Chris Miller on US technology policy in the early/mid-2010s:
“American tech policy was held hostage to banalities about globalization that were easily seen to be false. . . . in polite company in Washington and Silicon Valley, it was easier to simply repeat words like multilateralism, globalization, and innovation, concepts that were too vacuous to offend anyone in a position of power” (p. 298)
In my last post, I explored historical questions with ChatGPT. This time I decided to try out more analytical questions. The news is distinctly less good for ChatGPT. The first question is about my book, which is recent and obscure, so not a lot of internet content to draw on, and ends up being pretty blah. The second is about a massive question with lots of material, but I asked for a more definitive argument than ChatGPT seems interested in. The third is asking for a kind of critical literature review focused on an influential book, and it doesn’t go well at all. This all may be improved in a year or two, but at present suggests a real limit to what ChatGPT can do.
That said, I remain overall very impressed with its abilities – instantly writing a coherent, decent-Wikipedia-level summary of a historical trend or comparison of (famous) authors is a big deal and will be very useful to a lot of people, in ways good and bad. I definitely need to start thinking hard about the most useful kinds of papers to assign in the AI era – banning ChatGPT or whatever isn’t going to happen, so the key will be to figure out where value-added can be generated from an assignment. For instance, asking very specific analytical questions about relatively recent research is not (at least thus far) a ChatGPT area of strength.
Question: What are the strengths and weaknesses of Paul Staniland’s 2021 book Ordering Violence?
Answer: “Paul Staniland’s 2021 book Ordering Violence: How the Ancients and Moderns Created State-sponsored Crime is a detailed and nuanced analysis of the ways in which states use violence to maintain order, both historically and in the present day. Here are some strengths and weaknesses of the book:
Strengths:
Weaknesses:
My comment: well, this isn’t so great. Some of the weaknesses are perfectly reasonable points – gender and economic issues definitely don’t receive as much attention as others, and it is a state-focused argument. But the list of strengths are basically pulled from blurbs and the book description, and not very specific, certainly not enough to use in a paper. In addition to getting the sub-title of the book wrong, the book is actually almost exclusively about non-Western states, so Weakness #1 it identifies is the opposite. Definitely not something you’d want to base an essay on. The other critiques are very vague – how, precisely, would a gendered lens lead to different outcomes (note: I am not arguing that it wouldn’t, but instead that ChatGPT isn’t specifying), for instance? There isn’t much of an internet paper trail for ChatGPT to rely on; it seems like recent work or stuff without a big literature will be a real weakness for this tool.
Question: Is Max Weber’s approach to studying society more or less persuasive than Karl Marx’s?
Answer: “Both Max Weber and Karl Marx are important figures in the field of sociology, and their approaches to studying society have had a significant impact on the discipline. However, it is difficult to determine whether one approach is more persuasive than the other, as they differ in their assumptions, methods, and focus.
Max Weber’s approach is often referred to as interpretive sociology or verstehen sociology, which emphasizes understanding the subjective meanings that individuals attach to their actions and behaviors. Weber argued that social phenomena could not be fully explained by objective factors alone, such as economic or political structures, but also needed to take into account the subjective experiences and interpretations of individuals.
On the other hand, Karl Marx’s approach is known as historical materialism or dialectical materialism, which emphasizes the role of economic and material factors in shaping social relations and structures. Marx argued that society was divided into classes based on their relationship to the means of production, and that the struggle between these classes was the driving force behind historical change.
Both approaches have their strengths and weaknesses, and their persuasiveness depends on the context and the research question being asked. For example, Weber’s approach may be more persuasive when studying the subjective experiences and motivations of individuals, while Marx’s approach may be more persuasive when studying the economic and political structures that shape society.
Ultimately, the choice between these approaches will depend on the specific research question, the available data and methods, and the theoretical and philosophical perspectives of the researcher.”
My comment: this seems quite typical of these models – they don’t ultimately come down on a particular argument, but kind of summarize each to some extent, then say it depends or is up to the researcher. Useful in crafting a 90,000-feet-up summary, but not (yet) taking the next step. Useful up to a point, but a student would need to dig a lot deeper.
Question: which articles or books offer the most persuasive critiques of Stathis Kalyvas’ 2006 book “The Logic of Violence in Civil War”?
Answer: “As an AI language model, I cannot offer a personal opinion or provide a definitive answer to this question. However, I can suggest some resources that may help you in your search for critiques of Stathis Kalyvas’ “The Logic of Violence in Civil War”:
It’s worth noting that Kalyvas’ book has been widely praised and has had a significant impact on the study of civil war and political violence. However, as with any scholarly work, it is also subject to critique and debate.”
My comment: this is useless.
