Відео

Its hard to tell how close we are. At the moment it would require bespoke software, which isn't going to happen. Widely available AI tends to be good when it has enormous data sets to work with (chatbots, translation, computer programming). At the moment the data set that could be easily assembled to train the AI would be very limited. But you are not the first person to ask me this question so it is something people are considering.

That is a really good question. The experience of every educator I have met with suggests no. The most effective way to learn about history is to pick something you are interested in - something you are motivated to delve deep enough to begin to read primary sources and try to understand the debates behind the narrative. If you want to think in broader terms, history becomes more alien to your personal experience the further back in time you move, so most teachers find they are much more successful beginning with recent history and then moving backwards gradually building on that. I was working with a colleague who is doing some work on Chinese material culture and I had her start with the Qing (up to the nineteenth century) and work backwards. She got to elements of the Tang and mid first millennium today and commented on how lost she would have felt if she had begun there and tried to work forwards. I think that is a useful frame for general interest, but I would reiterate that enjoying the process is key, find something you care about.

Sorry, busy week. Let me answer the questions from broadest to narrowest. 1. Where are the coins? That's a genuinely good question, maybe even worth an episode. At one level we are not entirely certain (a conference is taking place this month on one aspect of the problem), but in general - most coins are in private hands for most coinages, the public collections are a selection from what is extant (and their catalogs are a selection from that). And probably even more coins are in the ground - the English coinage of the 1066 period was recently nearly doubled by the discovery of a single hoard (search for 'Chew Valley Hoard'). 2. The abbreviations are those used by Prasad for different collections, e.g. SML = State Museum Lucknow; BKB = Bharat Kala Bhavan (on the campus of a university in Varanasi). If memory serves all are Indian collections except the British Museum (BM). 3. D = nd/(n-d) is the easiest formula for calculating the number of dies that were originally used from the surviving sample. Because the completeness of the sample of coinage varies and many dies do not survive its necessary to apply a formula before comparing numbers.

I am aware of some of Silver's criticisms but am not sure which specific article you are referencing. Its really useful, not just for me but for other readers chancing on the comment, if you can leave a full link or a full reference.

Yes, very much. I find it a pity that we don't appear to have a full written account from Lichtman as we have had in previous elections (I first saw the 'official' call on a news outlet video via Chris Cillizza's channel, ua-cam.com/video/ysF9kZcIzao/v-deo.html, but there is an in depth analysis on Lichtman's own channel ua-cam.com/video/okYUg_Dx1iU/v-deo.html). Lichtman spends a lot of time arguing against the 'subjective' critique in his video, its worth noting that though some keys are subjective and some are ambiguous I'm not in that camp. And I'm inclined to agree with the case that Lichtman makes that you cannot make a judgment by reading the key, you have to do it with regard to the coding of the historical data. In both videos I saw Lichtman does a count skipping the foreign policy keys and gets to 3 false (1 Mandate, 3 Incumbency, and 12 Charisma). That means even if both foreign policy keys were false (10 failure, 11 success) the total would be 5 and Harris would be predicted a win. From 25 minutes in the longer video he then goes into the foreign policy. He calls 10 false (which given Afghanistan, Ukraine, and Gaza, seems beyond reasonable question). He calls 11 true, based on Ukraine, but this is clearly a dubious fudge. In Social History 88, which was published in January/February this year he called this clearly false with precisely the same information he has today. So, in his prediction I have issues with 11 which he is calling inconsistently with his own statements. I would say his read of the model is 5 false (he claims 4 false, but that contradicts his own prior analysis), so one more predicts a Trump win. Obviously I have problems with keys 2 and 9 (key 7 on policy I wouldn't personally express an opinion on without spending some time looking at the historical data). The contest key (key 2) is discussed at 8 minutes 30 into his own video, and the defence is that if a party unites after the contest it cancels a contest. This position is not defensible in a good faith way. There are 12 false contest keys in the historical data and it would require an argument that in all 12 of those cases the incumbent party chose not to unite behind its final candidate - which seems a hard case to make (the point of the historical analysis is not that parties refuse to support their candidate but that publicly splitting over who the candidate should be causes damage that cannot be repaired by good PR). The scandal key is more complex - but in the videos I saw he offered no serious defence - he mashed the age issue up with Hunter and then appealed to the 'bipartisan' criteria (Ysa and I came to that same criteria for assessing the Hunter Biden issue independently of Lichtman, and you could use that agreement as an argument in favour of the model - it suggests the apparent subjectivity some critics claim is lack of relevant skill in the assessment, or that would be the argument). The problem is that the age problem went bipartisan, and that is what provoked the contest that forced Biden out. In the long video (after 19 minutes) Lichtman briefly mentions 'corruption', but this doesn't really hold as Clinton's case was a sex scandal, not as previous scandals were -malfeasance and criminality (but to be fair even in the long video Lichtman makes no in depth argument on this point so I wouldn't want to try and read too much into it). The question with Biden's age is not whether or not it reaches the standard necessary to trigger the scandal key (it clearly does) but whether health qualifies as a scandal at all (for which we have no precedent in the historical data). So, yes, it still looks to me like Lichtman is fudging the keys to predict a Harris win when the model strongly suggests a Trump win. From the point of view of wanting to be right that is a smart move, Harris is more likely to win (at minimum she is likely to get the popular vote even if Trump is elected). Whether this is Lichtman consciously trying to be right or simply that we are close to the event and unconsciously he is smarting from 2016 (remember Lichtman has worked hard to fudge his own prediction. There is a critical examination of his attempt to obscure the difference between calling the popular vote, which the model does, and calling the winner, thepostrider.com/allan-lichtman-is-famous-for-correctly-predicting-the-2016-election-the-problem-he-didnt/). The fudging is clearly an issue. I find it a hard critique to raise against the predictions before 2016 (the analysis in the video, which was initially written before the Biden debate, changed dramatically after his remarks on Harris becoming the candidate). But that it appears to be the case here lends credence to many critiques that have been raised against Lichtman in the past. Personally I wonder if it might not be better to say the model predicts a Trump win but the model is probably wrong and then talk about why the model is probably wrong (both parties selecting clearly unsuitable candidates, competence meeting the bar for the scandal key for the first time, etc.). If you read older work by Lichtman you will see he would often finish his accounts of the model with provisos like that, pointing to factors that weren't present in the historical data set, and how they could upset the model.

