Martin W. Lewis, Author at GeoCurrents

Air-Conditioning Needs and Cartographic Failure at the Washington Post

Article-Grid, Environmental Geography, Northern California / By Martin W. Lewis

The Washington Post recently ran an article entitled “Addicted to Cool: How the Dream of Air Conditioning Turned into the Dark Future of Climate Change,” which features three maps of “Summer Days Requiring AC” in the U.S. at different periods of time (1981-200, 2001-2002, and 2060). As expected, the region needing air conditioning is projected to expand. Determining how many days actually “require” air conditioning is, however, an impossible task, as different people vary significantly in their cooling desires and demands, while housing design and shade considerations make big differences as well. Understandably bypassing such complexities, the newspaper used the heat index, a measurement of temperature and humidity, as a proxy. Unsurprisingly, their maps show that a large area of the country already needs summer air conditioning, and that in the decades to come the need for cooling will geographically expand.

The maps included in the article, however, are not impressive, to put it mildly. Their problems are particularly severe regarding California. As can be seen on the map detail of Southern California posted below, the Post’s mapping accurately shows the eastern deserts and the inland western regions as needing air conditioning on most if not all summer days. It also accurately depicts the highest elevation areas as rarely requiring it. But the same map also portrays the coastal zone as AC-dependent. This is not true. Downtown San Diego, for example, has an average July high temperature of 75 degrees Fahrenheit and an average July low of 66 (over the 1991-2020 period). Further north, in Rancho Palos Verdes, similar conditions prevail. According to Weather Spark, “The hottest month of the year in Rancho Palos Verdes is August, with an average high of 76°F and low of 64°F.” The same site also notes that “Over the course of the year, the temperature typically varies from 51°F to 76°F and is rarely below 46°F or above 84°F.” This is climate that very rarely calls for cooling.

The map is equally inaccurate in its depiction of Northern California. As can be seen on the map detail posted below, it does capture the cool summers characteristic of the coastal regions in and around San Francisco and Monterey bays. It completely misses the fact, however, that other coastal regions also have mild summers. The average July high and low temperatures in Point Arena, Santa Cruz, and Carmel are, respectively 65 and 50 degrees F.; 74 and 54; and 70 and 53 (all based on the climatological data found in the Wikipedia articles on these towns). In Point Arena, one is more likely to want heating than air conditioning in June and July, yet the map indicates that cooling is needed on most summer days.

Other odd features mar the map. As can be seen, the city of San Francisco is mislocated in the bay and on its eastern shore. The national map also features a faint white line that traces part of San Francisco Bay and would appear to indicate the actual coastline, at least in some areas. Was one map imprecisely overlaid on another?

Although these problems are serious enough, it is the map of projected air-conditioning needs in the year 2060 that truly fails. This can be seen easily on the paired maps showing current and projected AC requirements in California. Here much of the currently cool Big Sur coastal zone is projected to have much reduced air-conditioning needs by 2060. This region of projected cooling is bizarrely shown as extending over the Santa Lucia Range into the southern Salinas Valley, an area that now experiences warm summers (King City has an average July high of 85 degrees F.). Similarly, the currently warm inland area north of Santa Barbara is shown as being expected to have much cooler summers in 2060 than it does today, while with the rest of Southern California is projected to warm.

One can only wonder whether the cartographers in question actually examined these maps before publication, or, if they did, whether they have much of an understanding of the geography of climate. It often seems that journalists use maps as mere ornaments or, alternatively, to have the appearance of spatial precision without the substance. The maps in this article do little more than make the trite point that “more of us will need air conditioning as the climate warms.” Readers deserve better, especially from a once-great paper that is owned by the third richest person in the world.

Geographical Patterns of Income Inequality in the U.S. at the State and County Levels

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

I have long been intrigued by the geography of income inequality in the United States. As maps of the GINI coefficient show, income inequality is highest some of the country’s richest states (New York, Connecticut) and in some of its poorest (Louisiana, Mississippi). Similarly, some of the country’s most Democratic-voting states and some of its most Republican-voting ones are characterized by pronounced income inequality. Relatively low levels of income inequality are concentrated in an area that might crudely be described as the center-north-west, with four contiguous states occupying the lowest category on the map (Utah, Idaho, Wyoming, South Dakota). Low population density characterizes states with low income inequality. All of the states in the bottom two categories on this map except Hawaii have a lower-than-average population density. Politically, these states show the same mixed pattern that characterized the most economically polarized states. Although all the states in the lowest GINI category are bright red on electoral maps, two that fall into the next lowest category (Vermont, Hawaii) are bright blue.

A county-level GINI map clarifies the geography of U.S. income inequality and reveals some interesting patterns (unfortunately, the best map that I could find on this topic is somewhat dated). As can be seen, the elevated levels of income inequality found in northeastern states is largely an urban phenomenon. In the southeast, in contrast, some counties with high GINI coefficients are metropolitan (in southeastern Florida, for example), but others are markedly rural. In Western and Great Plains states characterized by relatively low income inequality, quite a few of their rural counties have high GINI scores.

In North Dakota, South Dakota, and Montana, some rural counties characterized by high income inequality also have a high percentage of Native American residents. To illustrate this correlation, I have placed a GINI map of the Dakotas next to one of indigenous population percentage. But there are a few striking exceptions to this pattern, two of which are noted on the map. As can be seen, Divide County, North Dakota has a small Native American population and a high GINI coefficient. Pete Morris’s agricultural explanation of income inequality, outlined in his comment on yesterday’s post, is probably relevant here as well. In contrast, Buffalo County, South Dakota has a large Native American population and a low GINI coefficient. Both of these counties have very small populations. Buffalo County is noteworthy for having the least populous county seat in the United States (Gann Valley, with a population of 14).

In the south-central region of the country, most counties with high levels of income inequality have large black populations. But again, interesting exceptions can be found. As can be seen, Jefferson County, Arkansas has a high percentage of Black residents and a mid-level GINI ranking. In contrast, Marshall County, Alabama has a very low percentage of Black residents and a high level of inequality. Jefferson County, intriguingly, is known for its concentration of “correctional facilities,” mostly located in and around Pine Bluff. Marshall County, Alabama, in contrast, is part of the Huntsville-Decatur Combined Statistical Area, a region noted for its many well-paid technical workers, owing largely to that presence of NASA’s Marshall Space Flight Center, the United States Army Aviation and Missile Command, and the FBI ‘s Operational Support Headquarters. Marshall’s largest city, Albertville, is mostly noted, however, as the home of the fire-hydrant-manufacturing Mueller Company. As noted by the Wikipedia article on the city, “Albertville holds the title of “Fire Hydrant Capital of the World.” To commemorate the one millionth fire hydrant, a chrome fire hydrant was placed outside the Albertville Chamber of Commerce.”

The next GeoCurrents post will examine the geography of income inequality in the country’s largest metropolitan areas.

Explaining Seeming Discrepancies on County-Level Income Maps of the United States

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

When working on a recent GeoCurrents post that involved maps of income in the United States, I noticed a few unusual patterns. A number of counties, for example, are mapped as having relatively high per capita personal income and relatively low median household income, whereas in others the opposite pattern obtains. In part this is a matter of household size, an explanation that works particularly well for Utah. Consider, for example Utah County, Utah which is characterized by relatively low per capita personal income, relatively high median household income, and a large number of people per household. In contrast, Grand County is characterized by relatively high per capita personal income, relatively low median household income, and a small number of people per household.

In Utah, the number of people per household correlates closely with religion. Members of the LDS church (Mormons) often have high fertility rates, leading to large households. Utah County, Utah, home of Brigham Young University, is usually considered the cultural center of the LDS faith. As can be seen on the second map below, Utah County has one of the highest fertility rates in the country. In contrast, Grand County has a relatively low fertility level (which is not shown in the map due to its small population) and the lowest LDS percentage in the state. Whereas Utah as a whole is roughly 62% Mormon, in Grand County the figure is only 26%.

These easy correlations, however, collapse when one examines North Dakota. As can be seen on the map below, Cavalier County has the highest per capita personal income in the state, which is why it is outlined with a heavy white line on the map posted here. But Cavalier County is also characterized by relatively low median household income and relatively few people per household. This seeming anomaly can be explained by taking into account the different way that the two income measurements are determined. Median household income is calculated by taking the income of all households in a county and finding the middle point; per capita personal income, on the other hand, is calculated by dividing the total income of all persons in the county by the population. If a county has a small population with a few very high-income individuals, the per capita personal income figure is inflated, whereas the median household income figure will remain low if most households have lower incomes.

If this explanation is correct, one would expect Cavalier County to have a relatively high Gini Coefficient, which measures the degree of inequality. The most recent GINI map of all U.S. counties that I was able to find (posted below) indicates that this is indeed the case. Overall, North Dakota is characterized by very wide range in GINI figures, which is probably largely an attribute of the small populations of most of its counties.

Regardless of its income level, Cavalier has not exactly been a thriving county over the past century. It had more than 15,000 people in 1920 and fewer than 4,000 in 2020.

