below, I have outlined Russian federal subjects with populations more than 50% non-Russian. As is noted above, the most non-Russian region, the North Caucasus, has a very low death rate. The mostly Mongolic speaking and traditionally Tibetan Buddhist Republic of Kalmykia has a slightly higher than average rate, as does the linguistically and religiously, diverse, but Muslim leaning, central Volga region. The highest rate is found in Turkic-speaking, Tibetan Buddhist Tuva. The death rate is also elevated in massive Sakha, a Turkic-speaking republic with a Christian plurality. But of the four federal subjects with death rates above 499 per 100,000, three have Russian majorities. But two of these, Buryatia and Chukotka, have slight ethnic Russian majorities (64% and 54% respectively). The same pattern is found in several other regions with high mortality rates, as can be seen in the second map below (on which federal subjects with more than a third non-ethnic-Russians are outlined).

Russian War Deaths and Areas Less than 50% Ethnic Russian map

Russian War Deaths and Areas Less than 66.6% Ethnic Russian map

Although regions with relatively high percentages of non-ethnic-Russians clearly have high mortality rates, it does not necessarily follow that these deaths were concentrated among their non-ethnic-Russian population. Acquiring such information is difficult and often impossible, but Alexey Bessudnov of the University of Exeter was able to use name-based classification to determine some ethnic mortality rates relatively early in the war. He found that for the country as a whole, Buryats and Tuvans had mortality “risk rates” roughly four times greater than those of ethnic Russians. But he also found that within Buryatia and Tuva, “fatality rates for the titular ethnic groups and ethnic Russians are comparable, suggesting that military recruitment is primarily influenced by regional socioeconomic factors rather than ethnicity.”

An AI (Grok) assessment of the situation essentially concurred with Bessudnov, arguing that ethnicity per se does not play a major role in the geographical patterns of war deaths. As it noted, “poor ethnic-Russian regions (parts of the Urals, Kirov, and so on) also have high death rates, which is why [the Russia 200 project] attributes the [war-death] map mainly to poverty, settlement size, and recruitment pressure rather than a policy of targeting minorities.”

The problem with this AI analysis is the fact that some strikingly poor regions in western European Russia have low death rates, as will be shown in the next post. “Poverty,” in other words, does not seem to be the main factor. AI, it would seem, still has limited geographical reasoning capacity.

Geographical Patterns and the Ethnicity of Russia’s War Deaths, 2022-2026 Read More »

Houthi Threats to Create a Greater Yemen and a Lesser Saudi Arabia?

Article-Grid, Border Disputes, Cultural Geography, Geopolitics, Sports / Martin W. Lewis

In late September 2026, the Yemeni journalist Hussain Albukhaiti, a Houthi-aligned commentator based in Sanaa, publicly asserted Yemeni claims to the Saudi provinces of Najran, Jizan, and Asir, claiming they will be conquered and annexed if the current war escalates. His unrealistic statement fits a long pattern of Houthi-linked rhetoric treating the three provinces, which were ceded to Saudi Arabia under the 1934 Treaty of Taif, as occupied Yemeni land. According to Wikipedia, the treaty gave Saudi Arabia only “temporary sovereignty” over these lands, but the Saudi government regards its control as permanent.

Currently, the Houthis control most of western Yemen, the most densely populated part of the country. They claim, however, to be the legitimate authority for the entire country. As a result, the Houthis reject southern secessionism and present themselves as defenders of Yemeni unity and sovereignty against external (especially Saudi and U.S.) influence, while running an authoritarian administration that suppresses rivals inside their zone. The map posted below shows the current geopolitical reality along with Houthi dreams of expansion.

It is not coincidental that much of this territory is primarily inhabited by Shia Muslims. Only southern Asir and some interior portions of Jizan, however, are inhabited by followers of Zaidi Shia Islam, the sect of the Houthis. Najran is mostly inhabited by Sulaymani Ismailis, who follow a distinctly different branch of Shiism.

These two Shia sects differ significantly in their current interpretation of jihad. Zaidi ideology emphases resistance to tyranny but adds the modern political program of anti-Americanism, anti-Zionism, and alignment with Iran, as expressed in the slogan “God is great, death to America, death to Israel, a curse on the Jews, victory to Islam.” Hussein Badr al-Din al-Houthi, founder of the Houthi movement, regarded armed jihad against the United States, Israel, and their allies as a religious priority. Ismaili traditions, in contrast, emphasize “greater Jihad,” or the internal struggle against the base desires of the self. Ismailis are associated with esoteric beliefs, and the largest Ismaili sect, that is the Nizaris, is noted for its liberalism, rationalism, emphasis on education, and cosmopolitanism.

It is difficult to say, however, what the Sulaymani Ismailis of Najran currently believe. Like other Ismailis, they have historically practiced taqiyya (or precautionary concealment of belief), understood as a practical response to persecution and a way to protect esoteric knowledge. Given the hostility of the Saudi government to Shia beliefs and practices, such concealment has become very important.

Houthi Threats to Create a Greater Yemen and a Lesser Saudi Arabia? Read More »

Is This the Worst Map in Current Circulation? No, It Is This One!

Article-Grid, Cartography, Cultural Geography / Martin W. Lewis

As I often mock the AI-slop maps that I see on X (“Twitter”), my account has apparently responded by sending even more garbage my way. One example recently caught my attention as perhaps the worst map that I had ever seen. Then, several days later, I encountered something even worse.

The first map, from the inaptly named blue-check account called “Logic Universe,” misplaces most of the countries that it highlights for having received massive loans. It labels the western United States as “Mexico,” northeastern France and environs as “Greece,” northwestern Russia, Finland, and the Gulf of Bothnia as “Ukraine,” and Brazil as “Argentina.” But that is the least of its problems. The base map that it uses is bizarre. It hallucinates a second Iberia and western Africa, which it has attached to South America. Its depictions of the Caspian Sea and the Gulf of Mexico are ludicrous. It does not, in short, even come close to being a legitimate map.

The second map comes from an account called “The Index Insiders,” which has some 20,000 followers and supposedly focuses on “Global Ranking, Facts, Wealth, Culture, Maps, Economy, Statistics & Data.” AI (Grok) tells me that its “signup metadata points to India.” Its map in question, which purports to show “U.S. Wars and Conflicts in the Past 30 Years,” at least depicts country shapes more or less correctly. But its labels are a hilarious mess. The United States is labeled “Canada,” France is labeled “Kosova,” Spain is labeled “Irax,” Turkey is labeled “Afghanistan,” Saudi Arabia is labeled “Yemen,” Egypt is also labeled “Yemen,” Libya is labeled “Syria,” Algeria is labeled “Libya,” and Peru is labeled “Somalia.” The color scheme, along with its minimalist key, tell us that the U.S. has been engaged in wars or other conflicts with most of its NATO allies over the past three decades, as well as with Switzerland, Austria, Morocco, Tunisia, and Saudi Arabia. Most egregiously, the map detaches Africa from Asia and joins it instead to western South America.

As far as county sizes and shapes are concerned, the worst offender that I have found was posted by a different account, one called “World Insights.” This pseudo-map of “famous landmarks” in Europe depicts Lichtenstein as roughly the same size as Germany. Enough said.

In a 2023 article called “Viral Cartographies and Fake Maps on Social Media Platforms,” geographer Jörn Seemann insightfully explores some of the pitfalls generated by fraudulent maps that are widely circulated online. As he aptly writes, “virtual and viral cartographies make it more difficult to discern fact from fiction or joke from ideology.” But the problematic maps that Seemann highlights are models of tongue-in-cheek clarity compared to the AI-slop maps that are circulating today. I don’t know whether to laugh or cry.