@@HistoricalPerspectiveRBr Thank you for a detailed response. I think a video where you and Ysa would go over the subject would be of interest to the UA-cam audience. For me it just doesn’t seem plausible that such things as the 11th and 12th foreign policy keys could have the same influence as the 5th and 6th economic keys. Besides, what I think even the significant foreign failure being arguable--at the end of the day, the US did pull out of Afghanistan, Ukraine did not fall, what was exactly a failure of the US in the context of the October 7 attack?--it just doesn’t seem right that the US public would give foreign policy the same weight as they would to the performance of the US economy--btw, the assumption that the majority of people objective evaluate the economy also seems incredibly doubtful. As far as I understand, his model has no recourse to explain how can the voters in one set of states could be more objective in evaluating such things than voters in the others. With historical margins for victory in some of the states being so small, prediction of the winner by looking at these overarching issues just seems so random.

@@HistoricalPerspectiveRBr for some reason UA-cam didn’t want to display my first response. Here, I’ll try it again. Thank you for a detailed response. I think a video where you and Ysa would go over the subject would be of interest. For me it just doesn’t seem plausible that such things as the 11th and 12th foreign policy keys could have the same influence as the 5th and 6th economic keys. Besides, what I think even the significant foreign failure being arguable--at the end of the day, the US did pull out of Afghanistan, Ukraine did not fall, what was exactly a failure of the US in the context of the October 7 attack?--it just doesn’t seem right that the US public would give foreign policy the same weight as they would to the performance of the US economy--btw, the assumption that the majority of people objective evaluate the economy also seems incredibly doubtful. As far as I understand, his model has no recourse to explain how can the voters in one set of states could be more objective in evaluating such things than voters in the others. With historical margins for victory in some of the states being so small, prediction of the winner by looking at these overarching issues just seems so random.