How Well Can ChatGPT Analyze a Complex Map? Part 2: Rich Counties in Red States

Article-Grid, Elections / By Martin W. Lewis

In the 2020 U.S. presidential election, the neighboring states of Wyoming, Idaho, and Utah all strongly supported Donald Trump, giving him respectively 70, 64, and 58 percent of their votes. But in all three states several counties opted for Joe Biden, a few by considerable margins (see the map below). All of these seeming exceptions are easily explained on the basis of demographic characteristics. Latah County, Idaho and Albany County, Wyoming, for example, are the sites of their states’ main universities, and college towns and nearby communities usually support candidates in the Democratic Party. Salt Lake County, Utah, in contrast, contains a major metropolitan area, which are also generally “blue” on electoral maps. Sparsely populated Grand County, Utah falls into neither category, but its main town, Moab, is a major center of outdoor recreation. Western communities with economies based on outdoor amenities also typically vote for Democrats. Teton County, Idaho falls into this same category.

The darkest shade of blue on this map is given to Blaine County, Idaho and Teton County, Wyoming. Their deciding demographic characteristic, as of Summit County Utah, is a variant of the environmental amenities theme, as all are sites of major ski resorts and their associated residential communities. What really distinguishes these three counties from their neighbors, however, is their wealth. As the somewhat dated map below posted shows, these are among the richest counties in the United States. With the exception of those whose wealth is largely derived from natural resources and farming, wealthy counties generally support Democratic candidates.

I wanted to see if ChatGPT could find and explain the seeming anomaly of low-density, white-dominated counties located in the conservative intermontane West strongly favoring Joe Biden over Donald Trump. I therefore made the following pointed, and hinting, request:

“Write a 500-word analysis of the geographical patterns found in the 2020 United States presidential election based on a map of all U.S. counties shaded by the margin of victory, paying attention to counties that were exceptions to the general geographical rules found on this map and taking into account per capita income.”

The essay that the chatbot gave in response is reproduced below. As can be seen, it fails to either locate or explain this interesting phenomenon.

Thinking that the chatbot would perform better if I limited its analysis to Utah, Idaho, and Wyoming, I gave it another chance. This response is also included below. Here again it failed its assignment. Although ChatGPT did notice and explain the seeming oddity of Teton County, Wyoming, it also included a major error in the same sentence: “County-level data shows that Trump won all but one county in the state, Teton County, which is known for its higher-income residents and a significant number of transplants from more liberal areas.” Perhaps ChatGPT did not notice that Biden also won Albany County, Wyoming because he took less than 50 percent of its votes. But Biden’s margin of victory here was still a reasonably solid 2.76 percent.

To sum up to this brief series, ChatGPT seems to be particularly bad at geographical analysis. Somehow, that does not surprise me.

ChatGPT Response 1:

The 2020 United States presidential election exhibited intriguing geographical patterns when analyzed using a county-level map shaded by the margin of victory. These patterns were influenced by various factors, including economic conditions such as per capita income. Let’s delve into the key observations, general trends, and notable exceptions:

General Patterns:

Urban-Rural Divide: A conspicuous urban-rural divide was evident. Urban and suburban areas, particularly along the coasts and in major metropolitan regions, tended to favor the Democratic candidate, Joe Biden. Conversely, rural and less densely populated areas, especially in the Midwest and the South, leaned toward the Republican incumbent, Donald Trump.
Economic Centers: Counties encompassing major economic hubs, like Manhattan (New York County), Los Angeles County, and Cook County (Chicago), predominantly supported Biden. These urban centers typically have higher per capita incomes due to their concentration of economic activity, educational institutions, and diverse populations.

Notable Exceptions:

Several counties deviated from the general patterns, and these exceptions can be partly explained by considering per capita income:

Rural Prosperity: In some rural counties, particularly in states like Wyoming, North Dakota, and parts of Texas, which benefit from natural resource industries (e.g., oil, gas, and agriculture), Trump received strong support despite their rural status. These exceptions underscore the economic diversity within rural America, where high per capita incomes can translate into political allegiance to Trump.
Urban Economic Variation: In urban areas, especially in Rust Belt cities like Youngstown, Ohio, and Scranton, Pennsylvania, Trump gained ground in counties with lower per capita incomes. Economic hardship and the decline of manufacturing jobs in these regions likely contributed to this shift, illustrating how economic conditions can influence political preferences.
Florida’s Diversity: Florida’s diverse economic landscape and a significant retiree population resulted in a mix of counties concerning per capita income and political leanings. While Miami-Dade County, with its high per capita income derived from tourism and international trade, experienced a shift toward Trump among some Hispanic voters, other counties with lower per capita incomes, such as Duval County (Jacksonville), swung toward Biden.
Pennsylvania’s Suburban Complex: In Pennsylvania, Montgomery County, a suburban county with a high per capita income due to its proximity to Philadelphia, leaned decisively toward Biden, aligning with the urban-suburban pattern. However, nearby counties like Bucks and Northampton, with lower per capita incomes, displayed stronger support for Trump, suggesting a complex interplay of economic and demographic factors.
Colorado’s Diversity: Colorado’s political landscape was influenced by factors like per capita income. While highly populated and affluent counties around Denver favored Biden, more rural counties with lower per capita incomes, especially in the Western Slope, remained staunchly Republican.
Urban-Rural Divide in Swing States: In pivotal swing states like Wisconsin, Michigan, and Pennsylvania, the urban-rural divide played a substantial role. Notably, in Michigan, Macomb County, with a median income below the national average, shifted toward Trump, while higher-income suburban counties near Detroit leaned Democratic.
Texas’s Suburban Shift: Texas saw suburban counties around major cities like Dallas and Houston, which typically have higher per capita incomes, shifting toward Biden. This shift challenged the notion of Texas as a solid Republican stronghold, emphasizing the impact of economic diversity in suburban areas.

In summary, the geographical patterns observed in the 2020 U.S. presidential election, as reflected in county-level margins of victory, revealed a dynamic interplay of economic conditions, demographics, and regional dynamics. While urban areas with higher per capita incomes generally favored Biden, exceptions in both rural and urban settings demonstrated the multifaceted nature of American politics. Understanding these exceptions is essential for a comprehensive analysis of the election results, as they underscore the complex relationship between economic well-being and political preferences in a diverse and evolving nation.

Prompt for second experiment: “Write a 500-word essay analyzing the 2020 presidential election returns from Idaho, Wyoming, and Utah using country-level data.”

ChatGPT Response 2.

Title: Analyzing the 2020 Presidential Election Returns in Idaho, Wyoming, and Utah

Introduction

The 2020 United States presidential election was one of the most highly anticipated and closely watched elections in recent history. While national-level analysis often dominates discussions, it is equally important to examine the results at the state level to gain a deeper understanding of the political landscape. In this essay, we will analyze the 2020 presidential election returns in the states of Idaho, Wyoming, and Utah using county-level data, shedding light on key trends, demographics, and political dynamics that shaped the outcomes in these predominantly conservative states.

Idaho: A Stronghold for Republicans

Idaho, known for its deep-rooted conservatism, continued its tradition of favoring the Republican candidate in the 2020 election. With a state-level margin of over 30 percentage points in favor of Donald Trump, it was one of the most solidly Republican states in the nation. County-level data reveals that the Republican dominance was consistent throughout the state, with Trump carrying the majority of counties.

One notable observation from the Idaho data is the urban-rural divide. While metropolitan areas like Ada County (Boise) leaned more Democratic, rural counties, particularly in the eastern part of the state, overwhelmingly supported Trump. This pattern reflects a broader national trend seen in many states, where urban areas tend to favor Democratic candidates, while rural areas lean Republican.

Furthermore, Idaho’s predominantly white population played a significant role in shaping the outcome. The lack of racial diversity in the state contributed to a more homogeneous political landscape, with racial and ethnic minorities having less influence on the election outcome compared to more diverse states.

Wyoming: A Republican Stronghold with Unique Dynamics

Wyoming, like Idaho, is considered a Republican stronghold, but it exhibits some unique dynamics. With a staggering margin of over 40 percentage points in favor of Trump, Wyoming was one of the most reliably red states in the 2020 election. County-level data shows that Trump won all but one county in the state, Teton County, which is known for its higher-income residents and a significant number of transplants from more liberal areas.

A striking feature of Wyoming’s election returns is its low population density. The state’s vast, sparsely populated rural areas play a crucial role in shaping the political landscape. In such regions, where traditional values and limited government intervention are often prized, the Republican message resonates strongly.

Another factor to consider is Wyoming’s energy industry, which is a significant contributor to the state’s economy. Trump’s promises to support fossil fuels and reduce environmental regulations likely bolstered his support among voters in this state, where energy policy is of paramount importance.

Utah: A Conservative State with Nuanced Tendencies

Utah is another conservative state, but it has shown more nuanced political tendencies. While Trump comfortably won Utah, his margin was smaller than in Idaho and Wyoming. County-level data reveals some interesting insights. Utah’s population centers, including Salt Lake County, leaned more Democratic, driven in part by a younger and more diverse demographic.

One notable aspect of Utah’s political landscape is the presence of a significant independent and moderate Republican constituency. This group, often referred to as “Never Trumpers,” was less enthusiastic about the president. The presence of prominent figures in the state’s Republican establishment who openly criticized Trump may have contributed to this trend.

Conclusion

In analyzing the 2020 presidential election returns in Idaho, Wyoming, and Utah using county-level data, it is clear that these predominantly conservative states exhibited unique political dynamics. Idaho and Wyoming remained solid Republican strongholds, with their rural areas and homogeneous populations contributing to the overwhelming support for Donald Trump. Meanwhile, Utah, while still conservative, showed more nuanced tendencies, with urban areas and a moderate Republican constituency influencing the election outcome.