Examples of Fake Geopolitical Maps on Twitter

Is This the Worst Map in Current Circulation? No, It Is This One! Read More »

Deeper Problems with Maps of “Excessive Drinking” in the United States

Article-Grid, Cultural Geography, North America, psychoactive substances / Martin W. Lewis

Oddly, maps of “excessive drinking” in the United States do not correspond particularly well with maps of per capita alcohol consumption. New Hampshire, Delaware, and Nevada, for example, are typically depicted as having the country’s highest levels of per capita alcohol consumption but only moderate levels of excessive drinking. In contrast, Iowa is usually depicted as having one of the highest levels of excessive drinking but only a moderate level of total alcohol consumption. (Unfortunately, maps of per capita alcohol consumption by county are not available.)

Maps of Excessive Drinking and Total Alcohol Consumption 1

Maps of Excessive Drinking and Total Alcohol Consumption 2

Such a seeming discrepancy can easily be explained by differences in drinking behavior. It is possible, for example, that on average Nevadans drink more than Iowans but generally do so in a more responsible manner, engaging less in “binge drinking.” Possible yes, but probable? I doubt it. I think it is more likely that the data used to make these maps are flawed. In the remainer of this post, I will critically examine the sources used for maps of “excessive drinking.” Admittedly, much of the evidence that I present was derived from “conversations” with AI; as such, my arguments should also be regarded with some skepticism.

The foundation of excessive-drinking maps is the data tabulated by the CDC’s Behavioral Risk Factor Surveillance System (BRFSS), derived from state-based telephone surveys. Sample sizes and exact questions, however, evidently vary by state and year, complicating comparisons over time and space. More significant, honest answers about such problematic behavior cannot always be expected. Supposedly, surveys routinely recover only a fraction of the amount of alcohol that sales data imply, often on the order of 30–60%. Such underreporting supposedly stems from forgotten drinks, underestimated pour sizes, and nonresponses by heavier drinkers. Many respondents, deemed “high impression managers,” want to appear respectable by reporting that they drink less than they actually do. According to one academic study, “high impression managers report 20 to 33% less consumption and are about 50% less likely to report risky drinking.” Such considerations worrisomely suggest that actual rates of excessive drinking are considerably higher than they appear to be on these maps.

But it is also known that the bias can run in the opposite direction, particularly among certain groups of college students for whom heavy drinking can be seen as a badge of belonging. Is it therefore coincidental that counties that are often mapped as having the highest rates of excessive drinking also have large college-student populations, such as Missoula and Gallatin in Montana and Dane in Wisconsin? More to the point, more under-reporting is to be expected in areas in which alcohol is deeply stigmatized (Utah and eastern Idaho; interior regions of the U.S. South) than in areas with alcohol-tolerant cultures (the upper Midwest and much of the interior West).

The definition of “excessive drinking” used in these surveys are also problematic. The CDC definition seems straightforward: either “binge drinking” (four or more drinks on one occasion for women, and five or more for men), or “heavy drinking” (eight or more drinks per week for women, and fifteen or more for men). Although the first threshold seems reasonable, the second is questionable. I doubt that most people would regard a woman who drinks a glass of wine with dinner every day, and two on Saturdays, as an “excessive drinker.” By the same token, is a man who drinks two beers a day, and three on Fridays, really a “heavy drinker?” Perceptions obviously differ widely from person to person. More important, actual alcohol consumption depends on what kind of beer the drinker in question consumes. A “light” beer, for example, typically has less than half the alcohol content of imperial IPA of the same size.

A similar problem is encountered with the concept of “binge drinking.” “Five or more drinks on one occasion” might seem like a high figure, but it depends on how long an “occasion” lasts. The standard figure for research purposes is two hours. This duration is used because a typical man who consumes five standard drinks (or 0.6 fluid ounces of ethanol) in two hours will reach a blood-alcohol concentration over 0.08%, which counts as legally “drunk” in most U.S. states. But the surveys on which excessive-drinking data are based do not seem to be rigorous in this regard. According to AI, “respondents are not walked through a two-hour clock. So five beers over a six-hour dinner can be coded the same as five shots in an hour. That is a real measurement gap, not a quibble.”

For both “heavy drinking” and “binge drinking,” individual weight and muscle-mass should also be considered. All other things being equal, a heavier person must drink more alcohol than a lighter person to reach the same blood-alcohol concentration, which is the real measurement of intoxication. A muscular person, moreover, will have a lower blood-alcohol concentration than a less-muscular person of the same weight who drinks the same amount of alcohol over the same period of time.

Ideally, alcohol-consumption surveys would account for such factors, but that is not necessarily the case. Considerations of weight and muscle mass are ignored, and a twelve-ounce beer is usually treated as standard unit regardless of alcohol content. After much back-and-forth, my AI “informant” concluded by telling me that the definition of excessive drinking employed by the CDC’s Behavioral Risk Factor Surveillance System (BRFSS) “is operational and crude on purpose. It is useful for comparing places and tracking a risk category. It is a poor description of how drunk someone got on a given night or how much alcohol they actually swallowed.”

Finally, alcohol-consumption surveys face the same problems of sample representativeness that bedevil all pollsters. Many if not most persons refuse to respond to such surveys, and it cannot simply be assumed that non-responders have the same experiences as responders. Researchers are well aware of this problem and try to compensate, but it is not clear whether they do so adequately when it comes to “excessive drinking.” According to AI:

BRFSS tries to account for refusal and uneven participation—but only in a limited way. It does not fully correct for the possibility that drinkers and non-drinkers refuse at different rates. Most people who are called never complete an interview. Recent median state response rates are around 44% (roughly 31–65% by state). That is normal for large telephone surveys now, not a secret. It assumes that, within those groups, people who answer the phone drink like people who hang up. Such an assumption is questionable.

All of this is not to argue that maps of excessive drinking in the United States are worthless. I find them interesting and regard them as partially valid. But they are more suggestive than definitive. I would also prefer that they be give more modest titles, such as “Rates of Excessive Drinking as Indicated by Public-Health Surveys.” A brief note on how “excessive drinking” is defined would ideally be included on all such maps.

Deeper Problems with Maps of “Excessive Drinking” in the United States Read More »

Could U.S. Counties with Low Levels “Excessive Drinking” Really Have High Levels of Alcohol-Induced Mortality and Low Levels of Health?

Article-Grid, Cultural Geography, Montana Atlas, North America, psychoactive substances / Martin W. Lewis

The map of excessive drinking in the United States that was recently discussed in GeoCurrents was originally posted on IntoxiStates.com and is based on information from the CDC’s Behavioral Risk Factor Surveillance System, as packaged by the University of Wisconsin’s Population Health Institute County Health Rankings & Roadmaps (CHR&R). It clearly indicates that Wisconsin has the country’s highest rate of excessive drinking and overtly states that among the many counties of the U.S., Gallatin in Montana occupies the top place. The IntoxiStates map is not dated, but a 2025 map based on 2022 data from the same source paints a different picture. Published in the The Hill, this map puts Montana in the top position and depicts Iowa as also having a higher rate of excessive drinking than Wisconsin. The county-based map from the same article indicates that Missoula in Montana is the ‘drunkest” county, with a 29% excessive-drinking rate. The most recent data from CHR&R’s own website puts both Missoula and Madison counties in Montana ahead of Gallatin, but just barely (26% vs. 25%). The “excessive drinking” data used to make these maps, in other words, change from year to year. More to the point, excessive drinking in Wisconsin does not seem to be nearly as pervasive or unusual as it appears to be on the IntoxiStates map.

The Hill Excessive Drinking Counties map

Montana Excessive Drinking by County map

The CHR&R’s Montana map is particularly interesting, as it reveals a clear county-level correlation between relatively low rates of excessive drinking and high percentages of indigenous residents. This association is significant, as it contradicts a widely held stereotype. It might not, however, be valid. It would be expected that low levels of excessive drinking would be associated with low levels of alcohol-induced mortality, but that is not what the Montana data indicate. A map of alcohol-induced mortality in Montana in the Covid-year 2021, derived from a map published in a PLOS Global Public Health article, shows a surprising negative relationship between excessive drinking and death by drinking. Another pair of maps indicates that Montana counties with high rates of excessive drinking are significantly healthier than counties with low rates. (Correlation, of course, does not imply causation.)