One could even plausibly argue that people have already been influenced by the system in that the scandal key is arguably something that can be flipped on an incumbent administration - - of course, whether this is because people are reading it out of the keys system or simply the political outcomes it produces it seems to be in a relevancy because it is self reinforcing. But it certainly highlights the importance of the incumbency key and the contest key for their heavily predictive power and suggests that parties should try to run for that second term or do everything possible to avoid an intro party fight. Again, though this simply does reinforce what we already want to believe about politics in many ways, a combination of conventional wisdom and idealism. In the deepest concern that I have about looking at elections this way is that it really does not encourage the kind of detailed policy politics that we also idealize- - largely because it almost suggests that elections and campaign, barely matter at all. There is another thread of thought here that the system may have been extremely capable during certain historical periods like the Cold War, but may be less applicable to more chaotic periods such as the one in which we appear to be living, where there even arguably is alignment taking place. As such, I think it is an interesting system to watch election after election to see whether these insights prove viable or not. I really think it’s more important to focus on these fundamentals that the keys seem to highlight rather than the Parlor game/gambling aspects - you may want to bet money on the keys prediction, but those of us in the USA are also essentially betting our futures on the election anyway.

Actually, as has been pointed on numerous times by people familiar with it, the 13 Keys system only predicts the winner of the popular vote, and not the winner of the election. And the electoral college system (the result) has only differed from the popular vote five times in US history. 2 of those occurred in the late nineteenth century and were included in Lichtman's training data for the model, 2 have occurred in the period the 13 Keys has been used as a prediction (one of which Lichtman called correctly in 2000, one wrong in 2016). Some people may have felt the popular vote had become irrelevant in 1888 having experienced two elections where its winner did not become president but if they did their fears were unfounded, more than a century passed before the two diverged again. One of the interesting questions is whether a political re-alignment is at work in the US, which will render the model invalid or irrelevant; and a failure of the electoral college to reflect the popular vote on a regular basis would certainly fall in that category.

You could certainly spend quite a bit of time on the 2016 prediction. People have. Some favour the same position as Ysa and are not generous with their interpretation, see for example thepostrider.com/allan-lichtman-is-famous-for-correctly-predicting-the-2016-election-the-problem-he-didnt/. You could also suggest that maybe it was hard to call, for example Lichtman did call key 4 incorrectly in 2016 (which in his defence is both not subjective and very hard to call, so its not hard to see in hindsight if it was called wrong), which given how he called key 2 at the time would have flipped the result. However you quickly get into the post-hoc problem. The reason no-one was impressed with the 13 keys in the 90s is because predicting things after they happen is easy, its the success of the model in predicting elections in advance that has garnered it attention. The problem from a historiographic point of view is that what Lichtman has done isn't enough to say this is not just one guy who got lucky - and its hard to see how more predictions of more elections, or re-examining old predictions would resolve that. Its all just enough to act as a caveat on conventional historical wisdom about why you cannot predict the future from historical patterns but not enough to actually serious challenge that.

Lichtman's 13 keys got Al Gore right - the election it got wrong was 2016 and Trump. A common misconception is the model predicts the winner of the electoral college - it doesn't - it predicts the winner of the popular vote; those have not aligned in four cases, two the model is trained on, and two the model predicted.

If you assume coin flips it would be 1/2^10 which is roughly 1 in a 1000. 9 out of 10 is about 20 times easier. But a lot rests on assumptions. That is for a coin toss, but pollsters do a lot better than 50/50. If you said (purely for an example) a pundit might make an educated guess and get 7/10 then 10 out 10 is about 1 in 40 (still unlikely but much better odds, and obviously 9 out of 10 is even better).

Well, actually, Ysa attributes it to luck - I cover precisely the issue that 9 out of 10 is surprisingly high early in the episode (and to correct your common misapprehension, it correctly predicted 2000, as it predicted Al Gore would win the popular vote, which he did). However, its not so high as to constitute evidence in its own right because it only slightly outperforms pollsters and there is the potential for a publication bias - but an episode on evaluating the likelihood of getting 'lucky' would be quite lengthy in and of itself; the only important point is that the performance is not so good that 'lucky' in the sense Ysa intends it (i.e. fudging numbers when it seems other factors might outweigh the model) can be ruled out.

@@HistoricalPerspectiveRBr It could very well be just luck. The poster considers this to be statistically unlikely, because he considers it an experiment with an n of one, i.e. Lichtman. But if we consider that thousands or millions of people in addition to Lichtman have been making these predictions since 1984, it doesn't seem so unlikely that there will be one who gets 9 out of 10 calls right by luck. Imagine if that many people were flipping coins, one guy getting 9 out of 10 calls correct would not be remarkable, but inevitable. He just wouldn't necessarily be named Lichtman. The argument is that Lichtman's system does not work, because it cannot work seems unpersuasive. In the first case, it is not clear that social events such as an election are so inherently unique in all aspects as to render prediction a priori futile any more than forecasting a rainstorm is. In addition, throughout most of the past several hundred years, the tools to make predictions, including probability and statistics, were in their infancy or non-existent. In the case of US presidential elections are we talking about a number of elections too small to be statistically significant or are we talking about the decisions of tens of millions of voters who might indeed be subject to the Law of Large Numbers, for example? In sum, a thoughful video.

