Talk:Great Divergence

From Wikipedia, the free encyclopedia
Jump to: navigation, search
Former good article nominee Great Divergence was a Social sciences and society good articles nominee, but did not meet the good article criteria at the time. There are suggestions below for improving the article. Once these issues have been addressed, the article can be renominated. Editors may also seek a reassessment of the decision if they believe there was a mistake.
June 23, 2010 Good article nominee Not listed
WikiProject History (Rated B-class, Mid-importance)
WikiProject icon This article is within the scope of WikiProject History, a collaborative effort to improve the coverage of the subject of History on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
B-Class article B  This article has been rated as B-Class on the project's quality scale.
 Mid  This article has been rated as Mid-importance on the project's importance scale.
WikiProject Economics (Rated B-class, Mid-importance)
WikiProject icon This article is within the scope of WikiProject Economics, a collaborative effort to improve the coverage of Economics on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
B-Class article B  This article has been rated as B-Class on the project's quality scale.
 Mid  This article has been rated as Mid-importance on the project's importance scale.

Rewrite because the approach is wrong[edit]

I appreciate the continuous efforts at improving the article, particularly of Kanguole, but I have to say that the basic problem of the article which was outlined here has not been remedied, but in fact worsened. Then we agreed that the focus must necessarily lie on the exceptional Western industrial development, its move away from the traditional constraints of agricultural modes of production, and not on the rest of the world which continued to follow traditional modes of production. These could serve as a baseline, but no more than that.

However, as it stands, the article extensively deals with non-industrializing countries and goes to great length trying to show that the Great Divergence took place as late as can be, developed as weakly as can be and was as coincidental as can be. In this vein, the reductionistic Coals and Colonies approach by Pomeranz, a minority view and critically discussed by other economists at length itself, still permeates much of the text.

So, how should the final article structure look like? We have to ask ourselves what is important and what not; what caused the Divergence, what changed and what remained the same? What was revolutionary and what remained traditional? In this context, I strongly advocated to conjure up the image of a scissor: The Great Divergence, that was a historical scissor where one blade (rest of the world) remained static, while the other (Europe) opened up. Therefore, the compelling question which should guide the article is not why the first blade remained in the position it has always been, but why and how the second one suddenly moved.

To put it bluntly: This article is by necessity through and through euro-centric because it describes an event of western history. These events began in Europe and America, were primarily caused by intra-Western processes and developments (many of which aren't even mentioned yet) and practically remained confined to Europe and America in the given time span. That has yet to be fully reflected by scope and contents of the article. Regards. Gun Powder Ma (talk) 20:16, 14 August 2010 (UTC)

I agree. The answer is to add all that material from a range of sources. Pomeranz cannot be ignored, but he should be one voice among many, and he should not always be given the last word. That will also involve cutting material currently in the article that is not directly related to the subject. Kanguole 00:09, 15 August 2010 (UTC)
Glad we agree. Pomeranz is a valid minority opinion, but not more than that. For the baseline comparison is very useful: David S. Landes: "Why Europe and the West? Why Not China?", Journal of Economic Perspectives, Volume 20, Number 2, Spring 2006, Pages 3–22, but note that DL has a dislike for quantitative economic history (in this respect he is the direct opposite of Maddison). Gun Powder Ma (talk) 17:48, 15 August 2010 (UTC)
Thank you Gun Powder Ma for making the article more neutral! —Preceding unsigned comment added by (talk) 18:44, 17 August 2010 (UTC)

Where is philosophy?[edit]

My thesis is that the divergence has nothing to do with coal or economics... but philosophy and culture. The thinkers of Europe shaped the societies to become liberal and capitalistic and it did evolve the political systems and thus the culture. Shape the culture and intelligence will follow. Example: Our culture endorse innovation which have given us the computer which has created new subcultures among the youth, which will eventually put these new cultural behaviours on the whole society which will then evolve and create more innovation which will create new culture.

But first comes culture. Shouldn't there be something about philosophy?