Montana Excessive Drinking Indigenous Population maps

Montana Excessive Drinking and Alcohol-Induced Deaths map

Montana Excessive Drinking and Health map

Montana does not seem to be alone in this regard. A pairing of the IntoxiStates map of excessive drinking with the PLOS Global Public Health map of alcohol-induced mortality in 2024 shows that several counties in other states have the same relationship. Curry County Oregon, Apache and La Paz counties in Arizona, and Torrence County New Mexico are all mapped as having low rates of excessive drinking and high rates of alcohol-induced death. Ethnic or racial differences do not seem to be a factor. Curry County is strongly white (83%), Apache County is strongly indigenous (75%), La Paz is highly mixed (61% white alone, 23% Hispanic/Latino, 14.5% American Indian) and Torrence is fairly evenly divided between whites and Hispanics (52% white alone, 42% Hispanic). As the final map in this post shows, many of the “driest” counties located in the interior U.S. south are also among the least healthy.

USA Excessive Drinking and Alcohol-Related Mortality map

Such surprising findings indicate that the “excessive drinking” data need to be examined more closely, as will be done in the next GeoCurrents post.

Could U.S. Counties with Low Levels “Excessive Drinking” Really Have High Levels of Alcohol-Induced Mortality and Low Levels of Health? Read More »

Does Gallatin County Montana Really Have the Nation’s Highest Percentage of Excessive Drinkers? If So, Does It Matter?

Article-Grid, Cultural Geography, North America / Martin W. Lewis

A map of U.S. counties that supposedly shows the percentage of residents who drink alcohol excessively has recently received a lot of attention. What immediately struck me when I first saw it was the prominence of Wisconsin, mapped as having a far higher proportion of problem drinkers than any other state. As I spent a year in Wisconsin and witnessed its drinking culture, I was initially inclined to regard this portrayal as reasonably accurate. But the more I examined the map, the more my doubts grew. The discrepancy between Wisconsin and its neighboring states is simply too stark to seem realistic. Several other states also seem to be overly differentiated from their neighbors. The low level of excessive drinking in Utah, for example, is understandable due to the prevalence of the LDS (Mormon) church, but several highly Mormon counties in southeastern Idaho are not mapped in the same way. Due to such seeming discrepancies, I plan to examine the data used to create the map and post my findings on GeoCurrents. For the time being, however, I will take this map as given and ask what it might reveal about the significance of regional drinking patterns.

It might be assumed that extremely high levels of excessive drinking would have some correlation with social disfunction. At the individual level, after all, immoderate consumption is associated with a host of health and behavioral problems. According to this map, more than a quarter of the (adult?) population drinks excessively in quite a few U.S. counties. With such high rates of drunkenness, wouldn’t these same counties be expected to exhibit some aspects of social disorder? By the same token, shouldn’t the soberest counties do relatively well according to social indicators?

What the map shows, however, does not always match this image. Many of the counties supposedly burdened by excessive drinking are prosperous, healthy, well educated, and safe. Prominent examples include the affluent Rocky Mountain counties of Colorado, Teton County, Wyoming, and Dane County, Wisconsin. By the same token, many counties in the South that are depicted as relatively free of excessive drinking are characterized by widespread poverty, low levels of health and education, and high crime rates.

Let us consider two counties on the opposite ends of the spectrum: Gallatin, in Montana, and McDowell, in West Virginia. Gallatin is mapped as having the highest rate of excessive drinking in the United States, at 26.8 percent. McDowell, in contrast, falls in the second lowest excess-drinking category. Although I have no experience with McDowell County, I know Gallatin County quite well. Its main city, Bozeman, is noted for its bars, brew pubs, and lively drinking culture. Yet it is also a safe and easy-going town embedded in a high-trust society. Gallatin County is the kind of place where people leave their trucks unlocked and running in parking lots on frigid winter mornings, unconcerned that they might be stolen. From what the data indicate, McDowell County in West Virginia is very different kind of place.

These differences can be easily elucidated through a series of maps. To begin with, Gallatin has a far higher educational level than McDowell. As can be seen, it falls into the highest tier of residents with college degrees, while McDowell falls into the lowest. Bozeman is home to Montana State University, noted for its engineering programs. It has a small but thriving tech sector, and many graduates stay in town, not wanting to leave such a pleasant and amenity-rich environment. As the second map shows, Gallatin County also has a high percentage of residents with graduate and professional degrees. Bozeman is, in other words, a “knowledge town.”

As the series of maps at the bottom of this post shows, Gallatin is far ahead of McDowell in other commonly used metrics. It is much wealthier, has a much lower homicide rate, far higher life expectancy, a strikingly lower rate of obesity, a much lower instance of tobacco-smoking, a far lower drug-overdose rate, and a significantly lower divorce rate. And as paired tables at the end of the series of illustrations show, Gallatin County is booming, attracting people from other areas at a rapid clip, whereas McDowell is losing residents at an alarming rate. In short, the “drunkest” county in the United States is remarkably successful by almost any standard, while one of the country’s “soberest” counties is withering away, plagued by numerous economic, social, and behavioral problems.

We must be wary, however, of reading too much into these comparisons, and I must again caution that the “excessive drinking” data used in this analysis may not be valid. Correlation, moreover, does not imply causation, and many other factors must also be considered. McDowell County, for example, has suffered grievously from the closure of its coal mines, which has no doubt intensified its many problems – although few of its remaining residents have evidently been driven to drink.

Yet for all of this, it is still safe to say that excessive drinking is a major cause for concern, at both the individual and societal levels. The difficulty comes in defining what “excessive” means. For a strident teetotaler, any alcohol consumption may seem excessive; for an unrepentant lush, excessive consumption is difficult to achieve. This is a complicated issue, as worrisome or dangerous levels of alcohol consumption vary from individual to individual and according to social context. At any rate, it remains to be seen how the author of this map defined the term. I suspect, however, that the definition used has a lower threshold than one that I would favor.

Despite these qualifications, I have no doubt that Gallatin County has a higher per capita level of alcohol consumption than the average American county. But is that necessarily a bad thing? The issue is debatable. Some scholars claim that moderate alcohol consumption has profound social benefits. Edward Slingerland goes so far as to argue that we would not have developed civilization without intoxication. I am not willing to go that far, but I can say that a geographical study of alcohol consumption in the United States offers no support for the position that drinking is intrinsically harmful.

Edward Slingerland’s Book Entitled “Drunk”

As a final note, Gallatin County also seems to have a significantly higher-than-average rate of cannabis consumption.* Dispensaries are ubiquitous; the Montana Department of Revenue lists 72 in and around Bozeman, a city with only around 60,000 residents. As is the case with alcohol, I do not see any evidence that the prevalence of cannabis has harmed Gallatin County, although I am sure than many residents would disagree vociferously.

*Reliable data on cannabis consumption is impossible to find. But an AI (Grok) assessment is worth considering: “Legal sales (good proxy for market intensity, not user prevalence) data from the Montana Department of Revenue consistently put Gallatin near the top of the state in monthly legal cannabis sales—often second only to much larger Yellowstone County. Combined with Gallatin’s population (~129,000) and very high dispensary density around Bozeman, per-capita legal sales are high. That is consistent with high use by residents plus tourist/visitor purchases. Montana as a whole also ranks among the highest states for legal cannabis sales and excise-tax collections per capita.

McDowell and Gallatin Demographically Compared

Does Gallatin County Montana Really Have the Nation’s Highest Percentage of Excessive Drinkers? If So, Does It Matter? Read More »

Mapping Religious Diversity

Article-Grid, Cultural Geography, Religion / Martin W. Lewis

The Pew Research Center recently published an interesting “religious diversity index,” which I have re-mapped (below). (Pew’s own map, also posted below, is in my view inadequate.) As can seen on either map, the lowest levels of religious diversity are generally found in the heavily Muslim belt of northern Africa and southwestern Asia. Exceptions are encountered here in countries that have large numbers of foreign workers, such as the United Arab Emirates (UAE). Different patterns would have resulted if only citizens had been counted: some 74–76% of the residents of the UAE are Muslim, but virtually all UAE citizens (Emiratis) adhere to the faith. Low levels of religious diversity are also found in a few predominately Christian countries (Romania, Moldova, Papua New Guinea) and in one Buddhist country (Cambodia).