Understanding these state-level dynamics is crucial for comprehending the broader political landscape in the United States. The 2020 election provided valuable insights into the interplay of demographics, regional characteristics, and political ideologies that shape electoral outcomes in these states. As we look ahead to future elections, it is essential to continue analyzing state-level data to gain a more comprehensive understanding of the American political landscape.

How Well Can ChatGPT Analyze a Complex Map, Part 1?

Article-Grid, Elections / By Martin W. Lewis

This second-to-last post in the current series on ChatGPT examines the chatbot’s ability to find and explain patterns found on a relatively complex map. ChatGPT cannot (yet?) make maps, but can it intelligently analyze them? To test its abilities, I asked it to write an analytic essay on the geographical patterns found on a map of the 2020 U.S. presidential election with counties shaded by the margin of victory (a map of this sort is posted below). It managed to capture the basic contours reasonably well, focusing on regional divisions and the urban-rural split. But it missed many of the more subtle patterns and failed to notice any exceptions to the general rules that it had inferred. I therefore began to refine my queries, asking for more details about specific patterns. I have been told that this “narrowing in” strategy is needed to get the most out of ChatGBT. I was not impressed, however, with the results. If anything, the chatbot lost coherence when given additional information and more pointed requests.

Although I ran a number of these experiments, I will discuss the results of only two, one today and another in the next and final post in this series. The query that I used for today’s post specified that ChatGPT should pay attention to exceptions found for the general patterns that it had identified.

The chatbot’s response to this prompt is reproduced below, along with my comments on its errors and points of confusion. My comments are [IN BRACKETS AND IN ALL CAPS].

ChatGPT Response:

The 2020 United States presidential election was a historic event that brought about a complex tapestry of geographical patterns when analyzing the margin of victory across all U.S. counties. While certain general trends [I ASKED ABOUT PATTERNS, NOT TRENDS] emerged from the map, there were notable exceptions that provided a deeper understanding of the diverse political landscape in the country.

General Patterns:

When examining the map shaded by the margin of victory, several broad geographical patterns became evident:

1. Coastal Strongholds for Democrats: Counties along the East and West Coasts, from Maine to Florida and Washington to California, predominantly voted for the Democratic candidate, Joe Biden, with significant margins of victory [THIS IS ESSENTIALLY UNTRUE. WITH JUST A FEW EXCEPTIONS, BIDEN ENJOYED A “SIGNIFICANT MARGIN OF VICTORY” IN COASTAL COUNTIES ONLY IN CALIFORNIA. FROM SOUTHERN DELAWARE TO SOUTHERN FLORIDA, MORE COASTAL COUNTIES VOTED FOR TRUMP THAN BIDEN (SEE THE FIRST MAP BELOW). AND WHAT ABOUT THE GULF COAST? IS IT NOT “COASTAL”?]. This trend aligns with the historical tendency for urban and coastal areas to lean left. [ONLY IF ONE HAS A VERY SHALLOW DEFINITION OF “HISTORICAL.” AND WHY DOES THE TERM “URBAN’ APPEAR IN A SECTION FOCUSED ON THE COAST/INTERIOR DIVISION?]
2. Inland Republican Strongholds: In contrast, many [MOST, ACTUALLY] inland counties in the Midwest and South showed robust support for the Republican incumbent, Donald Trump [THIS IS TRUE NOT JUST IN THE MIDWEST AND SOUTH]. These regions often featured rural or less densely populated areas, which have traditionally leaned conservative. [MANY INLAND, NON-METROPOLITAN COUNTIES IN THE SOUTH SUPPORTED BIDEN, OFTEN BY A WIDE MARGIN. ALL HAVE BLACK MAJORITIES. THE CHATBOT COMPLETELY MISSED THIS IMPORTANT EXPLANATORY FACTOR.]
3. Urban-Rural Divide: The urban-rural divide was pronounced, with densely populated urban centers and their surrounding suburban counties favoring Biden. On the other hand, rural counties consistently supported Trump, reflecting deeper cultural and demographic divisions.

Notable Exceptions:

While these general patterns were dominant, there were several counties that stood out as exceptions, challenging the conventional wisdom [THIS IS NOT A MATTER OF “CONVENTIONAL VERSUS UNCONVENTIONAL WISDOM,” BUT RATHER ONE OF “FINE-LEVEL VERSUS GROSS-LEVEL ANALYSIS”] of their respective regions:

1. Urban Counties in Red States: In typically conservative states, such as Texas and Georgia, urban counties like Harris County (Houston) and Fulton County (Atlanta) experienced a significant shift toward the Democratic Party [HARRIS COUNTY, TX DID NOT SHOW A “SIGNIFICANT” BLUE SHIFT IN THIS ELECTION, AS CAN BE SEEN ON THE MAP POSTED BELOW]. These counties, characterized by their growing and diverse populations, flipped or came very close to flipping from Republican to Democratic [THERE WAS NO “FLIP” OR “NEAR-FLIP” INVOLVED! HARRIS COUNTY HAS BEEN HABITUALLY SUPPORTING DEMOCRATIC PRESIDENTIAL CANDIDATES FOR SOME TIME.]
2. Rural Counties in Blue States: Conversely, some traditionally Democratic states like Minnesota and Oregon saw rural counties trending toward the Republican candidate [RURAL COUNTIES IN OREGON HAVE BEEN MOSTLY “RED” FOR SOME TIME. ALSO, MY QUERY DID NOT ASK FOR AN ANALYSIS OF ELECTORAL CHANGE]. For instance, in Minnesota, several northern and western counties that had previously supported Democrats shifted toward Trump [AS CAN SEEN ON THE MAP BELOW, SEVERAL NORTHERN AND WESTERN MINNESOTA COUNTIES SAW A “BLUE SHIFT” FROM 2016 TO 2020].
3. Florida’s I-4 Corridor: Florida, a perennial swing state [NOT ‘PERENNIAL; ONLY FROM 1992 TO AROUND 2020], demonstrated a unique geographical pattern [WHAT IS “UNIQUE” ABOUT IT?]. While the state as a whole leaned slightly toward Trump, the I-4 corridor, which stretches from Tampa Bay to Orlando, remained a closely contested battleground. This region, known for its political diversity and mix of urban and suburban areas [RURAL AREAS AS WELL], defied the state’s overall trend [A LOT OF ATTENTION HAS BEEN GIVEN TO THE MIXED AND CHANGING VOTING PATTERNS OF THE I-4 CORRIDOR; SEE THE SECOND MAP BELOW. BUT AT THE COUNTY LEVEL, THIS REGION IS NOT REALLY AN EXCEPTION TO FLORIDA’S GENERAL ELECTORAL PATTERNS, AND IT SAW ONLY MINOR ELECTORAL SHIFTS FROM 2016 TO 2020; SEE THE THIRD MAP BELOW].
4. The Rural-Urban Split in Swing States: In swing states like Wisconsin, Michigan, and Pennsylvania, the margins of victory in urban counties, such as Milwaukee, Detroit, and Philadelphia [THERE IS NO “DETROIT COUNTY”], were crucial in securing these states for Biden. Conversely, the majority of rural counties in these states remained staunchly Republican. The tension between urban centers and rural areas was starkly visible in these key battleground states.
5. Native American Influence: In Arizona and New Mexico, Native American reservations and communities played a significant role in the election. Several counties with sizable Native American populations, such as Apache County in Arizona, favored Biden. This underscores the importance of demographic diversity and tribal influence in these southwestern states [YES, BUT WHAT ABOUT NATIVE AMERICAN COMMUNITIES IN OTHER STATES, PARTICUARLY MONTANA AND SOUTH DAKOTA?].
6. Maine’s Split Decision: Maine stood out with its unique [NOT UNIQUE; NEBRASKA HAS THE SAME SYSTEM] allocation of electoral votes. While the state as a whole went for Biden, it was one of only two states (alongside Nebraska) [OK, BUT THIS STATEMENT CONTRADICTS “UNIQUE’ IN THE PREVIOUS SENTENCE] to split its electoral votes by congressional district. Trump managed to secure one of Maine’s congressional districts, emphasizing the granularity of political preferences even within a single state. [“EVEN” WITHIN A SINGLE STATE? THERE IS TREMENDOUS “GRANULARITY” IN REGIONAL VOTING PATTERNS IN EVERY STATE. BESIDES THAT, MY QUERY WAS ABOUT VOTING PATTERNS AT THE COUNTY LEVEL, NOT ABOUT ELECTORAL VOTES]

In conclusion, the geographical patterns found in the 2020 U.S. presidential election, as indicated by the margin of victory across counties, painted a diverse and multifaceted picture of American politics. While overarching trends pointed to the urban-rural divide and regional tendencies, notable exceptions illuminated the complexity of the electorate. These exceptions often hinged on factors like changing demographics, local economic conditions, and unique historical and cultural influences [LOCAL ECONOMIC CONDITIONS WERE NOT MENTIONED IN THE ESSAY], illustrating the rich tapestry of American democracy. Understanding these exceptions is crucial for a comprehensive analysis of the political landscape in the United States.