I’ve been reading and listening a lot about AI the last few weeks; not remotely my area but obviously important. I decided to try out the current version of ChatGPT. This is the first of a couple posts where I ask ChatGPT questions relevant to my research or teaching and see what happens. Today: history.
I asked four historical/factual questions about the same general topic, at different levels of detail or demandingness. The first pair are about the history of Nepal’s Communist movement (something I am researching), and the second about the political economy of technology/development in East Asia (I am reading Chip War right now).
For these kinds of historical questions that have a decent amount of internet-available information, ChatGPT is getting at least to the level of generating quality Wikipedia-level synthetic content. That may sounds like damning with faint praise, but it’s actually a huge deal – if I was a college sophomore crashing on a paper the night before it was due, these answers would all at least offer a solid starting point, saving me real time wandering news articles and Wikipedia entries to get the big picture. Specialists can poke and prod at many aspects of the answers below, but they are not on-face ridiculous.
As we’ll see in the next post, things get way more complicated as we move into other kinds of questions that are not about facts per se, but for “general historical narrative and the rough beginnings of a gesture toward explanation” ChatGPT is already doing some serious things. I did notice lots of network errors and crashes, which I guess is to be expected with a free beta that tons of people are trying out, but would not make me want to rely on ChatGPT to fuel an all-nighter.
Nepal
Easier question: what is the history of Nepal’s Communist movement?
Answer: “The history of Nepal’s Communist movement dates back to the early 1950s, when the country was ruled by a monarchy and was undergoing a process of political transformation. At that time, Nepal had a small and fragmented Communist movement, with several small Marxist-Leninist groups operating independently.
In the 1960s and 1970s, Nepal’s Communist movement underwent a period of rapid growth and transformation, driven by a number of factors including political repression, economic inequality, and the rise of leftist movements around the world. During this period, several Communist parties emerged, including the Nepal Communist Party (NCP) and the Communist Party of Nepal-Marxist-Leninist (CPN-ML).
In 1996, a decade-long Maoist insurgency began in Nepal, led by the Communist Party of Nepal-Maoist (CPN-M). The Maoists aimed to overthrow the monarchy and establish a socialist republic in Nepal, and they gained significant support from marginalized and oppressed communities, particularly in rural areas.
The insurgency was marked by significant violence, with both the Maoists and government forces accused of committing human rights abuses. The conflict ended in 2006, with the signing of the Comprehensive Peace Agreement between the Maoists and the government.
After the end of the conflict, Nepal’s Communist movement continued to evolve and transform. In 2018, two major Communist parties, the Communist Party of Nepal (Unified Marxist-Leninist) and the Communist Party of Nepal (Maoist Centre), merged to form the Nepal Communist Party (NCP). The NCP won a significant victory in Nepal’s parliamentary elections later that year, and formed a government with Communist leader KP Sharma Oli as prime minister.
However, in March 2021, the NCP split into two factions, with Oli leading one faction and former prime minister Pushpa Kamal Dahal leading the other. The split has led to significant political instability in Nepal, with ongoing protests and calls for new elections.”
My comment: as a Wikipedia-level answer, this is quite adequate, even if you can definitely have serious quibbles with some of the claims (the 1960s-70s certainly saw “transformation” in the Nepali Left, but “rapid growth” seems a a stretch given the continued dominance of the panchayat system and the crushing of 1970s ultra-left movements – but again, this is quibbling compared to the likely purposes of using ChatGPT to get this answer; same deal with the specific timing of the CPN unification (coalition vs. actual party merger) and the 2018 election).
Harder question: What were the factional politics within Nepal’s Communist movement during the 1990s, and what key issues divided the movement?
Answer: “During the 1990s, the Communist movement in Nepal was divided into several factions, each with its own distinct political ideology and agenda. The two main factions were the United Marxist-Leninist (UML) and the Communist Party of Nepal (Maoist) (CPN-M).
The UML was formed in 1991 after the merger of the Communist Party of Nepal (Marxist) and the Communist Party of Nepal (Marxist-Leninist). The UML believed in a gradualist approach to achieving socialism and advocated for democratic means to achieve its political goals.
On the other hand, the CPN-M was formed in 1994 after a split in the UML. The CPN-M was led by Prachanda and Baburam Bhattarai and believed in a more radical approach to achieving socialism. They advocated for a “people’s war” to overthrow the existing government and establish a socialist state.
The key issues that divided the Communist movement in Nepal during the 1990s included the role of parliamentary democracy in achieving socialism, the nature of the revolution needed to bring about socialism, and the relationship between Nepal and its two powerful neighbors, India and China.