 —Preceding unsigned comment added by (talk) 18:54, 17 August 2010 (UTC) 
I agree. Much, if not most of the rise of the West had nothing to do with the alleged readier availabilty of raw materials in Europe or unfortunate policy decisions of other political entities which allegedly put them at a disadvantage. To assume this is a thoroughly positivist mindset which just looks at things quantifiable or visible (and gets it wrong even there often enough).
Rather, it was the different mindset which made the difference, the Western rationalism. In advocating this view, we have a prime thinker in Max Weber, and Ricardo Duchesne has recently reaffirmed his conclusions here. And, of course, Karl Marx, not the one of Lenin and Stalin, but the 19th century economist, one of the very greatest of all times, who was a contemporary of the events and devoted much thought to this question. As I said, many of the most important topics the article has not even touched on. Gun Powder Ma (talk) 09:52, 18 August 2010 (UTC)
Credits to Gun powder Ma! You must be one of the most brilliant people on this planet! —Preceding unsigned comment added by (talk) 13:30, 19 August 2010 (UTC)
This article is definitely missing an explanation on the non-material differencies in not only philosophy but also politics/government and scientific tradition. Numbers divorced from their context can be very misleading. For instance, just to illustrate my point, the Incan capital at the time of the Spanish conquest of their kingdom was reputed to be far larger than any city in Europe with an enormous and disciplined workforce. However, the Incas also lacked writing and basic science among other things. If it was a matter of mathematics and sheer numbers regarding urbanization and possibly also agricultural output of products X and Y, then the Incas would have been in a superior position to Europe and we would be justifiably dumbfounded to this day as to why Europe and not the Incas led the world into modernity. Yet, I don't think that anybody would seriously dispute the fact that the non-material differences between the Incas and the Europeans were far more important than the material ones, namely the enormous legacy that Europe had amassed in politics, law, philosophy and science which made possible certain dynamics in Europe that were simply unthinkable in the Incan world. Certain such differences existed between Europe and China, between Europe and the Islamic caliphate, between China and Japan, between India and China, etc, and they all played out in different ways and contributed into shaping the different paths many of these countries and regions have taken since, some more significant and some less. Not including these non-material differences is tantamount to avoiding the subject that this article is supposed to be dealing with... Abvgd (talk) 22:20, 6 January 2011 (UTC)
Actually, the Incas didn't have any very large city. The civilization of the "city bigger than any in Europe" is actually the Aztecs. The Aztec capital, Tenochtitlan, is estimated to have had 100,000 inhabitants, about the size of the largest European cities of the time, and larger than any Spanish city at the time (Madrid had around 50,000-60,000 inhabitants). However, that was because the Aztec empire channeled a great proportion of the resources of Mesoamerica and concentrated them on this single city. Europe didn't have consumer cities, but instead had a network of cities that contributed to the development of the European economy. Urbanization is a very important indicator of economic development since it indicates the proportion of the labor force that needs to be employed on agriculture. While the largest cities in the world weren't in Europe during the early modern period, Istanbul and Beijing were the two largest, that was because these cities represented an artificial concentration of surplus agricultural production due to political factors, not because China and the Ottomans were more urbanized. For example, take Europe and China in 1650, Europe and China had 100 million inhabitants, but China's largest city was nearly twice the size of Europe's largest city (Beijing had 700,000 inhabitants, while Paris had 450,000 inhabitants), that wasn't because China was more urbanized than Europe, but because Europe had a less primal urban system. Overall, Europe had 10 cities over 100,000 inhabitants while China probably had only 4,5 cities over 100,000 inhabitants. China was consequently less urbanized. --RafaelG (talk) 22:49, 5 March 2011 (UTC)
My mistake, I meant the Aztecs but for some reason ended up writing the Incas instead... BTW, I fully agree with your analysis. This was what I had in mind when I wrote that numbers can be misleading when taken out of their context, but your comment was much more to the point. Pre-feudal societies usually had emperors who on a whim could redirect large portions of the population or economy into large-scale projects, no matter the costs or consequences. In feudal Europe with the division of power between monarchs, the church and the nobility, this was no longer possible or even desirable. Feudal Europe's economy was much closer to the rules of market economy, while China's economy had more in common with what is today associated with a Soviet-style planned economy. Planned economies are good at channeling resources into specific geographic or economic sectors (thus potentially creating the illusion of development or urbanization) but they are also lousy at managing resources as history repeatedly has shown. Abvgd (talk) 07:19, 17 April 2011 (UTC)