More Tests for ChatGPT

Article-Grid / By Martin W. Lewis

The previous GeoCurrents post on ChatGPT was perhaps unfair to the much-celebrated – and much-maligned – chatbot, as it focused on recent meteorological data that are evidently not included in its prodigious store of information. I therefore decided to see how it would perform on two other topics recently explored in the blog. (See the bottom of this post for the two ChatGPT essays.)

The first assessment is based on my post on the geography of Indian chess grandmasters. To give ChatGPT its best chance, I asked it to write essay designed for an academic audience. Although the results are reasonably good, the chatbot still misses the essential point. Although it claims that “While southern and western India have dominated the chess scene, the northern regions of the country have not lagged behind.” But as the map used in my original post (reposted below) shows, northern India – especially if one excludes West Bengal – has definitely “lagged behind.” The ChatGPT essay also fails to consider possible reasons why Chennai has produced so many grandmasters.

The second ChatGPT assessment is on the renewal of Taliban assaults on Afghanistan’s Hazara Shia community. Again, I asked for an essay aimed at an academic audience. This request is evident in the chatbot’s useless subtitle: “a complex analysis” (as opposed to a simplistic one?). Overall, however, ChatGPT did a reasonably good job of outlining the general context. But it gave no specific information on either recent or historical events. It produced the kind of essay that a well-informed person could have written without doing any additional reading. Overall, this is the kind of work what one might expect from an intelligent but lazy student who is unwilling to look beyond Wikipedia.

When it comes to somewhat obscure current events, this tiny sample makes it seem as if ChatGPT is not yet capable of producing a solid essay. It is also noteworthy that it gave a separate heading for each paragraph, something that an essay written for an academic audience would never do. I also find the chatbot’s trite moralizing somewhat off-putting. Although I do not disagree with the final two sentences of its essay on the Hazaras, I do not find them helpful.

Prompt: Write a 300-Word essay on the geography of Indian chess grandmasters for an academic audience

ChatGPT Response

Title: The Geography of Indian Chess Grandmasters

Introduction

India, a land renowned for its diverse culture and rich history, has emerged as a powerhouse in the world of chess over the past few decades. This remarkable transformation is characterized not only by the country’s success in international chess tournaments but also by the geographical diversity of its chess grandmasters. In this essay, we explore the geography of Indian chess grandmasters, highlighting the regions that have produced these exceptional chess players.

The Rise of Indian Chess

The roots of chess in India can be traced back centuries, with the game having been known as “chaturanga” in ancient Indian texts. However, it was in the latter half of the 20th century that chess gained prominence in India, largely thanks to the efforts of players like Viswanathan Anand. Anand, who hails from the southern state of Tamil Nadu, became India’s first grandmaster in 1988. His success inspired a generation of chess enthusiasts across the country.

Southern Dominance

The southern region of India, particularly Tamil Nadu and Andhra Pradesh, has been a fertile breeding ground for chess talent. Apart from Anand, other notable grandmasters like P. Harikrishna and Krishnan Sasikiran hail from this region. The city of Chennai, in particular, has played a pivotal role in nurturing chess talent and has produced numerous International Masters and Grandmasters.

Western Pioneers

Moving towards the western part of India, we find several chess players who have made their mark on the international stage. The state of Gujarat, for instance, has produced talents like Pentala Harikrishna and Tejas Bakre. These players have consistently performed well in national and international tournaments, contributing to India’s growing reputation in the world of chess.

Northern Aspirants

While southern and western India have dominated the chess scene, the northern regions of the country have not lagged behind. Delhi, the capital city, boasts notable chess players like Parimarjan Negi and Tania Sachdev. Their accomplishments highlight the geographical diversity of Indian chess grandmasters, dispelling the notion that chess talent is confined to specific regions.

Conclusion

The geography of Indian chess grandmasters is a testament to the widespread passion for the game across the country. From the southern states to the western and northern regions, India has produced a remarkable array of chess talent, with players making their mark on the international stage. This geographical diversity not only enriches the nation’s chess heritage but also underscores the inclusive and universal appeal of this ancient game in the modern Indian context.

Prompt: Write a 500-word essay on the Taliban’s Renewed Assaults on Afghanistan’s Hazara Shia Community for an academic audience

ChatGPT Response

Title: The Taliban’s Renewed Assaults on Afghanistan’s Hazara Shia Community: A Complex Analysis

Introduction

The Hazara Shia community in Afghanistan has faced a long history of persecution, discrimination, and violence. However, the return to power of the Taliban in 2021 has heightened concerns about their safety and security. This essay delves into the renewed assaults on the Hazara Shia community by the Taliban, examining the historical context, religious dynamics, and geopolitical factors that contribute to this ongoing crisis.

Historical Context

The Hazara Shia community, primarily residing in central Afghanistan, has been marginalized for centuries. Their distinct Mongol ancestry and adherence to the Shia branch of Islam have made them a target of discrimination and violence, particularly under various Afghan regimes. The Taliban, a predominantly Sunni Islamist group, has shown a historical animosity towards Shia Muslims, considering them heretics. The renewed assaults on the Hazara Shia community are rooted in this deep-seated prejudice.

Religious Dynamics

Religious differences have played a significant role in the persecution of the Hazara Shia community. The Taliban’s strict interpretation of Sunni Islam has led to their intolerance of Shia practices and beliefs. They perceive Shia Muslims, including the Hazara, as deviating from the “true” path of Islam. This religious dimension exacerbates the violence against the Hazara Shia, as the Taliban seeks to impose their interpretation of Islam on the entire country.

Geopolitical Factors

The geopolitical landscape of Afghanistan has further complicated the situation for the Hazara Shia community. Afghanistan’s strategic location has made it a battleground for regional and international powers, and the Taliban’s resurgence has been supported by neighboring countries such as Pakistan. These external influences have emboldened the Taliban and hindered international efforts to protect vulnerable communities like the Hazara Shia.

Human Rights Violations

The renewed assaults on the Hazara Shia community have been marked by severe human rights violations. Reports of targeted killings, abductions, and forced conversions have emerged since the Taliban’s return to power. The Taliban’s ruthless tactics have created a climate of fear and insecurity among the Hazara Shia, causing many to flee their homes in search of safety.

Impact on Civil Society

The targeting of the Hazara Shia community has had a chilling effect on civil society in Afghanistan. It has eroded trust in the Taliban’s claims of inclusivity and has created a sense of vulnerability among other minority groups and dissenting voices. The fear of persecution has stifled activism and free expression, further diminishing the prospects for a democratic and inclusive Afghanistan.

International Response

The international community has expressed concern over the assaults on the Hazara Shia community, with calls for accountability and protection. However, the effectiveness of these responses has been limited due to the Taliban’s control of the country and their disregard for international norms and conventions. Humanitarian organizations face significant challenges in providing aid and protection to the Hazara Shia community in such a hostile environment.

Conclusion

The renewed assaults on Afghanistan’s Hazara Shia community by the Taliban are deeply troubling and emblematic of the complex challenges facing the country. These assaults stem from historical discrimination, religious differences, and geopolitical factors, and have resulted in egregious human rights violations. It is imperative for the international community to continue advocating for the rights and safety of the Hazara Shia and all marginalized groups in Afghanistan. Addressing the root causes of this persecution will require a concerted effort to promote tolerance, inclusivity, and respect for diversity in Afghanistan’s future governance and society.

Assessing ChatGPT Based on Its Ability to Write a GeoCurrents Essay. Final Grade: D-

Article-Grid, Geographical Thought / By Martin W. Lewis

Over the past few months, ChatGPT and other forms of artificial intelligence (AI) have become something of an obsession in academic circles. Some of my colleagues are appalled by these development while others are thrilled, but no one is uninterested. Some professors think that AI will make it impossible to give take-home exams and will compromise student research papers and critical essays. But one colleague responded by quipping that at least we will no longer have to read incoherent papers. That might be true, but there is no guarantee that ChatGPT-assisted essays will not be trite and filled with misinformation.

With such controversies in mind, I decided to test ChatGPT against my most recent GeoCurrents post, which examined summer 2023 precipitation patterns in the United States. I tried several prompts of lesser and greater specificity. Two of the more representative ones are posted below, with my appraisals included [IN BRACKETS AND IN ALL CAPS]. What I received from ChatGPT was, to put it bluntly, rubbish. Evidently I do not have to worry that my blogging efforts can be outsourced any time soon to artificial intelligence (or artificial idiocy, as the case may be).

My Prompt: “Write a 700-word essay on precipitation patterns in the United States in the summer of 2023”

ChatGPT Results

Title: Summer 2023 Precipitation Patterns in the United States: A Weather Roller Coaster

Introduction

The summer of 2023 in the United States has been marked by unpredictable and extreme precipitation patterns [TRITE]. From scorching droughts in the West to relentless rainfall in the East [LARGELY UNTRUE], this summer has been a roller coaster of weather events. In this article, we will delve into the complex factors driving these precipitation patterns and their impacts on different regions of the country.

The West: Drought Persists

Summer 2023 began with a continuation of the severe drought that has plagued the western United States for several years. The states of California, Nevada, and Arizona, in particular, experienced another season of water scarcity [FALSE]. Low snowpack in the Sierra Nevada Mountains and reduced reservoir levels exacerbated the situation [ONLY OFF BY 180 DEGRESS].