High levels of religious diversity are found in several parts of the world. Most prominent are wealthy Western countries with large numbers of immigrants and many residents with no religious adherence. The northern and eastern countries of sub-Saharan Africa where Christianity and Islam coincide also rate high in the religious diversity index. In eastern Asia, Japan, South Korea, Taiwan, and Vietnam have high diversity scores, due largely to their combinations of Buddhism, secularism and, especially in South Korea, Christianity. Malaysia ranks high due mostly to the religious diversity of its Chinese population. For the same reason, Singapore has the world’s highest ranking, and by a considerable margin. I had thought that Suriname would come out on top, with its Christian, Muslim, Hindu, and animist communities, but it falls into second place. Neighboring Guayana and French Guiana also have high religious diversity rankings (but I mapped French Guiana as part of France, as it is an integral department of that country).

Overall, the Pew Religious Diversity Index is meaningful and well-formulated, which is why I decided to remap it. But I do have some criticisms. The first is its failure to include diversity within the major religious traditions. Hinduism, for example, is so diverse that some scholars view it more as a family of related faiths than as a single religion. Buddhism, Christianity, and Islam also have considerable internal diversity of beliefs and practices. The gap in Turkey (Türkiye) between Sunni Islam and Alevism is especially large, and if it had been considered, the country would have been given a higher diversity score. By the same token, Tonga would have ranked higher if members of the Church of Jesus Christ of Latter-Day Saints (Mormons) had been separated from Trinitarian Christians.

Another problem, which is probably insurmountable, stems from the failure to consider religious syncretism and other combinations of faith traditions. Papua New Guinea, for example, is overwhelmingly Christian on the surface, and thus has a low diversity ranking. But most Papuans strongly retain elements of their traditional beliefs, which are themselves very diverse. Other countries have similar mixtures. A scholar who worked in the west African country of Guinea once told me that the first thing you need to understand about the country is the fact that it is 90 percent Muslim and 90 percent animist. A similar issue is encountered in Japan, where most people are relatively secular but still maintain some Buddhist and Shinto beliefs and practices.

Putting all faiths other than Christianity, Islam, Hinduism, Buddhism, and Judaism into a single “other religions” category is another problem, although it is again one that is difficult to avoid. Chinese “folk religion” (religions?) supposedly has several hundred million adherents. It is also notable that Sikhism is lumped together with “other religions,” even though it has roughly twice as many adherents as Judaism.

As a final note, the Pew Religious Diversity Index is derived from sophisticated statistical analysis, which is explained on the website. The following passages outline the methodology employed:

Two factors determine a country’s religious diversity score: the extent to which its population falls into a single religious group, and the degree to which the remaining population is dispersed across other religious categories. When a single religious group constitutes a higher proportion of the population, the country receives a lower RDI score.

Meanwhile, among countries where the largest group comprises the same share of the population, places in which the remaining population is distributed more evenly across groups score higher on the RDI. For example, in a country in which 70% of the population belongs to a single religious group, the index score would be 4.9 if the remaining 30% were concentrated in just one category. However, the score would rise to around 5.7 if the remainder were spread evenly across five or six different groups

Mapping Religious Diversity Read More »

In Victorian Britain, Geography Was an Essential Pillar of Education, as Demonstrated in Charles Dickens’ Bleak House

Article-Grid, Cultural Geography, Geographical Education, Historical Geography / Martin W. Lewis

As I periodically contemplate the abysmal state of geographical education and knowledge in the United States, I can’t help comparing it to the situation that existed in earlier times. Geography had long been an essential component of basic education, but it was largely eliminated in the U.S. in misbegotten post-WWII educational “reforms.” We now face a situation in which even college-educated Americans are hard-pressed to identify more than a few countries on a world political map.

In Victorian Britain, in stark contrast, a basic understanding of world geography was essential for every educated person. Young ladies as well as young gentlemen were expected to have a good deal of familiarity with the map of the world. They could even feel shame if they lacked it.

This is, I realize, a strong claim that demands justification. In this post, I offer only one piece of evidence. But it is, I think, a strong one. It comes from a passage in Charles Dickens’ novel Bleak House. Here we find the character Caddy Jellyby, a middle-class teenager who is forced to work as her mother’s secretary, comparing herself unfavorably to Ada Clare, a young lady she has just met. Caddy is speaking to another new acquaintance, the saintly Esther Summerson, the main protagonist and partial narrator of the novel.

“An orphan. Ain’t she? But she knows a quantity, I suppose? Can dance, and play music, and sing? She can talk French, I suppose, and do geography, and globes, and needlework, and everything? … I can’t, … I can’t do anything hardly, except write. I’m always writing for Ma.” (Bleak House, Chapter 4, “Telescopic Philanthropy.” Emphasis added)

Caddy Jellyby on the Importance of Geography

Bleak House is often regarded as Dickens’ greatest novel, and for good reason. In a 2022 essay in Atlantic, Charlie Tyson argues if Dickens had died in 1850, after David Copperfield but before Bleak House, he would be remembered mainly as a sentimental storyteller. In Bleak House, Tyson argues, Dickens crafted something entirely new: a multi-plotted “novel of networks” that captured the complexity of industrial and metropolitan life in London, and that combined the social-problem novel, the educational novel (bildungsroman), and the mystery novel. Tyson also claims that the book anticipated Kafka’s vision of a nightmarish bureaucracy and even much later postmodern experiments in fiction writing.

I hesitate to disagree with Tyson’s interpretation, as I am no expert in English literature. I would argue, however, that Bleak House is not primarily a novel of London and urban life, as it is far more comprehensive than that. It equally encompasses rural life, ranging from the opulent manor at Chesney Wold in Lincolnshire, to the park and gamekeeper’s lodge at the same estate, to a miserable brickmaker’s cottage in the countryside, roughly 20 miles north of London. Bleak House itself – which is anything but bleak – is near the same cottage, in a bucolic area outside of St. Albans in Hertfordshire.

In short, I agree that Bleak House is a magnificent portrayal of urban life, focused on the Holborn neighborhood of central London, particularly Chancery Lane and Lincoln’s Inn. (Not coincidentally, Dickens lived there as a young man.) But it is equally a story about rural life; without Chesney Wold and the rustic lands around St. Albans, it would be half the novel that it is, both literally and figuratively. It would also need a different title. Krook’s Rag and Bottle Shop, perhaps?

Main Location in Dickens’ Bleak House map

Main Urban Locations in Dickens’ Novel Bleak House map

Historical Map of the Main Urban Location in Dickens’ Bleak House

Source of the illustration-detail of Caddy Jellyby and Esther Summerson: https://victorianweb.org/art/illustration/phiz/bleakhouse/3.html

(Note: For those interested, I recommend both reading Bleak House and listening to it as an audiobook. It is available for free at LibriVox. I also recommend the version read by Mil Nicholson. In my humble opinion, Dickens is the greatest novelist, and Mil Nicholson is greatest reader of novels. It is a real art, and she does it superbly.)

In Victorian Britain, Geography Was an Essential Pillar of Education, as Demonstrated in Charles Dickens’ Bleak House Read More »

A Leftwing Stronghold in a Rightwing Region of Colombia: The Once-Vital Riverport of Honda and the Nearby “Red City” of Líbano (with a nod to Alexander von Humboldt)

Article-Grid, Cultural Geography, Elections, Latin America / Martin W. Lewis

As discussed in recent GeoCurrents posts, the northern and central parts of Colombia’s Andes region – the core area of the country – voted heavily for the narrowly victorious rightwing candidate, Abelardo do le Espriella, in the 2026 election. Several important left-voting exceptions were also noted: Bogotá with its industrial “suburbs,” and indigenous-majority districts. Another cluster of left-leaning municipalities located in northeastern Tolima Department (see the map below) calls for further investigation. As it turns out, some interesting historical and geographical factors are at play here.