The UML believed in participating in parliamentary NETWORK ERROR“
My comment: It had been slow-going to this point, and ChatGPT ran out of steam. I had it regenerate an answer but it had another error. I don’t know if this was an idiosyncratic general network problem, or specific to this question, but it was definitely making some progress and I could see it offering a pretty detailed answer in future iterations.
Taiwan/East Asia
Easier question: what are the geopolitical dynamics of semiconductor production in Taiwan?
“The geopolitical dynamics of semiconductor production in Taiwan are complex and multi-faceted. Taiwan is a major producer of semiconductors, which are critical components used in a wide range of electronic devices, including smartphones, computers, and automobiles. Given the strategic importance of semiconductors, their production and distribution have become key elements of global power politics.
One of the key geopolitical dynamics in the semiconductor industry in Taiwan is the role of China. China is a major consumer of semiconductors, and has been seeking to increase its domestic production of these components in order to reduce its reliance on imports. However, China’s semiconductor industry is still relatively underdeveloped, and it relies heavily on imports from Taiwan. This has created a complex dynamic between the two countries, with Taiwan seeking to protect its semiconductor industry while also trying to maintain positive relations with China.
Another key geopolitical dynamic in the semiconductor industry in Taiwan is the role of the United States. The US is a major consumer of semiconductors, and has been concerned about the potential impact of Taiwan’s semiconductor industry on national security. The US government has been seeking to promote domestic semiconductor production and has also been pressuring Taiwan to increase transparency and security measures in its semiconductor industry.
In recent years, tensions between the US and China have also added to the geopolitical complexities of the semiconductor industry in Taiwan. The US has been using its economic and political power to pressure Taiwan to limit its semiconductor exports to China, while China has been seeking to increase its control over Taiwan’s semiconductor industry through economic and political means.
Overall, the geopolitical dynamics of semiconductor production in Taiwan reflect the complex interplay between economic, political, and security considerations. The country’s strategic importance as a producer of semiconductors has made it a key player in global power politics, with the actions of Taiwan and other countries shaping NETWORK ERROR
There was an error generating a response”
My comment: Both times I asked I ended up with a network error. Still, even before ChatGPT crashed, this is a useful beginning of an answer.
Harder question: What explains the divergent developmental strategies of the “Asian Tigers”?
Answer “The “Asian Tigers” refer to a group of four countries in Asia: South Korea, Taiwan, Hong Kong, and Singapore, that experienced rapid economic growth and development between the 1960s and 1990s. While these countries share some similarities in terms of their developmental strategies, there are also important differences in their approaches.
There are several factors that may help explain the divergent developmental strategies of the Asian Tigers:
In summary, the divergent developmental strategies of the Asian Tigers can be explained by a combination of historical context, political institutions, human capital, economic policies, and geography and natural resources.”
My comment: there’s not a real argument here compared to a laundry list of stuff, and lots of kind of vague claims (“may have influenced”), but as a rough starting point it’s not useless at all in my opinion.
Blair Kamen is one of Chicago’s major architectural critics/experts. Here he offers six ideas on how Chicago’s downtown “will rise again.” I particularly liked the first point:
“The lesson: Confident downtowns do not panic. They follow in the tradition of Daniel Burnham. They plan. They make big plans, Burnham style. And they make small plans, which wisely recognize that different streets, and even different sections of streets, represent different but interrelated parts of a larger urban ecosystem.”
A common, and I think broadly accurate, critique of US foreign policy in Cold War Asia was its tendency to exaggerate the extent of Communist power within countries. Which makes it always interesting to find a countervailing case, in which the US did not buy the Communist card being deployed by a local actor.
An example is the State Department’s assessment of King Mahendra’s December 1960 coup in Nepal; after a meeting with the king on December 20 in which he claimed his coup was a result of needing to overcome corruption and communism, the Embassy offered its analysis:
“In analyzing this coup d’etat, for this is what we believe it to be, we feel that the King’s motives in taking the precipitate action he did were guided less by the issues of corruption and Communism than by a growing fear that his own personal position and prestige were dwindling and that if he did not act soon, it might be too late. [2 lines of source text not declassified] While it is doubtless true that there has been corruption in high places and evidence, some true and some fabricated, will be presented to prove this, and, less likely, there may be discovered some vague connections with Communist activity, the real motive behind the move was the preservation of the monarchy and the Shah dynasty in its absolute form. Although the King protests that the decision was his alone, we are convinced that it was aided and urged by the group around him, which may also have misled him. This group includes members of his and his wife’s family, remaining Class A Ranas, hereditary Generals and reactionaries and “feudal remnants” generally, who, themselves, are concerned over the survival of their privileged positions. Added to these forces are those land owners and others who stood to suffer financially from the enforcement of the recent tax and land reform laws.”
Paul Staniland 