The primary drivers of this prolonged drought include a lack of winter precipitation [NO], higher temperatures leading to increased evaporation [NOT IN CALIFORNIA], and climate change-induced alterations in atmospheric circulation patterns. The West’s arid conditions have not only strained water resources but have also contributed to an increased risk of wildfires, as dry vegetation becomes more susceptible to ignition [NOT IN THE SUMMER OF 2023].

Despite some scattered showers and thunderstorms [THAT HAPPENS EVERY YEAR], many parts of the West faced water restrictions, mandatory conservation measures, and heightened concerns about future water availability.

The Midwest: Abundance of Rain

In stark contrast to the parched conditions in the West, the Midwest experienced an unusual abundance of rainfall in the summer of 2023. States like Iowa, Illinois, and Missouri witnessed frequent and heavy downpours, causing localized flooding and waterlogged fields. [IOWA WAS VERY DRY]

The primary factor behind this excessive rainfall was a persistent low-pressure system that remained nearly stationary over the region. This system acted as a moisture conveyor belt, pulling in warm, humid air from the Gulf of Mexico and generating frequent thunderstorms [NOT THIS YEAR].

While the surplus of moisture had negative consequences for agriculture, leading to waterlogged crops and delayed planting schedules [NOT IN MOST AREAS], it also had a silver lining. It helped alleviate concerns about drought and replenished aquifers, ensuring a more secure water supply for the coming months.

The South: A Tale of Contrasting Precipitation

The southern United States had its own share of precipitation variability in the summer of 2023. States along the Gulf Coast, such as Texas, Louisiana, and Mississippi, experienced a mix of heavy rain events [MOST OF TEXAS AND LOUISIANA WERE VERY DRY] and periods of drought.

The Gulf of Mexico played a central role in driving these weather patterns. Warm sea surface temperatures contributed to increased moisture in the atmosphere, resulting in heavy rainfall and occasional flooding in coastal areas [NOT SO MUCH THIS YEAR]. Conversely, the same warm waters fueled the development of tropical storms and hurricanes, which could bring substantial precipitation but also destructive winds and storm surges [THIS SENTENCE REFERS TO GENERAL CONDITIONS, NOT THOSE OF 2023].

In the interior South, areas farther away from the Gulf Coast often experienced drier conditions and occasional heatwaves [DUH]. These fluctuations created challenges for farmers, who had to navigate between excessive moisture and parched fields, making it a season of mixed blessings .

The Northeast: Unpredictable Rainfall

The northeastern United States, including New York, Pennsylvania, and New England, witnessed a summer characterized by unpredictable rainfall patterns. While some areas enjoyed near-normal precipitation, others faced extended dry spells punctuated by sudden, heavy rain events [OVERALL, THIS AREA RECEIVED MUCH HIGHER THAN NORMAL PRECIPITATION].

One of the driving factors behind this variability was the position of the jet stream. Its meandering path influenced the movement of weather systems, leading to localized variations in rainfall. Areas under the jet stream’s influence saw more frequent and intense rain events, while those outside of it experienced drier conditions [REFERS TO GENERAL CLIMATOLOGY, NOT CONDITIONS IN 2023].

This inconsistency in precipitation had consequences for agriculture, water management, and recreational activities [THIS IS TRUE EVERY YEAR]. Farmers struggled to predict planting and harvesting schedules, while water resource managers had to adapt quickly to shifting water supply conditions.

Conclusion

The summer of 2023 brought a diverse range of precipitation patterns to the United States. While the West continued to grapple with drought and the associated challenges, the Midwest saw an abundance of rain that alleviated drought concerns but presented its own set of problems. The South experienced a mix of heavy rain and drought, depending on proximity to the Gulf Coast, while the Northeast faced unpredictable rainfall [ALREADY STATED].

These diverse patterns highlight the complex interplay of atmospheric and climatic factors that influence regional weather [TRITE]. In a world affected by climate change, extreme weather events are becoming increasingly common, emphasizing the need for adaptive strategies in agriculture, water resource management, and disaster preparedness [BOILERPLATE].

As we move forward, it is crucial to monitor and understand these changing precipitation patterns and their impacts on various sectors of society. Building resilience and implementing sustainable water management practices will be essential to mitigate the challenges posed by future weather extremes, ensuring a more stable and secure future for the United States. [I WONDER HOW OFTEN THESE TWO SENTENCES ARE REPRODUCED BY CHATGPT FOR WEATHER-RELATED TOPICS]

ESSAY GRADE: D-

My More Specific Prompt: “Write a 700-word essay on 2023 summer precipitation patterns in the United States, paying particular attention to the U.S. corn crop”

ChatGPT Results

Title: Analyzing 2023 Summer Precipitation Patterns in the United States: Implications for the U.S. Corn Crop

Introduction

Summer precipitation patterns play a crucial role in shaping agricultural outcomes, especially in the United States, where agriculture is a significant contributor to the economy [UTTERLY TRITE]. The U.S. corn crop, in particular, is highly sensitive to variations in precipitation, as it is a staple crop used in various industries, from food production to biofuel [EQUALLY TRITE]. In this essay, we will examine the 2023 summer precipitation patterns in the United States and their implications for the U.S. corn crop.

The Importance of the U.S. Corn Crop

Before delving into the specifics of summer precipitation patterns, it is essential to understand the significance of the U.S. corn crop [NO, IT IS NOT]. Corn, or maize, is one of the most widely cultivated crops in the United States, with millions of acres dedicated to its growth. It is a versatile crop, serving as a food source for humans and livestock, a raw material for various industrial processes, and a critical component of the biofuel industry. As such, fluctuations in corn production can have far-reaching economic and societal consequences.

2023 Summer Precipitation Patterns

To assess the impact of summer precipitation patterns on the U.S. corn crop in 2023, we must first examine the climatic conditions during that period. In recent years, climate change has led to increased variability in weather patterns, making it crucial to monitor and analyze the effects of these changes on agriculture.

The summer of 2023 witnessed noticeable variations in precipitation patterns across the United States. While the Western and Southwestern regions experienced extended drought conditions, the Central and Eastern parts of the country saw above-average rainfall [POOR GENERALIZATION]. These divergent patterns had significant implications for corn cultivation.

III. Impact on Corn Production

Drought in the West and Southwest

The Western and Southwestern regions of the United States, including states like California, Arizona, and New Mexico, experienced severe drought conditions during the summer of 2023 [NO]. These regions are not traditionally major corn-producing areas but do contribute to overall national production. However, the prolonged drought significantly impacted corn yields in these states, leading to decreased production. Farmers in these regions faced challenges related to water scarcity, increased irrigation costs, and reduced crop quality.

Above-Average Rainfall in the Central and Eastern U.S.

Conversely, the Central and Eastern regions of the United States, including states like Illinois, Iowa [NO], and Ohio, received above-average rainfall during the summer of 2023. While adequate rainfall is generally beneficial for crop growth, excessive precipitation can lead to waterlogging and crop diseases [TRITE]. Corn, in particular, is susceptible to water stress and fungal infections when exposed to prolonged periods of high moisture [WATER STRESS IN PLANTS IS CAUSED BY A DEFICIENCY OF WATER].

Adaptation and Resilience

In the face of increasingly erratic climate patterns, farmers and agricultural scientists have been working to adapt and develop more resilient farming practices. These efforts include the use of drought-tolerant corn varieties, improved irrigation techniques, and enhanced soil management practices.

In regions affected by drought, farmers have adopted more water-efficient irrigation methods, such as drip irrigation [DRIP IRRIGATION IS NOT A SIGNIFICANT FACTOR IN U.S. CORN PRODUCTION], to conserve water resources. Additionally, some have shifted towards drought-tolerant corn varieties that can withstand water scarcity to a certain extent [COULD YOU BE A LITTLE MORE SPECIFIC?].

In areas with excess rainfall, farmers have focused on improving drainage systems and selecting corn varieties resistant to waterlogging and fungal diseases. These adaptations aim to mitigate the negative impact of excessive moisture on crop yields [OVERLY GENERAL].

Government Policies and Support

The U.S. government plays a crucial role in supporting farmers and managing the impact of climate-related challenges on agriculture. In response to the 2023 summer precipitation patterns, the government implemented various policies and programs to assist farmers affected by drought or excess rainfall . These measures included financial assistance, crop insurance programs, and research initiatives aimed at developing climate-resilient agricultural practices [NONE OF THESE INITIATIVES WERE IMPLEMENTED IN REPONSE TO 2023 SUMMER PRECIPITATION PATTERNS] .

Conclusion

The 2023 summer precipitation patterns in the United States had a significant impact on the U.S. corn crop [BEYOND TRITE]. While drought in the Western and Southwestern regions led to decreased yields and increased production costs, above-average rainfall in the Central and Eastern states presented challenges related to waterlogging and crop diseases [GROSSLY OVERGENERALIZED]. Adaptation and resilience-building efforts have become essential for farmers, with the support of government policies and programs.

As climate change continues to alter weather patterns, it is crucial for the agricultural sector to develop sustainable and climate-resilient practices. The U.S. corn crop, as a linchpin of the nation’s agriculture and economy [WHILE VERY IMPORTANT, THE CORN CROP IS HARDLY THE LINCHPIN OF THE NATION’S ECONOMY], will continue to face challenges and opportunities in the years to come, making it imperative for stakeholders to adapt and innovate in response to evolving climate conditions.