Let us begin with the historically important municipality of Honda. Although situated well within Colombia’s high-elevation Andes region, Honda is actually lowland town (elevation 751 feet/229 meters), as it is located in the deep valley of the Magdelena River. Although it currently has only some 25,000 residents, Honda was once one of Colombia’s most important transportation hubs. Before the early twentieth century, Bogotá had no road or railroad connections to Colombia’s Caribbean ports. Instead, people and goods had to travel up the Magdelena River to its head of navigation at Honda. Honda was connected to Bogotá by a mule road that dates to the mid-1500s. According to Wikipedia, the town’s “golden age” was between 1850 and 1910, when “all the imported goods and articles arrived in Bogotá through the port of Honda.” It has been speculated – at least by AI – that Honda’s historical connections with left-leaning Caribbean ports, particularly Barranquilla, are linked to its current leftwing voting pattern.

The old road from the river port to the capital city, the Camino Real Honda–Santa Fe de Bogotá, traversed some rough landscapes while gaining 7,923 feet/2,415 meters of elevation from its beginning to its end. The local topography and elevation differences can be seen the map posted below, which was cobbled together from two excellent Wikipedia physical maps. The route of the road is visible in the next figure posted below, which is based on a detail taken from a map of the late 19^th century.

Historical Road from Honda to Bogota map

In 1801, the great German geographer Alexander von Humboldt took an arduous canoe trip to Honda and then proceeded overland to Bogotá, where he stayed for two months. In Bogotá, von Humboldt worked with the prominent botanist and mathematician José Celestino Mutis, a key figure in the Spanish-American Enlightenment. I tried to access Humbolt’s assessments of Honda from his personal diaries, using AI as a crutch. I learned that “These emphasize practical travel difficulties, the surrounding vegetation, commerce on the river, and scientific measurements more than architectural or social details of the town.” Humboldt did note, however, that the inhabitants of Honda were “disfigured by goiters,” presumably due to iodine deficiency.

Northern Tolima, Colombia and Left-Voting Municipalities Map

Most of the other leftwing-voting municipalities of northeastern Tolima Department are located on the Magdelena River south of Honda and hence have somewhat similar historical and geographical conditions (see the map above). The exception is Líbano, a plateau town of some 36,000 inhabitants. I was initially surprised by Líbano’s leftwing election returns, as it is a coffee-growing district that was established in the mid-1800s by settlers from Antioquia Department. Today, the plateau zone of Antioquia and Colombia’s coffee-growing areas more generally are noted for their political conservatism. This is a familiar story to me, as I studied in graduate school under James Parsons, author of Antioqueño Colonization in Western Colombia (1968, University of California Press). As stated at the beginning of the book:

The temperate uplands of the northernmost Andes of western Colombia are the home of the energetic and thrifty Antioqueños, the self-styled “Yankees of South America.” They are shrewd, aggressive individualists whose extraordinary colonizing genius and vigor have made them the dominant and most clearly defined population element of the republic. Their long and effective geographical isolation in the interior highlands of Colombia is reflected in a determined conservatism and a marked cultural particularism. Being Antioqueño means more to them than being Colombian.

James Parsons Antioqueno Colonization in Colombia map

My perplexity was, however, relieved by further investigation. Líbano, it turns out, has a fascinating and unusual political history, one that still heavily influences its political orientation. As AI (Grok) performed extremely well in this instance, I will simply paste in its interpretation rather than put it in my own words:

Líbano (Tolima) has a distinctive historical tradition of radical Liberalism, early socialist/communist organizing, and lasting progressive political networks that set it apart from many nearby highland coffee-growing municipalities, which more often lean conservative or right-leaning. This explains why it backed Iván Cepeda (Pacto Histórico) in the 2026 presidential runoff while much of the surrounding area supported Abelardo de la Espriella.

Líbano was founded in the mid-19th century by Liberal settlers from Antioquia/Viejo Caldas (led by figures such as Isidro Parra). It developed as a prosperous coffee municipality in the northern cordillera of Tolima and became a known Liberal stronghold (“the Red City”). Liberal governments encouraged its growth as a party redoubt.

In July 1929, roughly 300–armed artisans, peasants, and workers (linked to the Partido Socialista Revolucionario and inspired by the Russian Revolution) briefly seized the town and nearby areas in what historians describe as one of the earliest organized communist/socialist insurrections in Latin America. It was quickly suppressed, but the episode embedded a durable local identity of agrarian radicalism, labor organizing, and left-leaning dissent. [Former] President Petro has publicly referenced the “Líbano bolchevique.” This memory continued to shape local politics and even influenced later guerrilla fronts.

During mid-20th-century partisan violence, Líbano remained a predominantly Liberal area that experienced guerrilla self-defense activity and later leftist organizing. It produced or sheltered progressive figures and retained networks tied to agrarian struggles, coffee-worker issues, and opposition to conservative dominance.

The influential Jaramillo political family (notably Alfonso Jaramillo Salazar, a longtime Liberal leader, physician, governor, congressman, and health minister, plus relatives such as Guillermo Alfonso Jaramillo, a Petro-era minister, and Mauricio Jaramillo) built strong local clientelist and progressive ties. In recent cycles, “jaramillismo” and Liberal factions in the area aligned with or voted for Pacto Histórico candidates rather than traditional or right-wing options. Local coverage treated Líbano as a key northern Tolima bastion for Cepeda/Pacto Histórico.

A Leftwing Stronghold in a Rightwing Region of Colombia: The Once-Vital Riverport of Honda and the Nearby “Red City” of Líbano (with a nod to Alexander von Humboldt) Read More »

Afro-Peruvians and the 2026 Election in Peru

Article-Grid, Cultural Geography, Elections, Latin America / Martin W. Lewis

The three northwestern coastal departments of Peru, Tumbes, Piura, and Lambayeque all voted solidly for the victorious rightwing candidate, Keiko Fujimori, in the 2026 election, giving her, respectively, 64.4%, 57%, and 58.9% of their votes. By Peruvian standards, these three departments are roughly average in terms of social and economic development (see the maps in the previous post). All three have agriculturally oriented economies, and all have long favored conservative, market-oriented policies.

One distinctive socio-cultural feature of northwestern Peru is its relatively large proportion of Afro-Peruvians. Black Peruvians constitute 11.5 percent of the population of Tumbes, 8.9 percent of the population of Piura, and 8.4 percent of the population of Lambayeque, the highest figures in the country. One might expect Afro-Peruvians to favor leftwing candidates, which is certainly the case in Colombia regarding Afro-Colombians (as can be seen on the final two maps posted below). Unfortunately, racial electoral data in Peru is not available. It has been noted, however, that Afro-Peruvian issues received little attention in the 2026 platforms of all Peruvian political parties. If Black voters in Peru did favor the leftwing candidate, Roberto Sánchez, then the dominant mestizo population of the northwest must have voted even more strongly for Fujimori than the total departmental tallies would indicate.

Afro-Peruvians and the 20026 Election Map

Afro-Colombians the the 2026 Election map

Afro-Peruvians and the 2026 Election in Peru Read More »

Oxapampa: A Conservative-Voting Province of Partial Germany Ancestry in the Andean Highlands of Central Peru

Article-Grid, Elections, Ethnicity, Historical Geography, Latin America / Martin W. Lewis

Comparing an elevation map of Peru with a map of its 2026 presidential election shows that most highland area voted for the defeated leftwing candidate Roberto Sánchez. Tension between the people of the Andean plateaus and those of the economically dominant coastal region and its metropolitan center of Lima are deeply engrained, giving the highlands a distinctly leftwing orientation. But the same maps also reveal a few exceptions to this rule. The most interesting one is the province of Oxapampa in the eastern part of the department of Pasco.