I'm going to leave this comment in place because it is such a wonderful example of problems with the internet as an educational tool. A viewer encounters a fairly uncontroversial statement of fact that contradicts their prior belief and immediately refuses to ever engage with that source again. In case you are reading the comments without listening to the video the model predicts the winner of the popular vote (not who will become president, though the two are usually the same) and in 2016 it predicted Trump but it was Clinton that won.

@@HistoricalPerspectiveRBr Which version of reality are you living in? Hillary did not win in 2016.

@@HistoricalPerspectiveRBr ua-cam.com/video/o4MVuRIiD4s/v-deo.htmlsi=UvomKX3NIoGg_NU2

@@HistoricalPerspectiveRBr ua-cam.com/video/DVsDEXZX0b8/v-deo.htmlsi=BEz7hb0txKxvlWD2

@@HistoricalPerspectiveRBr ua-cam.com/video/DVsDEXZX0b8/v-deo.htmlsi=BEz7hb0txKxvlWD2

As your first post stemmed from a basic misunderstanding of the 13 keys model you should probably not take offence that any claim about how obvious a particular element is will be taken with a pinch of salt. Listen carefully to the video Lichtman prepared and to this. On that part of the scandal key (no.9) that we both directly address we both give exactly the same answer, and even use the same word 'bipartisan' as part of our judgement despite the remarks being independent of each other (and 'bipartisan' not being in the text of the keys) - as you might expect two professional historians to do faced with the same criteria (its actually not a bad point to use as positive evidence for the keys). However, the video you looked at by Lichtman simply does not address the question that was being discussed here. "Answers to some of the questions posed in the Keys require the kind of informed evaluations that historians invariably rely on in drawing conclusions about past events." Is what Lichtman wrote on the matter. Like many people who misunderstand the keys you try to reduce what Lichtman himself identifies as complex and nuanced judgements to the simple reading of numbers (others do this with polling and the third party key by over-interpreting the rules of thumb presented in publications). No doubt you do so because you fear the danger of subjectivity, but rather than avoid it you make special pleading easier. Publicly forcing the presumptive nominee to stand down against their will a priori rises to the standard of a 'serious contest for the incumbent-party nomination' and since Lichtman has not addressed that (his comments on it are directly discussed in the video) it presents a compelling case for anyone wanting to draw the same conclusion as Ysa. Its admirable to want to correct (I assume you meant correct and not 'conciliate') misunderstandings, but to do so you need to actually understand the thing yourself. There is a valuable exercise that will benefit you, and that is to try to imagine why two people who understand something better than you can come to two different conclusions from the same data without ascribing a mistake or a lack of insight to either (performing this exercise helps students develop the kind of nuance that historians value and its particularly useful for understanding historiography).

Unfortunately you are quite wrong. If you listen to the video you will find there is a lengthy discussion of your misunderstanding, which is quite widespread. The model (unless you believe the whole enterprise fundamentally compromised Lichtman has made no predictions) predicts the popular vote, and it predicted this incorrectly in 2016 (it predicted it correctly in 2000).

@@HistoricalPerspectiveRBr ua-cam.com/video/o4MVuRIiD4s/v-deo.htmlsi=UvomKX3NIoGg_NU2

@@KenechukwuAniagboso ua-cam.com/video/DVsDEXZX0b8/v-deo.htmlsi=BEz7hb0txKxvlWD2

I think the intuition that if you know enough it surely must be possible is very natural. Though as you acknowledge in the last paragraph we never do know enough. The 13 keys model is interesting precisely because it asserts that we don't need to know very much - for example, while it feels like that attack on Trump was a huge event with massive real and potential ramifications the model says not. Ultimately I prefer the comfort of the theoretical defence that events either are (or are best studied) as unique phenomena over a single outlier, though I find the 9 out of 10 record a more disturbing challenge than Ysa does (there are theoretical critiques which could be mounted, including publication bias, but which would require a lot of leg work).