ESSAY GRADE: D-

As GeoCurrents posts are always accompanied my maps, I asked ChatGPT what it could do in the regard. The result is posted below. AsChatGPT evidently has no access to recent climatological date, this was perhaps an unfair test. My next post will therefore assess the chatbot’s ability to write a different kind to essay.

The Very Wet and Very Dry Summer of 2023 in the United States (and Its Consequences for the U.S. Corn Crop)

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

The 2023 meteorological summer (June-August) in the United States was both very dry and very wet, with extreme precipitation variation. As the map posted below shows, this was the driest summer since (at least) 1895 over several widely scattered parts of the country, including western Oregon, southern Arizona, western New Mexico, southern Louisiana, southeastern Texas, and the southeastern corner of Minnesota. The map also shows a small area near California’s Bay Area as having experienced its driest summer since 1895, but this particular depiction is misleading. Like most of the rest of the state, this region experiences negligible summer precipitation, getting a little over a tenth of an inch on average (see the table posted below for Antioch, CA). The difference between 0.01 inches and 0.15 inches may be large in percentage terms, but it means next to nothing in regard to actual conditions.

The 2023 summer drought in southern Minnesota, Iowa, and adjacent areas is concerning, as this area forms the heart of the U.S. corn belt. The official Drought Monitor Map shows severe to extreme drought over much of the same area. Curiously, this map also shows extreme long-term drought in parts of Kansas and Nebraska that experienced well above average summer precipitation this year. Some critics complain that the U.S. Drought Monitor is too slow to revise its mapping as conditions change.

Given such dry conditions over much of the corn belt, one might expect a reduced U.S. corn harvest in 2023. But according to an August 8 Reuters headline, “U.S. Farmers Expect Corn Harvest Could Be Second-Biggest Ever,” with the article explaining that “rains during July shepherded the crop through its critical development phase, offsetting dry conditions early in the season and hot summer temperatures.” A Successful Farming article outlines the problems faced by Minnesota farmers due to a three-year drought, while noting that technical improvements have mitigated the damage:

Bob Worth, president of the Minnesota Soybean Growers Association and a corn and soybean farmer in Lincoln County, Minnesota, said that improvements in seed technology is saving his farm from complete disaster despite the dry conditions: “If we had the same hybrids that we had in my early farming years,” Worth said, “we wouldn’t have a crop with as little rain as we’ve gotten.”

Despite its relatively healthy 2023 corn crop, the United States just lost its position as the world’s top corn exporter. The title was handed over to Brazil, as was that for soybeans a few years ago. According to a recent Bloomberg report (cited in Farm Policy News):

For more than half a century, US farmers dominated the international market for corn, shipping more of the critical crop than any other country to feed the world’s livestock, fill its stockpiles and manufacture its processed foods. No more. In the agricultural year ending Aug. 31, the US handed the corn-exporting crown to Brazil. And it might never get it back.

Other parts of the United States saw record-breaking precipitation in the summer of 2023. As the first map in this post shows, much of northern New England, eastern Michigan, southern California, central Wyoming, eastern Colorado, and south-central Montana experienced their wettest summer since 1985. Tropical Storm Hilary was responsible for most of this precipitation in southern California, and for some of it in central Wyoming and south-central Montana.

If the summer 2023 precipitation map had included the first three days of September, its depiction of the southwestern quadrant of the United States would have been quite different, showing much higher figures for eastern Arizona, Nevada, and parts of northern California (see the map below). A rare early-September cut-off low-pressure system rotated off the cost of far northern California during this period, bring unprecedented precipitation to many areas, as well as widely mocked misery to campers at the Burning Man Festival in northwestern Nevada.

Note on this map the extraordinarily steep gradient on the seven-day late-August to early-September rainfall map in northern California. Here areas getting less than five percent of average precipitation in this period are almost adjacent to those getting over 600 percent. This seeming anomaly was partly the result of a sharp precipitation cut-off in the recent storm, with entrained bands of rainfall hitting some areas repeatedly while leaving nearby areas completely dry. But it is also, again, a consequence of the extremely dry average conditions in this region at this time of the year.

The Geography Indian Chess Grandmasters

Article-Grid, Education / By Martin W. Lewis

The prolific blogger, economist, and public intellectual Tyler Cowen recently reposted a map on the geography of Indian chess grandmasters (reproduced below). The map was originally posted on X [Twitter] by “The_Equationist” under the heading “This chart speaks for itself.” I am not sure that it does, nor do several of Cowen’s commenters. But for those who understand the basic spatial contours of socio-economic development in India, the map does show a familiar pattern. Almost all Indian grandmasters come from the more economically vibrant and educationally developed states of southern and south-central India. Conversely, few come from the densely populated and economically lagging Ganges Valley, India’s historical core, or indeed from any other northern part of the country. Based on socioeconomic development alone, one might expect a few more grandmasters from Punjab, Haryana, and Delhi, but the pattern remains unexceptional.

What is especially striking is that three states in particular stand out for producing grandmasters: West Bengal (labeled here simply as “Bengal”), Maharashtra, and Tamil Nadu. But again, for those familiar with the geography of Indian intellectual life, this pattern is not particularly surprising. West Bengal has long been famed as India’s intellectual center, particularly in fields in the humanities. This feature of Indian geography has occasionally hit home. I few years ago, I was peripherally involved in a search at Stanford University for a new faculty member specializing in the history of modern South Asia. Well into the process, a South Asian member of the search committee asked whether the rest of us had noticed that all the top candidates were Bengali, hailing from West Bengal. I had not noticed.

Tamil Nadu, which really stands out on the map, is also famous for the intellectual achievements of its inhabitants, which veer in a more scientific and technical direction. Such intellectuality is particularly notable among the Tamil Brahmins (or “Tam Brahms”), who have been subjected to some reverse discrimination in their homeland and are thus well represented abroad. Maharashtra does not have the same intellectual reputation as the other two states, but it is, in many regards, India’s economic standout and its center of popular-culture production.

One of Cowen’s commenters, Sathish, mentions that almost all of Tamil Nadu’s grandmasters come from its largest city, Chennai (formerly Madras), noting as well that 15 of the 29 studied at the same school. Another commenter, sxb, provides the necessary background information:

It started in the 50s with the first Indian international Master, Manuel Aaron. He started the Tal chess club in Chennai, brought in Soviet chess books and even a Soviet Grandmaster for a month with more visiting and Anand, the Indian prodigy, was his pupil. Basically all the local players worthy of note went through that chess club. It blossomed from there. Anand was born and lives in Chennai and has been a great role model for chess in India and got many youngsters interested in the game.

As geographers have long noted, place matters. So does personality, as this story so well demonstrates.

The Taliban’s Renewed Assaults on Afghanistan’s Hazara Shia Community

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

I was surprised to recently read that the Taliban are trying to marginalize the Persian language in Afghanistan, given its near majority status, stature, and role in inter-ethnic communication (see the previous GeoCurrents post). On resuming power in 2021, moreover, the Taliban had promised to pursue less brutal and divisive policies. In their first stint (1996-2001), they had viciously attacked the country’s minority Shia population, mostly found among the Hazara people in central Afghanistan. Some observers viewed these campaigns as almost genocidal (see this earlier GeoCurrents post). But as Radio Free Europe framed the Taliban’s new attitude:

After regaining power, the Sunni militant group tried to assuage Hazaras’ fears of discrimination and persecution. The Taliban visited Shi’a mosques in the Afghan capital and deployed its fighters to protect ceremonies marking the Shi’ite month of Muharram.

Given the Taliban’s previous animosity toward Shia Islam, I had expected that that any reversal of its newly formulated toleration program would be directed against the Hazaras, and perhaps also the smaller non-Hazara Shia communities in western Afghanistan. Yet I had encountered nothing of the sort. It turns out, however, that my reading on this subject had been far too limited. The July 17, 2023 Radio Free Europe article cited above went on to note that:

Last week, the Taliban prevented Shi’a from celebrating an important religious festival. The militants have also restricted the teaching of Shi’a jurisprudence in universities in Afghanistan. In February, the Taliban reportedly banned marriages between Shi’a and Sunnis in northeastern Badakhshan Province.

The Shi’ite community has accused the Taliban of failing to prevent deadly attacks on Hazaras by the rival Islamic State-Khorasan (IS-K) extremist group. Meanwhile, rights groups have documented the forced evictions of Hazaras by the Taliban, a predominately Pashtun group, in several provinces.

Other reports are even more worrisome. As noted by Jurist, “The Asian Forum for Human Rights and Development (FORUM-ASIA) issued a statement on Friday calling for an end to the systemic killing of Shia Hazaras in Afghanistan in order to prevent a possible genocide under Taliban rule.” According to Agnès Callamard, Amnesty International’s Secretary General, “These violent deaths are further shocking proof that the Taliban continue to persecute, torture and extra judicially execute Hazara people.”

Also shocking is the lack of reporting on this situation in the American media. Afghanistan’s Shia population is massive, constituting something between 9 and 29.5 percent of the country’s population of 40 million (see the two highly divergent pie charts below). Continuing attacks on these people could quickly generate a humanitarian crisis of the first order.