Highlands of Peru in the 2026 Presidential Election

Oxapampa Province, Pasco Department, 2026 Election Map

Pasco is as a sparsely populated department, with fewer than 300,000 inhabitants, that is rich in mineral resources. Its capital city, the mining center of Cerro de Pasco, sits an elevation of 14,210 feet (4,330 meters), making it one of the loftiest cities in the world. Pasco’s eastern province, Oxapampa, is situated on the eastern side of the Andean divide, with elevations ranging from 17,400 feet (5,300 meters) at the highest peak (Huaguruncho) to 660 feet (200 meters) in the Amazonian lowlands of the east. Most of Oxapampa’s roughly 100,000 inhabitants live the mid-level elevation zone in the western part of the province, which is the area that supported the rightwing candidate Keiko Fujimoiri in the Peruvian presidential election of 2026.

Oxapampa’s conservative voting behavior may be linked to its settlement by German immigrants in the mid 1880s. The unusual history of the province is outlined in the Wikipedia article on the village of Pozuzo:

In the 1850s, Peruvian President Ramon Casilla proposed building a railroad from the capital city of Lima across the Andes Mountains, and onward to the navigable rivers of the Amazon Basin. Part of the plan (the only part realized) was to import German and Austrian Catholics to settle on the eastern slopes of the Andes and in the tropical rainforests of the Amazon region, then populated only by indigenous people. In 1855, the Peruvian leader signed an agreement with German traveler Kuno Damian von Schütz-Holzhausen to recruit 10,000 Germans and Austrians to emigrate to Peru and establish colonies on the eastern flanks of the Andes. The Peruvian government agreed to pay the cost of the passage from Europe to Peru of the emigrants and the cost of transporting them overland to Pozuzo and other areas designated for their settlement. …

On arrival [in Peru], Schütz found that the Peruvian government had done little to prepare for the arrival of the colonists and their transportation to Pozuzo. The colonists undertook a difficult journey on foot and mule to reach Pozuzo, starting from the port of Huacho, reaching an elevation of more than 4,700 metres (15,400 ft) crossing the Andes, to Cerro de Pasco, onward to Acobamba (Ambo) and, constructing the road en route, to Pozuzo. The town was established in 1859 by 172 of the original 302 colonists who had departed Europe together. Many had deserted en route or had died or been killed during the journey.

The population of the German settlers rapidly increased in the early decades, but their descendants eventually intermarried extensively with other Peruvians. Their culture gradually became more Peruvian than German. Some aspects of German culture, however, have persisted, helping make Oxapampa an important tourism destination in recent years. An AI assessment of the current ethnic composition of the province runs as follows:

In the town of Oxapampa, the German cultural presence is significant but not dominant. Many residents have partial German/Austrian ancestry, reflected in surnames (e.g., Schmidt, Müller), Tyrolean-style architecture, festivals, food (e.g., local beers), and some traditions. However, extensive intermarriage with mestizo, indigenous (e.g., Yanesha), and other groups means “full” or dominant German descent is limited. Nearby Pozuzo (founded by the original colonists) has a stronger enclave feel, with estimates of ~1,000 people of more direct Austro-German descent in a district of ~5,000–9,000.

German Influence in Oxapampa Province, Peru

Oxapampa: A Conservative-Voting Province of Partial Germany Ancestry in the Andean Highlands of Central Peru Read More »

Language & Ethnicity in the 2026 Peruvian Presidential Election

Article-Grid, Elections, Linguistic Geography / Martin W. Lewis

Not surprisingly, language and ethnicity were major factors in the 2026 Peruvian presidential election. As can be seen in the maps posted below, the defeated leftwing candidate Roberto Sánchez had his best results in areas where Spanish is not the majority mother tongue. In Peru’s southern highlands, Quechua remains the most common first language, especially in rural areas. Sánchez’s campaign highlighted the concerns of the historically marginalized indigenous peoples. In particular, he promised to transform Peru into a “plurinational” republic, recognizing the distinct national identities of non-Spanish-speaking communities and abandoning the idea of the country as a “nation state” defined around a singular political identity. The move from “nationalism” to “plurinationalism” has already been institutionalized in Bolivia and Ecuador, and has been proposed as well in Chile, Argentina, Guatemala, and Costa Rica. Although popular in indigenous communities and among the left more generally, plurinationalism is opposed by the right and is not popular in many of Peru’s Spanish-speaking communities.

Spanish Languages & Peru’s 2026 Election Map

Quechua Languages & Peru’s 2026 Election Map

Indigenous languages are also spoken across the lowlands of eastern Peru, but most have small numbers of speakers who are not well integrated into the national society. The Spanish language is dominant in most of this sparsely inhabited but rapidly growing region. As the map of lowland indigenous languages posted below shows, only two provinces in eastern Peru have clear indigenous majorities: Purus (in Ucayali Department) and Condorcanqui (in Amazonas Department). Interestingly, these two provinces had completely different voting patterns in the 2026 election. Condorcanqui overwhelmingly supported Sánchez, whereas Purus strongly supported Fujimori. The vote-count in Purus, however, is almost certainly not representative of the province’s indigenous character. Only some 3,200 people live in Purus, and its indigenous communities are small, scattered, relatively isolated, and divided among several different language groups. It seems likely that the votes cast in this province came mostly from the economically dominant Spanish-speaking population. It is a different story in the more populous province of Condorcanqui, which has almost 50,000 inhabitants. Most of them are members of the Awajún (or Aguaruna) ethnic group, who number some 70,000 and have achieved a significant degree of political organization. As noted in their Wikipedia article:

In the later half of the 20th century, the arrival of Protestant and Jesuit missionaries, the building of roads, and the construction of an oil pipeline created substantial tension between the Aguaruna people, poor agricultural colonists, state agencies, and corporations. In response to new threats to their way of life the Aguaruna began to organize a political and social response to defend themselves on the basis of principles consistent with other rights of indigenous peoples.

Amazonian Languages & Peru’s 2026 Election Map

Given such political organization among the Awajún/Aguaruna people, it is not surprising that Condorcanqui overwhelming supported the leftwing candidate Roberto Sánchez in the 2026 Peruvian presidential election.

Although almost all non-Spanish-speaking areas of Peru supported Roberto Sánchez in the election, it does not follow that Spanish-speaking areas correspondingly voted for Keiko Fujimoiri. The coastal strip of southern Peru and the large inland city of Arequipa in the same area, for example, are strongly Spanish speaking, yet gave Sánchez solid support.

Arequipa, Spanish Languages & Peru’s 2026 Election Map

Language & Ethnicity in the 2026 Peruvian Presidential Election Read More »

Could a Druze State Be Emerging in Southern Syria? If So, What Would It Encompass?

Article-Grid, Ethnicity, Geopolitics, Regionalism, Southwest Asia and North Africa / Martin W. Lewis

Since the fall of Syria’s Assad Regime in December 2024, I have been wondering what would become of the Druze people and the areas in which they are the majority population. Particularly important is the Jabal al-Druze, or “Mountain of the Druze,” which is clearly visible on satellite images of southern Syria (see the two maps posted below, with different international borders). Syria’s new president, Ahmed Hussein al-Sharaa, now portrays himself as a moderate and has vowed to protect the Druze and other minority groups. But al-Sharaa, who recently went under the nom du guerre Abu Mohammad al-Julani, was once a member of al-Qaeda and was a founder of the Al-Nusra Front, another hard-core Islamist group deeply hostile to the non-Muslim Druze.

Jabal al-Druze in Its Geopolitical Environment, Satellite Imager Map 1

Jabal al-Druze in Its Geopolitical Environment, Satellite Imager Map 2

Not surprisingly, the situation in the Jabal al-Druze remains tense. On March 20, 2026, Israel bombed Syrian military sites in the vicinity, claiming that it did so to protect the Druze. Roughly 150,000 Druze live in Israel, where their men serve in the national defense forces (IDF]); according to Wikipedia, “Druze men have one of the highest enlistment rates in the country, at over 80%.” After the attack on Syrian military assets, Israel’s defense minister said that “We will not allow the Syrian regime to exploit our war against Iran and Hezbollah to harm the Druze. If necessary, we will attack with greater force.”