​@@HistoricalPerspectiveRBr Yeah, though if he's fudging his 13 keys according to a larger number of unacknowledged and perhaps subconscious factors, it becomes less impressive. Still it is interesting to think about how, if the model really works, then maybe one or both candidates actually getting assassinated wouldn't even change the outcome. That definitely feels counter intuitive to me, like something out of Isaac Asimov. Which makes it pretty cool if it's actually true

As far as I am aware there is nothing new but these things tend to build up slowly and un-noticed and then suddenly you see a renewed interest in the problem. If my intuition is right then there has been a shift in focus and people working on the general problem have shifted from origins and place to the question of how people saw and interacted with the images. I am reading a book by Jas Elsner at the moment on Amaravati, and it seems like that is a definite theme both in that and for recent articles quoted (ie things I would not have read at the time of preparing this episode). Sorry my response was so slow, Ysa and I have been racing to prepare something on the presidential elections form 5pm today.

I try to like and comment on every video cause this channel is so much more deserving of views than the stuff the algorithm usually promotes. I hope it gets really popular someday

I have never read the book on slavery. I know it was very well received when it was written but don't know how Romanists would reflect on it today. I suspect a modern scholar might be inclined to problematise 'slavery' itself and ask to what degree things we do or do not label that way are equivalent. A few years ago an American colleague was challenged by an under-graduate student in lectures on the Ghurids and Mamluks, they insisted that it was inappropriate to say 'slave' as it condones the dehumanising institution and that instead they should have said 'enslaved person'. I think its a little too easy to simply dismiss the students concern because replacing 'slave kings' with 'enslaved kings' would obviously be confusing. But the question of the degree to which the Mamluk or Ghulam systems where slave soldiers rose to become rulers, the Roman system where freedmen were amongst the richest citizens, and African-American slavery which was rooted in race, are or are not similar, and the language we use to talk about it, would be a much larger part of the discussion now. Historians write about the past but we always write for our times.

@@HistoricalPerspectiveRBr Finley actually argues a lot of what you are saying, the subtitle is after all 'and modern ideology'. He clearly says for him that a slave must be a commodity. i think a lot of people are sloppy with the word, but I think 'a commodity that is also a person' is a good definition that carries a lot of legs for comparative purposes.

Cart brakes that engage once the cart is off the owner's lot are used primarily as theft deterrent. It is VERY popular for the homeless to abscond with this property of the merchant and use it to, you know, improve their lives; this is used as a sort of mobile hold-all for everything they own.

@@bothewolf3466 Yeah it's definitely mainly homeless people I see taking carts. Plus people where I live people constantly take cans out of recycling bins to sell, and having a cart helps a lot with that. When you're homeless or low income it's quite obvious that a shopping cart is worth more than a quarter

That is interesting. Yes, money is very much locked into social conventions and norms. Anthropologists and numismatists have independently have taken issue with a lot of the assumptions of rationality in economics (to be fair a lot of economists have). I am reminded of an example given by Steven Levitt in which he pointed to a nursery that introduced a fine for parents who collected their children late, and discovered that the fine increased the number of late parents (they now saw the fine as a charge and this freed them from the moral restraint which has made them unwilling to inconvenience the staff).

@@HistoricalPerspectiveRBr Yeah, the "homo econimicus" of Econ 101 is a rare sight in real life. But when times are tough, people get desperate and they lose faith in the social contract, so you see it more and more. The last few years where I live things have been especially bad, lots of people have started thinking that anything not bolted down is free. Things seemed to be getting a little better but then given the debate a few days ago, I'm not so sure anymore. At least I have videos about Indian history to take my mind off the present situation.

I can imagine a medievalist doing a whole channel dedicated to Menocchio - I'd listen to that. Yes, there is quite a bit I could expand on here, and it took some discipline to keep it this short, not sure if/when I would get to looking at it again but will certainly consider it.

​@@HistoricalPerspectiveRBr Glad to hear that!

This is often greatly exaggerated. The effect of deliberate destruction was often quite modest compare to simple re-use (stone is expensive and re-using existing pieces is very common, examples survive of mosques where the walls were made from older temples simply by turning panels around and carving on the back, much more to do with convenience than conviction). The lack of surviving wood (terracotta survives in quite significant numbers but it has not been as well studied) is a serious issue, as is the lack of surviving painting. The lack of these ephemeral arts deprives us of artistic forms that would actually have been very common and which probably acted as an important vector for the transmission of ideas. And yes, there are occasional references indicating some Buddhists were resistant to Buddhism but it is curious that no debate on this was preserved in the surviving literature.