Although most Shia Muslims in Afghanistan follow the Imami, or Twelver, majority sect of the faith, a significant minority adhere instead to the minority Ismaili sect. Most Afghan Ismailis are also ethnic Hazaras, but a few are ethnic Tajiks (both of which are Persian-speaking peoples). One might expect that the Taliban would be especially hostile to the Ismailis, given their heterodox, esoteric, and cosmopolitan orientation. Scant information, however, is readily available on this group. But according to one prominent 2001 report:

Ismailis in Afghanistan are generally regarded with suspicion by other ethnic groups and for the most part their economic status is very poor. Although Ismaili in other areas such as the northern areas of Pakistan operate well-organized social welfare programs including schools, hospitals and cooperatives, little has been done among Afghan Ismaili communities. Considered less zealous than other Afghan Muslims, Ismaili are seen to follow their leaders uncritically.

The news searches that I conducted for information on Ismailis in Afghanistan mostly returned articles about a recent deadly knife attack by an Afghan refugee on women in an Ismaili center in Lisbon, Portugal. A Shia Wave article, however, notes that the Taliban are trying to convert Afghan Ismailis to Sunni Islam, and evidently with some success. I was disappointed to find no information on Afghan Ismailis on the website of the Aga Khan Foundation, the well-funded and highly effective humanitarian wing of the global (Nizari) Ismaili community. The Foundation does highlight its extensive humanitarian work in Afghanistan, but sidesteps the country’s sectarian divisions.

Afghanistan’s Language Policies and the Cartographic Marginalization of Persian (Farsi)

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

A fascinating recent article in The Diplomat (“Decoding the Taliban’s Anti-Persianism” by Javeed Ahwar) outlines how the new government of Afghanistan is attempting to sideline the country’s dominant Persian language in favor of Pashto, the main language of its Taliban rulers. As Ahwar writes, “One of the first things that Taliban did after taking over in August 2021 was to remove Persian from public billboards,” noting that the new policy “came at the cost of marginalization of the Persian-speaking population and undermining their rights to political participation.” As he further argues, the drive against Persian is rooted in a quest for Pashtun political dominance and reflects the overriding influence of the harsh South Asian Deobandi school of Sunni Islam, which favors Urdu over Persian. It also stems from the desire of the Taliban to disconnect the people of Afghanistan from the extraordinary riches of the Persian literary tradition, much of which is too cosmopolitan and secular for their taste.

Although the word “Afghanistan” literally means “land of the Pashtuns” – the Pashto-speaking people – Persian is still the country’s first language. As can be seen in the table posted above, roughly half the people of Afghanistan speak Persian as their mother tongue, with more than three quarters speaking it as either their first or second language; the comparable figures for Pashto are 40 and 48 percent. Persian is clearly the country’s main language of interethnic communication.

The importance of Persian in Afghanistan, however, is commonly obscured both in written depictions and especially on maps. Consider, for example, the Wikipedia map “Languages of Afghanistan,” posted below. Although it purports to be a language map, it is actually a map of ethnic groups that are only partly based on linguistic affiliation. Importantly, the term “Persian” does not appear on the map. Instead, the Persian-speaking population of Afghanistan is divided into three ethnic groups: Hazara, Aimak, and Tadjik [Tajik]. Although minor dialectal differences separate these groups, they are mostly divided by religion (the Hazaras being Shia) and traditional lifestyle (the Aimaks being pastoralists). The map also minimizes the extent of the Tajik people, especially in western Afghanistan. The second map rectifies these problems, but still avoids the term “Persian” in favor of “Dari,” following the long-established usage of the Afghan government. But Dari is merely the main variant of Persian spoken in Afghanistan, and in written form is essentially identical with Farsi, as Persian is labeled in Iran. The terms Persian and Farsi denote the same language and are themselves variants of the same word, “Persian” being “Farsi” as translated into Greek; the term “Dari,” on the other hand, is generally thought to mean “court,” implying the “language of the court.” Occasionally one finds a map that portrays most of northern Afghanistan as Farsi-speaking, such as the Deviant Art map by H.G.[TRKM_JOSE] posted below. Such depictions, however, are relatively rare in scholarly and journalistic sources. Finally, some maps, like the last one posted below, do show northern Afghanistan as Persian speaking.

According to the erudite scholar Nile Green, current holder of the Ibn Khaldun Endowed Chair in World History at UCLA, “the impulses behind renaming of Afghan Persian as Dari were more nationalistic than linguistic.” The Wikipedia article on Dari summarizes Green’s argument as follows:

Afghanistan’s Persian-speaking population still prefer to call their language “Farsi”, asserting that the term “Dari” has been imposed upon them by the dominant Pashtun ethnic group as an effort to detach Afghanistan from its deep-rooted cultural, linguistic, and historical connections with the wider Persian-speaking world, encompassing Iran, Tajikistan and parts of Uzbekistan.

The broader context and consequences of Afghanistan’s anti-Persian/Farsi policies will be considered in the next GeoCurrents post

India’s Kota Factory: Producing Enhanced University Entrance Exam Scores

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

To an American viewer, the premise of the Indian television show Kota Factory (available on Netflix) might seem absurd. The show is set in a city of over a million inhabitants that is economically based on a single “industry,” that of tutoring teenagers preparing for college entrance exams. Yet the city of Kota, located in India’s northwestern state of Rajasthan, is real, as is its focus on exam preparation. More than 150,000 students are reported to come annually to this city, most of them staying for two or three years. Dozens of exam-coaching companies – private schools, essentially – compete for students, most of whom live in cramped hostels. The aim of most students is admission into the Indian Institutes of Technology (IIT), a highly prestigious 23-campus institution with branches scattered across the country.

Kota Factory is shot in black-and-white, with camera angles and other techniques reminiscent of those of an art film. It is best described as a serious sitcom. According to the Wikipedia article on the show, its creator “aims to change the popular narrative surrounding Kota and preparation for [entrance exams] in Indian pop culture to a more positive one via the show.” Despite the popularity of studying in Kota, the city’s educational industry has acquired a somewhat negative perception due partly to a string of well-publicized suicides linked to its extraordinarily competitive atmposphere.

I find this show particularly interesting for the light that it sheds on the differences between the Indian and American educational systems. In Kota, a successful teacher – meaning one whose students have a relatively high acceptance rate at IIT – can have something of a celebrity status. The show’s most interesting character is a charismatic physics lecturer who is harsh but caring, deeply involved in his students lives. He is precisely the kind of mentor who can successfully shepherd shell-shocked 16-and 17-year-old pupils through the grueling process. As noted by the Wikipedia article on the show, this character is played by “Jitendra Kumar as Jitendra Kumar a.k.a. Jeetu Bhaiya: a fictionalised version of himself.” (Kumar was a civil engineering student at IIT Kharagpur before going into acting.)

Kota’s rise as an educational coaching center began in 1980s, when it was an industrial town producing, among other goods, polished building stone and synthetic fabric. At the time, however, many of its local industries were failing, a process that accelerated after the Indian economy opened to global forces in 1991. Adversity, however, generated some creative responses. As explained by Neelam Gupta in a 2016 Governance Now article:

The story begins with Vinod Kumar Bansal, who was once an employee of JK Synthetics. … As he had good mathematics skills, children from the neighbourhood flocked to him for help. One day he was diagnosed with muscular dystrophy, a condition that rapidly weakens one’s muscles and bones. This disease restricted his movement and he could no longer work in the factory. Fearing he might lose his job, Bansal started giving maths tuitions at his home. In 1986, one of his students cracked the IIT joint entrance exam (IIT-JEE). Next year, some 100 students had joined his classes and by 1998, Bansal was coaching more than 1,000 students. Soon he built a big building in the city’s industrial area and set up Bansal Classes. In 2000, one of his students topped and 300 others cracked the IIT-JEE. It created a stir. “There was a long queue outside my study centre and I finally selected 18,000 students that session,” he recalls.

As often happens, Bansal’s success generated spin-off companies founded by his own employees. Students increasingly flocked to Kota partly because few Indian secondary schools offer adequate preparation for the demanding higher-education entrance exams. It is perhaps not coincidental that the state of Rajasthan, in which Kota is located, has one of India’s worst educational systems (see the literacy map below). Another draw is social. Many students find that it is much easier to devote themselves to intensive study in a city where tens of thousands of people their own age were doing the same than they would at home, as the show itself emphasizes. By 2010, 40,000 students a year were coming from all over northern India to study in Kota. Enrollment jumped a few years later when some key entrance exams changed from a state-level to a national basis. The annual figure, as noted above, is now around 150,000.

Not surprisingly, other Indian cities are hoping to profit from the educational coaching boom. A 2018 Hindustan Times article features the headline, “Why Are There ‘Education Malls’ in Ranchi? Inside the New Kotas of India.” Ranchi, like most rising coaching centers, and Kota itself, is a sizable but secondary city located in India’s poorer and educationally lagging north-central belt (see the map posted below). An exception is Madurai in the south, which has long been noted as a major university city and as the “cultural capital” of the state of Tamil Nadu. As Wikipedia notes, “Madurai has been an academic centre of learning for Tamil culture, literature, art, music and dance for centuries.”

As the creators of Kota Factory know what they are writing about, one can even learn a little science from watching the show. Episode Three, which focuses on inorganic chemistry and the many exceptions to its rules, including exceptions to the exceptions, is the prime case in point. The prolonged diatribe of Mayur More (playing the student Vaibhav Pandey) about the frustrations of studying the subject is one of the most amusing and riveting bits of acting that I have ever seen – and by far the most intellectually informed.