Shortly after these events, Mario Nawfal, an influential but controversial Lebanese-Australian entrepreneur and political commentator based in the UAE, made some strong claims on X (“Twitter”) about an emerging Druze state being “midwifed” by Israel in Syria’s Suwayda Governorate, which includes the Jabal al-Druze (see the figure posted below). I find Nawfal’s comments intriguing yet exaggerated. But I view that map that he used to illustrate his post as highly misleading.

Mario Nawfal’s Map of an Emerging Druze State in Syria

Nawfal’s map does a decent job of showing two sizable areas with Druze majorities in southern Syria, which he depicts in green (compare his map with a map of Druze distribution posted on the blog Informed Comment, which is reproduced below). The problem lies in his depiction of the area between these two Druze-majority zones. He has colored this area brown but has given it a green outline while placing within it the label “Autonomous Druze Territory” along with a depiction of the Druze flag. Such mapping implies that any emergent Druze state or autonomous territory would include this area. This prospect is not only highly unlikely, but any attempt to create it would almost certainly result in a deadly and protracted conflict. The brown area on Nawfal’s map includes Syria’s Daraa Governorate, which is overwhelmingly Sunni Muslim (around 97 percent by most estimates) and hostile to both Druze and Israeli geopolitical designs. It also has more than twice the population of the Druze-dominated Suwayda Governorate (over one million as opposed to fewer than 500,000 inhabitants). (Note that M. Izady’s detailed map of language and religion in the area shows a sizable Christian area in eastern Daraa. But Christian numbers here were small to begin with and have dropped significantly in recent years.)

Druze Distribution Map from Informed Comment

M. Izady Map of Language and Religion in Southern Syria

In my opinion, the situation of the Druze in both Syria and Lebanon deserves far more attention than it has been receiving. As if often case, the more intense conflicts in the larger region tend to crowd out important but less heated geopolitical struggles.

Could a Druze State Be Emerging in Southern Syria? If So, What Would It Encompass? Read More »

Where Is Khwarazm Located in the Current Geopolitical Framework & How Did It Get There?

Article-Grid, Central Asia, Culinary Geography, Historical Geography / Martin W. Lewis

As explained in the previous GeoCurrents post, the historical region of Khwarazm, located south of the (former) Aral Sea in Central Asia, was a major intellectual hub in late-ancient and medieval times, producing two of the world’s greatest scholars, al-Biruni and al-Khwarizmi. It also supported several powerful states based on its extensive irrigated agriculture, the rich fisheries of the Aral Sea (or Lake of Khwarazm), and long-distance trade. Yet it is now a largely ignored region characterized by a lagging economy, political tension, and extreme environmental degradation. Khwarazm’s current troubles will be explored in a later GeoCurrents post. This essay explores a more basic issue: where exactly is historical Khwarazm situated in the geopolitical framework of today? The answer to this simple question is more complicated that it might seem.

The Khwarazm region is clearly visible on satellite images, where it appears as green splotch of verdant farms set amid brown deserts and the desiccated salt-beds of the former Aral Sea. But how closely does this current oasis-like area correspond to historical Khwarazm? A map of the Khwarazm’s ancient fortresses seemingly indicates that the historical region’s core territory lay in the southeastern portion of the modern irrigated zone. A map of the cities of Khwarazm in the early Islamic period, however, shows that Khwarazm during this period largely corresponds to the modern agricultural area.

I had long thought that historical Khwarazm was largely coterminous with the modern “Autonomous Republic of Karakalpakstan” in Uzbekistan. The Wikipedia article on the region, after all, tells us that the territory of Karakalpakstan“covers the classical land of Khwarazm, which in classical Persian literature was known as Kāt.” In actuality, Karakalpakstan covers less than half of this “classical land” (see the maps below). Much of Khwarazm is in what is now the Daşoguz region (province) of northern Turkmenistan, and a large portion is in Uzbekistan’s Khorazm (or Xorazm) region (province) (see the second figure posted below). Because of this political division, it is not easy to find the current population of historical Khwarazm. To do so, one must add the figures from Karakalpakstan (2,015,000), Uzbekistan’s Khorazm region (1,958,000), and Turkmenistan’s Daşoguz region (1,550,000), which yields a total population of around five and a half million people. Given the relatively small size of its cities, modern Khwarazm has a high rural population density, which can easily be appreciated by closely examining satellite images of the region.

Karakalpakstan and the Former Aral Sea map

Historical Khwarazm and Divided Khwarazm Today, map

It must be asked why such a coherent area would be geopolitically divided, split between two countries and, in one of those countries, between a supposedly autonomous republic and a regular region (province). The answer is found in the internal geopolitics of the Soviet Union. From 1511 to 1920, Khwarazm formed the Khanate of Khiva, ruled by Uzbek Khans. In 1873, this state was stripped of much of its territory and reduced to the status of a Russian protectorate. But it remained largely independent, as indicated by many maps made at the time (see below). In 1920, three years after the Russian Revolution, Khwarazm was transformed into the Khorezm People’s Soviet Republic, a first-order division of the Soviet Union. Three years later its status was downgraded to that of a “Socialist Soviet Republic” (SSR) and the next year it was dissolved, its land divided between the Uzbek and Turkmen SSRs and the new “Karakalpak Autonomous Oblast.” In 1936, the “autonomous oblast” was joined to the Uzbek SSR, which eventually became Uzbekistan. According to Wikipedia, “the history of the short-lived [Khorezm People’s Soviet Republic] remains murky and the way its government functioned is not largely clear.” It also unsurprisingly informs us that “its officials fervently opposed the delimitation plans of the Khorezmian Republic (which had to be carried out under Stalin’s orders) but in the end they were forced to concede.”

Khanate of Khiva Depicted as Independent in the Early 1900s map

The ostensible reason for the Soviet division of Khwarazm was the ethnolinguistic differentiation of the region. According to often-abused Leninist principles, each major ethnic group or nationality in the Soviet Union was supposed to have some measure of cultural and political autonomy. But by the early twentieth century, Khwarazm was no longer culturally homogenous and was, as a result, divided. This is a complicated story, however, that needs some unpacking.

As discussed in the previous post, ancient Khwarazm had its own Iranian language, which gradually yielded to Persian in the early Islamic period. Somewhat later, as the military power of Turkic-speaking pastoralists grew, their dialects began to spread. After the Mongol conquest of 1219-1221, Khwarazm came under the rule of the Chagatai Khanate, named after, and initially ruled by, the second son of Genghis Khan. Under the patronage of this powerful state, a new Persianized Turkic literary language emerged, which was generally called “Turki” by its speakers and later deemed “Chagatai by Western scholars. This language gradually differentiated into two modern tongues, Uzbek* and Uyghur. Formal Chagatai continued to be employed as a literary language, however, until the early twentieth century. Its final flourishing occurred in the Khanate of Khiva, based in Khwarazm. As noted in the Wikipedia article on the language:

The reign of Muhammad Rahim II [of Khiva] marked the peak of a cultural revival, during which “more than a hundred works were translated, mostly from Persian into Chagatai Turkic.” Muhammad Rahim II introduced printing Khiva in 1874. He was also “a munificent patron” and wrote [Chagatai] poetry under the pen name Feruz.

Although Persianized Turki (Chagatai/Uzbek) became the common language of Khwarazm, pastoral nomads speaking non-Persianized Turkic dialects continued to infiltrate, raiding and often settling. In the Daşoguz region, one of their tongues, Turkmen, was established as a minority language. In the northern periphery of Khwarazm, the ever-shifting channels of the Amu Darya delta created an unstable environment well suited to a combination of semi-nomadic herding, farming, and fishing. Here pastoralists from the north, speaking a dialect closely related to Kazakh, settled in the eighteenth and early nineteenth centuries. They would come to be known as the Karakalpak people, their language labeled with the same term.