This is a list of the first universities in the world:

  1. University of Bologna, Italy – teaching from 11th Century recognised as University 1088
  2. University of Paris, France – recognised as University 1150
  3. University of Oxford, England – teaching known from 1096 (potentially 825) recognised as University 1167
  4. University of Modena and Reggio Emilia, Italy – recognised as University 1175
  5. Bosnian Church University, Moštre near Visoko, Bosnia (today Bosnia and Herzegovina) – recognised as University 1175

There are a lot more but these are the first. All in Europe. something to add? —Preceding unsigned comment added by (talk) 21:53, 9 September 2010 (UTC)

Is this really relevant given the period this article covers? Nev1 (talk) 23:13, 9 September 2010 (UTC)

I'm not sure it would be accurate to include universities here. Different cultures had different educational systems. The problem lies in the definition of a university. Univiersities have become the de facto standard today as the Western culture influences the rest of the world. Studying where the first universities were feels misrepresentative of the overall picture as it discounts foreign schools just because they did not call themselves a 'university'. (talk) 11:48, 5 January 2011 (UTC)

Comparative graph of populations[edit]

Comparative populations (log scale) of China and Continental Europe between 1000 and 1975

Volunteer Marek, what do you mean by "sloppy (at best) graph"? I agree that it needs careful reading to appreciate the magnitude of the difference between the two (perhaps changing the scale might help), but it does show the population cycles and growth rates in both places. Kanguole 15:30, 27 August 2011 (UTC)

I don't know what discussion might previously have been had on this graph, and I don't have a copy of Feuerwerker, but:
  • The borders of China have varied somewhat over the last thousand years. Is this a count of heads within today's borders, or within a variable area?
  • What exactly is meant by "continental europe" here, and again is there any possibility of distortion due to shifting borders &c? (For instance anybody calculating the population of continental europe by adding up population estimates of constituent countries might find that colonial populations creep into some years totals but not others).
  • Population estimates are much more reliable than they used to be. Displaying population as a simple line masks the massive uncertainties in earlier populations. Is it feasible to show some error bars?
  • Log scales can be intrinsically misleading, and great care should be taken before presenting data on a log scale to a non-geeky audience; is it possible to use a linear scale?
bobrayner (talk) 16:24, 27 August 2011 (UTC)
The data is Feuerwerker's. I don't think he gives a precise delimitation of "China", but even if he omitted the additional areas now in the PRC from earlier figures it would make little difference, as their population then was tiny compared with the Chinese empire. On the other hand, "continental Europe" is a fairly clear geographical term. Error bars might be appropriate, but the source provides no data on which they could be based. The trouble with a linear scale for an exponential variable like population, especially over such a large timescale, is that recent high values overshadow everything else. A linear scale also obscures growth rates, which are central to this article. Kanguole 23:18, 28 August 2011 (UTC)

Regardless of how you define continental Europe and for any reasonable definition of China, it's just not the case that these two regions had comparable populations 1900-2000. Yet the graph makes it look like they're almost at parity today. And yes, I know what a logarithmic scale is. But graphed this way, the diagram *obfuscates* the tremendous differences, rather than highlighting it, as it should. If you think that the exponential graph would overshadow everything else (for China probably, for Europe maybe), it might be better for the purposes of this article to graph the ratio. And if you want to graph growth rates (the slopes of the above lines), then graph growth rates. Another possibility would be to have two (maybe side by side) graphs, one for the pre-industrial era (say, pre 1850) and one for the post. I got the pop data on China handy (by province, from the Chinese gov) for the post war period, but I do not have Feuerwerker's numbers. If you have it could you send it to me?Volunteer Marek (talk) 08:19, 1 September 2011 (UTC)