Favorable Climatic Conditions and the Flourishing of the Tibetan Empire, 618-842 CE

Article-Grid, Historical Geography / By Martin W. Lewis

I have long been perplexed by the Tibetan Empire, which existed from 618 to roughly 842 CE. The Tibetan Plateau is a sparsely populated land with a challenging physical environment. Not surprisingly, for most of recorded history it has played a minor geopolitical role. Yet for more than 200 years, the Tibetan Empire was something of a superpower, fully competitive with the Tang Empire of China, the Muslim caliphates of the Middle East, and other great Eurasian states. As the maps posted below indicate, the Tibetan Empire’s territorial sway extended, at various time, well beyond the plateau itself. At its height it covered the Tarim Basin of northwestern China (Xinjiang), the Yunnan Plateau, and more.

The most extensive depiction of the Tibetan Empire that I have found was published in a recent Science Direct article (Hou et al., 2023). On this map, posted below, the Tibetan Empire in 820 CE is shown (with a dashed yellow line) as having included most of what is today Afghanistan, Uzbekistan, Tajikistan, and Kyrgyzstan, as well as much of Bangladesh and lowland north India. Most maps of the empire show it as having exercised only indirect power over a much smaller part of northern South Asia. Animated maps, moreover, indicate that its power over this area was brief. The map posted in Hou et al. thus seems to be exaggerated, as is so often the case in the mapping of empires.

Regardless of cartographic hyperbole, the article in question does offer an intriguing and largely convincing explanation for the rise and fall of the Tibetan Empire. As the authors’ investigation shows, between 600 and 800 CE the Tibetan Plateau experienced both relatively warm and wet conditions, allowing both agriculture and pastoralism to thrive. Such circumstances, in turn, provided the economic basis for a strong and expansionistic state. Importantly, this period was the only time in the past several thousand years that saw both warmth and “exceptional” humidity on the lofty plateau. Significantly, the collapse of the Tibetan Empire coincided with a return to drier conditions. As the authors write:

A closer examination at the precipitation records revealed that periods when Tibetans actively invaded Tang territory mostly coincided with humid periods, while conversely, Tang invasions into Tibetan territory occurred during relatively dry periods on the TP [Tibetan Plateau].

The peak of the drought at approximately 840 AD coincided with the collapse of the Tibetan Empire. Cold and dry events would be expected to lead to a decline in crop production, breaking the balance between population and resource supply.

Other major episodes of imperial expansion in Central Eurasia also seem to have been linked to favorable climatic conditions. A 2014 article, for example, found something similar in regard to the Mongol Empire. As the authors summarized their findings:

A 1,112-y tree-ring record of moisture shows that in opposition to conventional wisdom, the climate during the rise of the 13^th-century Mongol Empire was a period of persistent moisture, unprecedented in the last 1,000 y. This 15-y episode of persistent moisture likely led to a period of high grassland productivity, contributing fuel to the Mongol Empire.

A century ago, leading geographers argued that climate and other aspects of the physical environment determine the course of human history. Such overweening determinism weakened the discipline, contributing to its subsequent marginalization. Today, however, historians and physical scientists are reviving the study of the influence of climate – and of climate change – on the course of history, but are doing so in more modest, rigorous, and productive manner.

Environmental determinism reached its height in the early twentieth century in the works of Yale geographer Ellsworth Huntington. In 1907, Huntington argued in The Pulse of Asia that a long period of intensifying droughts beginning several thousand years ago forced Central Asian pastoral nomads out of their homelands, impelling them to conquer sedentary states located in more climatically favorable areas. As he put it (page 393-394):

If the rainfall fell from 20 inches to 10, the number of sheep would decrease from 60 to one. Manifestly, if such a change took place in the course of a few hundred years, most of the inhabitants would be obliged to migrate. As the nomads pressed outward from the dryer central regions of Asia, we can imagine how they were obliged to fight with the neighboring tribes whom they tried to dispossess. The older inhabitants and the newcomers could not all live together; new migrations would be a necessity; and the confusion would spread in every direction. Meanwhile, Europe, after this long period of blighting cold, was becoming warm and habitable, and the migrants pressed into it, horde after horde. No one tribe could stay long in its chosen abiding-place, for new bands of restless nomads pressed upon it. Rome fell before the wanderers. Nothing could stay their progress until the turn of the tide.

Per chance, though this is only vague conjecture, the legends of King Arthur and his Knights bear a hint of what might have occurred all over central Europe if it had not been for the influx of Barbarians. England, in its remote corner of Europe, far from the dry plains of Asia, responded at first to the influence of improved climatic conditions, until it, too, was reached by the migrating hordes of invaders. Meanwhile, in the most densely populated part of Asia, another movement of nations had begun, presumably because of the distress due to rapid desiccation. The Arab migrants carried with them the fanatical faith of Mohammed, and were by it inspired to remoter conquest.

Such conclusions are directly opposite of those now being advanced by scientifically informed historical studies. One could not, of course, expect Ellsworth Huntington to have had a modern conception of environmental and historical processes and conditions in 1907. Still, to have argued, if only as “vague conjecture,” that the mythical glories of Camelot might have been replicated over central Europe in the early Middle Ages, had it not been for the drought-driven barbarian invasions, is more than a stretch. Unfortunately, the excesses of Huntington’s determinism led geographers to largely abandon the idea that the physical environment has any significant influence on human history, much to their detriment. Today, scientific studies of the changing physical environment are clarifying our conception of the human past are helping to solve historical mysteries.

GeoCurrents Summer Break

Article-Grid, Site News / By Martin W. Lewis

Dear Readers,

Dues to travel and family considerations, GeoCurrents is currently on a summer hiatus. Blogging will resume in mid to late August.

Mismapping the Achaemenid Persian Empire and the Caucasus

Article-Grid, Cartography, Historical Geography / By Martin W. Lewis

As noted in a recent post, maps of empires tend to exaggerate their territorial extents, and the Achaemenid Persian Empire (550-330 BCE) is no exception. Most maps of this important empire depict it as covering all or almost all the South Caucasus region, with its border extending to the crest of the Greater Caucasus range (see the top maps from a Google image search posted below). Some show it as pushing even further to the north, encompassing the historically Circassian lands to the north and west of the Caucasus and sometimes even extending completely around the Black Sea (see below).

There is little if any good evidence, however, that the Achaemenid Persian Empire ever included the Kingdom of Colchis, located mainly in what is now the western half of the Republic of Georgia. The Wikipedia map of the early Georgian states posted below gives a much better depiction of the geopolitical situation of the time. The notion that this ancient Persian empire extended to the crest of the Greater Caucasus range derives essentially from a passage written by the ancient Greek scholar Herodotus. Although there is much to admire in the works of Herodotus, it has long been known that many of his assertions were far from accurate. It is for good reason that Lloyd Llewellyn Jones recently decided that it was necessary to write a book on the Achaemenid Empire based mostly on Persian sources, rather than on Herodotus and other Greek writers. But Jones, unfortunately, also maps western Georgia as having been under Persian control.

There is, however, some scholarly disagreement about which polity (or polities) had ultimate sovereignty over what is now western Georgia between 550 and 330 BCE. The Wikipedia article on the history of the Republic of Georgia provides an excellent summary:

Between 653 and 333 BC, both Colchis and Iberia survived successive invasions by the Iranian Median empire. The case is different for the Achaemenid Persians, however. According to Herodotus (3.97), Achaemenid power extended as far as the Caucasus mountains, but the Colchians are not included in his list of the twenty Persian satrapies. Nor are they referred to in the lists of Achaemenid lands (dahyāva) given in the Old Persian inscriptions of Darius and his successors. In Xenophon’s Anabasis (7.8.25; probably an interpolation) the tribes of Colchis and East Pontus are referred to as independent (autónomoi). On the other hand, Herodotus mentioned both the Colchians and various Pontic tribes in his catalogue (7.78-79) of approximately fifty-seven peoples who participated in Xerxes’ expedition against Greece in 481-80 BC. As the Encyclopaedia Iranica states, it is thus probable that the Achaemenids never succeeded in asserting effective rule over Colchis, though local tribal leaders seem to have acknowledged some kind of Persian suzerainty. The Encyclopaedia Iranica further states, whereas the adjoining Pontic tribes of the nineteenth satrapy and the Armenians of the thirteenth are mentioned as having paid tribute to Persia, the Colchians and their Caucasian neighbors are not; they had, however, undertaken to send gifts (100 boys and 100 girls) every five years (Herodotus 3.97).

The giving of gifts and the supplying of troops by a polity to a much more powerful neighboring empire, however, does not in itself indicate inclusion in that empire. It must also be noted that careful historical cartographers, such as Thomas Lessman, do not map western Georgia as having been part of the Achaemenid Persian Empire (see the map below).

The issue at stake here is not merely that of the inaccurate mapping of empires. What I am more concerned about is historical amnesia about the Caucasus, coupled with its pervasive historical misrepresentation. To put it simply, this key region of the world does not get its due in most historical and geographical accounts. All too often, it is simply appended to one or more empires based in other lands. Many such empires did covet the region, and in some periods they did control, directly or indirectly, large parts of it. But the Caucasus also had its own kingdoms and other polities, which deserve recognition.

I recently gave a keynote address about such issues at a conference on the Black Sea region held in Batumi in the Republic of Georgia. I hope to convert this talk to a video later this year; if I do so, I will post it on this website.

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

Author name: Martin W. Lewis