One could therefore argue that the Soviet political division of Khwarazm was instituted to reflect the region’s ethnolinguistic divisions, yielding a Karakalpak-speaking region in the north and east (Karakalpakstan), a Turkmen-speaking region in the center (Daşoguz in Turkmenistan), and a core Uzbek-speaking region in the southeast (Khorazm in Uzbekistan). Significantly, although all of these language are Turkic, they belong to different branches of this large family.**

Such a view, however, is fundamentally misleading. Under Stalin, the theoretically autonomous areas of Soviet minority groups were effectively stripped of their autonomy and were in many cases re-engineered to prevent disfavored peoples and regions from gaining too much power. It was probably for this reason that the original “Khorezm People’s Soviet Republic” was disassembled and divided between what would eventually became Turkmenistan and Uzbekistan. The Daşoguz region, for example, had a large Uzbek majority when it was assigned to the Turkmen Soviet Socialist Republic. It took aggressive “Turkmenization” programs, under both the Soviet Union and independent Turkmenistan, to gave it a Turkmen-speaking majority. Yet up to this day, roughly a third of the people of Daşoguz still speak Uzbek as their mother tongue. Uzbek probably remains the majority language of the larger cities of this mostly rural region. Interestingly, some language maps portray the settled portion of Daşoguz as Turkmen speaking while others portray it as Uzbek speaking (see below).

However the processes of ethno-linguistic change and geopolitical machination occurred, the common language and culture of Khwarazm was largely erased, along with its standing as discrete unit of political geography. This undermining of the identity of Khwarazm would eventually be accompanied by the weakening of its economy and ecological foundation, as we will see in the next GeoCurrents post.

* The naming of these languages has a confusing and contradictory history. Historically, “Uzbek” referred to a Kipchak Turkic language, closely related to Kazakh, that was spoken by nomadic pastoralists who conquered much of the agricultural zone of Central Asia. Most of their descendants adopted the Persianized Karluk Turkic language of the majority population, variously called “Turki,” “Sart,” or, for its literary form, “Chagatai.” The original Uzbek language, however, continued to be spoken by small grounds of pastoralists. In the early Soviet period, the majority Karluk language was renamed “Uzbek,” while the original Uzbek language essentially disappeared due to the decline of nomadic pastoralism in the region.

** Turkmen is an Oghuz Turkic langue, belonging to the same branch as Turkish and Azerbaijani. Karakalpak is a Kipchak Turkic language, belonging to the same branch as Kazakh and Kyrgyz. Uzbek is a Karluk Turkic language, belonging to the same branch Uyghur.

Where Is Khwarazm Located in the Current Geopolitical Framework & How Did It Get There? Read More »

Khorasan, Greater Khorasan, Greatest Khorasan… and the Geography of Medieval Scholarship

Article-Grid, Cultural Geography, Historical Geography, Southwest Asia and North Africa / Martin W. Lewis

The modern region of Khorasan covers three provinces in northeastern Iran: North Khorasan, Razavi Khorasan, and South Khorasan. Razavi Khorasan, the name of which refers to Ali al-Rida, the eighth Imam of Twelver Shia Islam, is by far the most populous and important of the three (see the map below). Before 2004, this entire region constituted a single Khorasan province. Historically speaking, however, the term “Khorasan” – “land of the rising sun” – covered a much larger area. In the strict sense of the term, it also extended across northern Afghanistan and most of Turkmenistan, and is usually mapped accordingly. In this scheme, Khorasan terminated at the Oxus River (Amu Darya), with the lands beyond it called “Transoxiana” in Western sources and sometimes Mā Warāʾ an-Nahr (“the lands beyond the river”) in Persian and Arabic sources. But Persian and Arabic geographers often further extended Khorasan across the Amu Darya to include much of what is now Tajikistan and Uzbekistan.

Iran’s Old and New Provinces of Khorasan map

Khorasan in this “greater” sense might be regarded as the eastern portion of the historically Persian-speaking zone. As noted in the Wikipedia article on “Khorasan Province:”

Some of the main historical cities of Persia are located in the older Khorasan: Nishapur and Tus (now in Iran); Merv and Sanjan (now in Turkmenistan); Samarkand and Bukhara (both now in Uzbekistan); Herat and Balkh (now in Afghanistan); and Khujand and Panjakent (now both in Tajikistan).

Historical Greater Khorasan and the Current Extent of the Persian Language map

As the map posted above shows, most of this “older Khorasan” is inhabited to this day by Persian speakers, although the language is commonly called “Dari” in northern Afghanistan and “Tajik” in Tajikistan. Storied “Uzbek” cities such as Bukhara and Samarkand are probably still majority Persian-speaking (as the mother tongue), although official statistics claim that most residents primarily speak Uzbek, a Turkic language. Turkmenistan, on the other hand, is no longer part of the Persian language zone. Its once-great Persian cities, such as Merv and Sanjan, were destroyed centuries ago.

As was noted in the previous GeoCurrents post, the historical region of Khorasan is commonly regarded as forming the intellectual core of the Persian-speaking world from roughly 800 to 1200 CE/AD, the time of its greatest scholarly achievements. The Wikipedia article on the region lists 10 prominent scientifically minded scholars from the region who lived during this period. As the maps posted below show, four of them were from what is now the Iranian region of Khorasan (all from the province of Razavi Khorasan), two were from what is now northern Afghanistan, a region once called Bactria, and two were from Transoxiana, or Mā Warāʾ an-Nahr, a region once called Sogdia or Sogdiana. All eight were thus natives of “Greater Khorasan” in the extended sense of the term.

Birthplaces of Medieval Scientists from Khorasan map

Birthplaces of Medieval Scientists from Khorasan, map 2

The remaining two scholars were from Khwarazm, an important region focused on the inland delta of the Amu Darya (Oxus) River just to the south of the (former) Aral Sea. This region has rarely been included in Khorasan. AI (Grok), however, tells me that that the noted tenth-century Persian geographer and traveler Istakhri did classify it in such a manner, reflecting “the flexible, sometimes expansive use of “Khorasan” in medieval texts, where it could loosely denote eastern Iranian/Central Asian lands east of the central Iranian deserts.”

The term “Khorasan” can thus be regarded as possibly denoting four separate but nesting regions: modern Khorasan (northeastern Iran); greater Khorasan in the strict sense (adding much of Turkmenistan, and northern Afghanistan); greater Khorasan in the extended sense (adding Transoxiana or Mā Warāʾ an-Nahr); and “greatest” Khorasan (adding Khwarazm).

If Khorasan is to include Khwarazm, in cannot be defined as the eastern portion of the historically Persian-speaking region, as Khwarazm was not, at least during its intellectual heyday, Persian speaking. Its native tongue at the time was Khwarezmian, an Eastern Iranian language more closely related to Pashto (the language of the Pashtun people) than to Persian/Farsi. (It was even more closely related to Sogdian and Bactrian, which had been spoken in Transoxiana and northern Afghanistan respectively before being replaced by Persian.) But Khwarazm had been part of the greater Persian cultural sphere, and by the thirteenth century its own language was also being replaced by Persian. Turkic dialects later spread, and today the people of the region speak the Turkic language Karakalpak (closely related to Kazakh), and are ethnically deemed Karakalpaks (literally meaning “black hats.

The two scholars from Khwarazm noted on the map of medieval scientists from “Khorasan”, al-Buruni and Al-Khwarizmi, deserve further consideration and will therefore be examined in the next GeoCurrents post.

Khorasan, Greater Khorasan, Greatest Khorasan… and the Geography of Medieval Scholarship Read More »

The Indo-European Controversy: Facts and Fallacies in Historical Linguistics	Diversity Amid Globalization: World Regions, Environment, Development (6th Edition)
The Myth of Continents: A Critique of Metageography	Green Delusions