I have to admit I am not sure what to recommend in this regard. The relationship between anthropology and history is fairly deep, with the two disciplines drawing on each other in different ways right the way back to the early twentieth century. I think Google Scholar is your friend here, I was able to pull up a wide variety of works discussing different aspects (scholar.google.co.uk/scholar?hl=en&as_sdt=0%2C5&q=historicisation+anthropology&btnG=) and I would suggest that if you find something that seems relevant look at other things by the author and that cite that work. That is how I usually chase down ideas I am interested in.

I'm glad you found some of it useful. There is a lot there to respond to so I'll circle back round to this when I've a bit more time. Its only Bush senior that is being discussed, so anywhere implying otherwise is me mis-speaking during the video. There is a suggestion on reading at the very end after 1:30:00 and the titles. Disciplinary boundaries are complex - there are interesting things that could be said about the differences between history of economic thought, economic history as practiced in economics, and the history of trade and other economic activity as practiced in history and archaeology (too much for me to cover here) - the only caution I would raise is to never take an example like this (about a specific problem) and try to read it as indicative of the entire discipline that the problem is part of.

‘Economic history’ can cover a lot of activities, all of which blur into each other and overlap. In this specific case we are mostly talking about ‘history of economic thought’ which is certainly a sort of intellectual history though mostly done by people who would not otherwise work on other types of intellectual history. These terms are woolly but usually archaeological research on ancient shipping would probably just be referred to as marine archaeology - likewise if I wrote about Indian Ocean trade, which is also an economic subject, people would not usually call me an economic historian. Yes, I’ve encountered a variety of attributions of ideas to Khaldun. It is true, as I touch on briefly here, that the development of European and Islamic thought are very entwined (in both directions, the divide is more in our conception than the physical realities of the medieval world). But its also true that it is very easy to find ideas if you look for them and I don’t think that practice has been restricted to economics. The connections with Frank Herbert’s Dune are striking. It is obvious Herbert draws heavily on Muslim and Arab history, and given the reading involved presumably must have directly or indirectly encountered Khaldun. However, a quick search done early in the preparation of this episode yielded no discussion of the connections. Finally, on two points you raise I would like to point up the theme of intellectual honesty. I totally get that some of my audience would prefer me to talk more about Khaldun rather than Khaldun’s reception, but I need to be honest with myself about my expertise (much more in historiography, intellectual history, etc. than in 14th C. North Africa or Arabic literature) and reception is what I can intelligently talk about. The second is the deeply entwined connection between politics (in a broad sense) and history. These are inseparable things (which is not quite the same as saying that all history is partisan or that objectivity is impossible) and I think its better to acknowledge than obscure them - I didn’t come to this politically charged topic by accident as a result of an abstract intellectual exercise (for better or worse). Anyway, thank you for the response. I always find it surprising and humbling whenever anyone tells me they have been anticipating my work.

​@@HistoricalPerspectiveRBr I appreciate the thorough reply. I still think the amount of insults against Johnson goes a little beyond what's fair, but I do get your point about intellectual honesty. Of course the ideal is completely unmotivated, dispassionate searching for truth, but someone who isn't at all motivated and passionate isn't going to make an hour and a half video about Ibn Khaldun, are they? And a political motivation isn't necessarily worse than motivation driven by a desire for popularity, views, ad revenue, and sponsorships, which is the cause of a lot of sloppy work and misinformation on this website. So, bottom line: keep up the good work! I'm looking forward to the next video.

Sort of. Skocpol and Somers cite a number of works, and you absolutely do encounter this kind of analysis but there is something of a disciplinary cleavage. That is, someone with my background would be very likely to distrust this method, the people who do it (not universally but its a definite trend) tend to be in political science, or sociology, or anthropology departments. Regarding coins. Yes, its nearly universal that numismatists agree the Lydian and Chinese origins are independent of each other. Whether the Indian invention is independent is much more debated (see ua-cam.com/video/phHah0NqW3I/v-deo.html). No, there is no agreement on why coinage was invented, its debated a lot. And very few people look at the problem comparatively (I recently reviewed a booked called White Gold by the American Numismatic Society, and while I was positive about the individual contributions, I criticised the framing of the volume, which suggested you could work out why coinage was invented by looking only at why the Greeks adopted the Lydian tradition).