The graph clearly shows a significant difference in 1975 (it doesn't continue to the present). But that is peripheral – this article isn't about the 20th century. I don't see the point to any of your suggested alternatives: they would show less data, or data less relevant to the topic of this article. Feuerwerker's numbers are on the cited page of the source, but I'll repeat them here:
Year Continental
-200 26 42
1 31 53
200 36 63
400 31 53
600 26 50
800 29 50
1000 36 66
1100 44 105
1200 58 115
1300 79 96
1400 60 81
1500 81 110
1600 100 160
1650 105 140
1700 120 160
1750 140 225
1800 180 330
1850 265 450
1900 390 475
1950 515 590
1975 635 1000
Perhaps the difference in 1950 isn't what you think it should be, but a linear plot wouldn't change that. Kanguole 11:59, 1 September 2011 (UTC)
The graph clearly shows a significant difference in 1975 - yes if one looks at the very end of the graph. But right before it (1974?, 1950?) it looks like they're at parity. Also, these numbers don't have anywhere near the variance that would justify the use of the (more confusing) log scale. If we graph it in levels it looks just fine.
And I think my suggestions were good. In fact your claim that this article isn't about the 20th century supports that. If the article isn't about the 20th century then neither should be the graph. Just keep the stuff up to 1850 or so in there and that will make the graph much clearer. Likewise, using the ratio would present a more easily understood picture.
And yes, there is some discrepancy between these numbers and usually-accepted numbers from other sources, at least for 20th century post 1960 data.Volunteer Marek (talk) 02:57, 2 September 2011 (UTC)
Are you saying you have information that contradicts Feuerwerker's numbers for 1975, or is it his relatively close numbers for 1950 that are the problem? Kanguole 06:12, 2 September 2011 (UTC)
Yes, Feuerwerker's numbers are contradicted by standard numbers from international organizations for 1975, though this could be due to this whole "how are you defining China/Europe" thing. Anyway, unless you really think that in 1950 the two regions were at parity, that's not the main issue here. And to say it again, the log-scale here is just not useful.Volunteer Marek (talk) 07:28, 2 September 2011 (UTC)
Feuerwerker gives pairs of numbers for 1900 and 1950 (close together, whatever scale you use) and a pair of numbers for 1975 (significantly separated). These numbers seem to be at the root of your "almost at parity" complaint. Which of these 6 numbers are you challenging, and what's the basis for discounting Feuerwerker here? Kanguole 08:13, 2 September 2011 (UTC)
Forget about the numbers for a second. What I'm challenging is the graph. Like I said (what, 3? 4? times already?) it is not very helpful here. A different display of the data (assuming it's correct) would be much better.
More generally, what is the graph supposed to illustrate and why is it pertinent to this article?Volunteer Marek (talk) 09:39, 2 September 2011 (UTC)
BTW, the same numbers are in McEvedy and Jones Atlas of World Population History. I'm not sure which one is the "original" source, or as I suspect, both sources got their numbers from another source (for European pop I'm guessing it's from Josiah Russel, which is pretty outdated).Volunteer Marek (talk) 12:56, 2 September 2011 (UTC)
Comparative populations (linear scale) of China and Continental Europe between 1000 and 1975

Here's the linear plot of the same numbers. As before the populations are close in 1950 and significantly different in 1975, following the underlying data. The linear plot emphasizes the difference at the end, but that's about all it does, and that's somewhat peripheral here. The 20th century, like the period before 1400 or so, is useful context, but not central to an article on the Great Divergence. The log plot gives a more informative overview of the populations over the whole period, because in addition to showing the absolute values of the two populations over time it indicates rates of growth, and significant fluctuations in earlier periods are not flattened out to fit in the later exponential growth. Kanguole 00:15, 3 September 2011 (UTC)