That is why I find this so fascinating - without numbers I would have assumed that Marxism would attract way more that East Asian phonology. I sort of get why it might want Jevons as the Jevons Paradox has entered popular discourse and I think the pattern matching the algorithm does it quite crudely statistical - but I think an episode (done in the right tone) and just how staggeringly racist so many late 19th/early 20th century academics were might be more 'fun'.

That encompasses a complex and interesting set of questions, which I probably could not cover in a whole episode. The first is to do with the difference between between primary and secondary sources. Herodotus will always be relevant because the work is a primary source for the period it describes, other works will always need to refer to it because it will never be superseded as evidence. But, yes secondary sources definitely have a useful lifetime, and that can be complicated. Pop histories and general introductions are trying to communicate what is known about questions contemporaries care about using language that is accessible - since all of those three things change over time, they tend to date very quickly (they also tend to be very useful for the relatively short period before they become dated for similar reasons). Specialist works often date more slowly, the notes in Stein's publication of the Rajatarangini, published in the 1890s, are still valuable as secondary literature. This is partly to do with how much new evidence accumulates, how radical theoretical changes are, etc. For example, I suspect if you picked up a copy of Bury (a century old or more) for information on the chronology of Classical Greece it would be basically sound, but the chronology in any book more than 20 years old on Early Historic India is pretty much worthless (it will mislead more than it enlightens). Works like Bloch's Historians Craft tend to have a longer shelf-life, because they are reflecting on practice and that changes more slowly. This is why Barbara Tuchman features in this set of talks despite being a popular historian, because her reflections on what she does have more staying power than her pop histories (in some ways more than other theorists because she is unusual among popular historians to have written those sorts of reflections). But even for method there is a definite shelf-life. Eventually what is useful gets so thoroughly embedded its better learnt through more recent texts. I can't imagine ever doing an episode on someone in the generation before Bloch in this way, because historical practice changed so dramatically in the early twentieth century. I don't think Ranke's methodological reflections would have much insight to offer, however important they were at the time. But some would disagree - Nathan Hill (I reference Nathan because his thoughts are available on UA-cam) thinks Marx is still relevant (I'm doubtful but can see an argument). A lot of theorists think Hegel is still relevant (I'm confident they are wrong), and as I will cover in an episode this year some academics think a 14th century North African writer is still relevant (I think they are nuts). So, yeah, most of Bloch's work is probably no longer 'worth reading' in that sense. But some of it still is - the trick of course is to know which is which.

@@HistoricalPerspectiveRBr Thanks, I didn't expect a response so quickly. Looks like it's a cheap paperback, which is also an important consideration for me cause I'm not rich, so I think I'll check it out.

⁠@@roundninjaif you have any interest in world war 2 history I highly recommend strange defeat. It’s a short, primary document providing a French perspective of the war. Marc Bloch is a pleasure to read and is often required reading for 20th century french history classes.

@@coles11q Thanks for the tip! I was able to find the book online and it's really interesting so far.

who gives a crap

@ History is a real discipline that has standards and Tuchman’s work is very poor

They are usually referred to as the Panchalas. Of the top of my head I cannot actually remember why. The coins they issue are round and die struck with an incuse square, depicting a god - and its usually a pun on the king's name (the name of the king will often end in mitra, so meaning something like 'friend of' the god that is being referred to). The archaeology of the coins is a mess, they have no standard catalogue, and the chronology is often all over the place, but there are a group of coins marked 'Achyuta' which are associated with a reference to a king in an inscription of Samudragupta in the fourth century, and the earliest show no Kushan influence, suggesting they are first century or earlier. In a lot of references to them a break is artificially inserted in the series, allowing the Kushans to occupy the region for a century or more - personally, I think that is unlikely and that the coins are probably a continuous series.

Well, looking at my shelf I have some T'ang poems and some Yuan plays, but I also have that experience of the dynastic histories being pushed as opposed to poems and plays being something South Asian specialists would rather work on. I think its a two-fold move, both conflating a lack of survival for a lack of production, and also tending to favour certain things in a way that exaggerates their importance (if you looked at academic literature you would think Indian Buddhist texts vastly outnumber Indian Jain texts, but of course they don't, its just the relative size of the modern faiths and the transmission, and thus preservation in translation, of early Buddhist texts encouraged a focus on one over the other).

@@HistoricalPerspectiveRBr Very good point about Buddhist versus Jaina texts.