Oy, I'm afraid this is going to come down to the aesthetics. Looking at the linear plot I find it a lot more informative than the log scale plot. Here are some specific reasons why:
1. Before the 20th century (actually 1950 or so) China's population was basically an upward shifted version of Europe's. Both graphs show this but the linear graph shows it more clearly.
2. Like you say, the 20th century is not the subject of this article. So why not just drop the observations for 1975 and 1950 (actually just dropping 1975 will do most of the work). Once you get rid of these two there's just no reason to use a log scale plot.
3. What is this "more information" that the log plot gives? The rate of growth? Not really unless a reader is mathematically trained enough to know to look at the slope of the graph. For the average reader that's not going to be true. And if you really want to show the growth rate, then, as I said, before, just make another graph with the growth rates.
4. The fact that fluctuations in earlier periods were of minor magnitude compared to the high growth that came later is itself an interesting piece of information and the linear graph shows it.
At the end of the day though, the questions to ask here are: "What does this graph illustrate about the Great Divergence?" and "Does it illustrate it adequately?".
The Great Divergence is about the differences in living standards that arose between China and Europe (and its offshoots) - population itself is only tangential. If we assume both China and Europe in the pre-industrial era were "Malthusian" then including data on population densities could be useful as in that case population growth proxies for the rate of technological innovation. This is what I think this graph is aiming at showing but it does so in a very unclear and sloppy way. But in that case, there's no reason to include numbers for the post-industrial era (essentially 1900, 1950, 1975... maybe 1950 should be kept). And once you get rid of these "high" numbers there's no reason to use a log scale.
So again, can you provide an explanation for why this graph (and the sentence accompanying it) is even here?Volunteer Marek (talk) 02:04, 3 September 2011 (UTC)
The graph provides background information on conditions in China (and Europe) in the period under discussion, as well as earlier and later periods for additional context. In a Malthusian setting, with incremental productivity increases being consumed by expansion of the population, the latter is directly relevant. It might be interesting to graph growth rates, densities or other variations, but I know of no sources on which we could base such graphs.
China's population history before 1950 was actually quite different from an upward shifted version of Europe's, though this is harder to see from the linear plot. Those earlier fluctuations were highly significant in proportion to the populations of the time, even if they are miniscule compared with modern populations.
As for truncation, to omit the 1975 levels (or the 1975 and 1950 levels), would give the misleading impression that Chinese and European populations were converging, which was your original complaint. Omitting levels after 1850 would give the equally misleading impression that the high rate of growth to that point was sustained (an issue directly relevant to the Great Divergence). Kanguole 13:32, 4 September 2011 (UTC)
The growth rates can be calculated from this data. In fact, that's what the log-scale graph essentially shows.Volunteer Marek (talk) 23:39, 4 September 2011 (UTC)
And if the graph of the levels before 1950 shows that the populations were converging (not really) then that IS in fact relevant here.Volunteer Marek (talk) 23:41, 4 September 2011 (UTC)

Error bars[edit]

Population estimates are much more reliable than they used to be. Displaying population as a simple line masks the massive uncertainties in earlier populations. Is it feasible to show some error bars? - this is generally not going to be possible (bad data means not just unknown level, but also unknown measurement error), particularly since I'm guessing a lot of this "line" is linear extrapolation. I don't know what form these data take but if they are high enough frequency and NOT just filled in with linear extrapolation then you could do a moving average or some other filtering and hope the measurement error is just noise which will average out. But it doesn't look like it - there's a lot of smoothness there already.

More generally, best data on historical populations is from England. And even there the quality stuff goes back only to 16th century. Semi-reliable data goes back, sporadically, to the 13th. But even rest of Europe has very very very sketchy data for population, with a lot of these numbers you see out there basically just pulled out of thin air. China's probably even worse in that respect (at least in terms of what is accessible to Western researchers). One thing that could be noted is the singular years when standardized coordinated censuses began to be taken. 1841 for England (though that one was sketchy, the first decent one was 1851) and only the end of the 19th century for rest of Europe. China - I don't know off the top off my head.Volunteer Marek (talk) 08:31, 1 September 2011 (UTC)

A small but significant piece of pedantry: The 1841 census covered the United Kingdom (more or less), not just England. Let's be careful with demonyms :-) bobrayner (talk) 08:38, 1 September 2011 (UTC)
Lol, yes. But the Wales and Scotland ones are even less reliable than the English one so often times demographers just use the 1841 English one.Volunteer Marek (talk) 08:43, 1 September 2011 (UTC)