New comment: As I have noted somewhere in these discussions, I propose that as a a starting point economic 'cycles' should have a strong role in periodisation, because without those, we are moving outside the Kondratieff framework. Periods would then be simply named after perceived leading technologies, they would represent 'paradigms' but what drives the transition from one to the other would be difficult to argue.

To may knowledge, the most comprehensive book on long waves is Long Cycles, Prosperity and War in the Modern Age, by Joshua S. Goldstein, Yale University Press 1988. Luckily, the whole book is available online (link above). There, in Part One, Debates, chapter 4, Goldstein compares all the existing data series and discusses their merits. There, I found this periodisation from p.90 quite convincing:

Industrial Growth Rates after Van Duijn 1983, in Goldstein, p.90

	             UK   U.S. Germany	France Japan	
1840s-1870s    3.0   6.2	4.3     1.7     ? 
1870s-1890s    1.7   4.7       2.9     1.3     ?  
1890-1913        2.0 5.3  	4.1     2.5     2.4
1920-1929       2.8  4.8 	        0        8.1     3.4
1929-1948      2.1	3.1	0      -0.9    -0.2
1948-1973      3.2	4.7	9.1     6.1     9.4

I have tried to plot this as a bar chart with the periodisations as a baseline on x axis using gnuplot but failed. It would be good to be able to make our own charts and curves if we have raw data available.

NOTE: This document was updated using the 'create new revision' option in 'publishing options' at the very bottom of the edit mask for this entry. All group members should also be able to edit this and other pages of the emerging 'book'. When changing existing entries the 'create new revision' button creates wiki-style versioning.

old entry:

NOTE: as Brian has noted, there are some discrepancies between my timeline and for instance the table by Carlota Perez. Yet the main difference is that I follow the Freeman/Soete model, which Carlota Perez also does, with the only difference, that she dates the beginning of a great surge at the point of invention of the technology, whereas Freeman and Soete take the reaching of maturity as its beginning. What I have also done is to link the techno-economic cycles with phases of expansion and contraction. What this exactly signifies needs to be clarified - need to go back to literature, yet mainly based on the notion of the overall tendency of the profit rate and of economic growth. Compared to my initial periodisation, I have added some more fine grained categories towards the end, because this is where discussions will occur, about where we really are and which terms to use for which period.

I think also we need to layer somehow the 50-60 year Kondratieff cycles with the longer world system cycles.

This Timeline could also be arranged as a 'book'
Timeline 1890 - 2010

1890 - 1914
Belle Epoque

1914/18 - 1938/45
Age of Extremes

1938/45 - ca 1967/73 Fordism/Keynisianism

1971/73 - 1993 Post-Fordism, Toyotism

1993- 2000 Information society, Network Society

2000s -07/ 09 regulatory crisis, pure Financialism

201? maturity phase


better late then never

Hi Michael

welcome on these pages here. It seems you have been taking an interest in waves/cycles for quite a while. Yes, we are aware of Perez' reasons for her type of periodisations, but it is good to have it explained in such a condensed version. The question remains now, which periodisation are we going to use? Which time-spans, which names? Are we going to link them to a simgle factor such as the profit rate or a combination of economic factors or do we stay away from links to economic explanations of that sort? We have discussed various aspects, maybe now is the time to propse a periodisation scheme on which we can agreenfor now and then give meaning to those 'paradigms'


perez vs. soete

Sorry I'm late in the conversation, and you probably covered this already,

my understanding of the difference is,

- the big crises, sudden systemic shocks, are both the end and beginning of the cycles for kondratieff cycle, they go to a high point, where also a crises occurs which breaks the upcycle

- for perez, the kondrafieff first half upcycle, is the period of maturation and expansion of a previous technology wave, and when it reaches saturation, it starts to decline and the downcycle starts,

- but for her, precisely because of saturation, a new technological combination emerges, so that kondratieff's downcycle is her upcycle of uptake of the new tech as an emerging minority technology for advanced financial capital, which otherwise can only engage in speculation

- the big crises then, occurs because the old paradigm can't use the new emerging one, but at the same time, creates room for the new one to become the new dominant paradigm, and so a new upcycle can start


Goldstein and Aglietta

The table from Goldstein is very convincing. I totally agree with the new comment above, the timeline should follow the cycle of contraction and expansion and use the best economic data available, otherwise there is no objectivity. The identification of "Lead Technologies" should be tested by looking for figures on the growth of the sector in question over long periods. As it turns out, Goldstein published a piece in 2005 updating his work since the late 80s when the book was published. The predictive power of his idealized long-wave cycle is impressive:

Of course I haven't digested the book yet, but what does not seem to appear are the questions of socio-political reaction to economic change, and particularly not the concept of a regulation crisis and its outcome. It is interesting to quote Aglietta from A Theory of Capitalist Regulation:

"The regulation of capitalism must be interpreted as a social creation. This theoretical position will enable us to conceive crises as ruptures in the continuous reproduction of social relations, to see why periods of crisis are periods of intense social creation, and to understand why the resolution of a crisis always involves an irreversible transformation in the mode of production. The concept of a rupture only makes sense in a theory that takes qualitative changes into account - an indispensable condition in the social sciences, where the systems under study cannot be represented by systems of equations whose variables can be continuously differentiated." (p. 19)

It's important to maintain the tension between an econometric approach like that of Goldstein and the attempt at a holistic approach like the Regulation Theory, which is a lot more useful for cultural critique. Actually maintaining that tension is the whole point of Regulation Theory, because as Aglietta goes on to say after the quote above, his aim is to show how requalifications of the entire social and cultural system occur in order to maintain the basic invariants of capitalist social relations.

For more on regulation theory and its approach to periodization, see chapters 2, 3 and 11 of this book by Bob Jessop and Ngai-Ling Sum:

4-category book is looking great!

Fabulous to see the platform side of this unfolding. We shall have to discuss in detail the subheadings of the categories and see if they are pertinent. In fact, we'll probably have to do it twice, once at the start and then once a little later when we start to fill them with research. I assume that once we have an intial consensus, you can then duplicate the hyperlink structure, adding specific names to the periods, no?

>>comment armin: not 100% sure what you mean by the last sentence but daccord in principle

For example we might call the postwar period FORDISM (or something else) and the four metacategories Assembly-Line Production, Keynesian Welfare State, Liberal Consumerism, Cold War Hierarchy. In other words, terms that are specific to that period (which is why I don't really like Fordism....)

>>comment armin: maybe you could start changing the time-line above so we get an agreed framework first this way, then start creating chapter headings

Once we start getting to the primary categories, ideally I would like to have a way of putting up both books and images into the archive in such a way that they are behind a password, because then I think we could really put a lot of research materials inside here and make it a functioning library.

>>comment armin: on one hand it is possible to link entries with the biblio module. This are not the full books but only the references database

I think it needs to be behind a password and maybe it is even good to put one of those bot-scripts in the technopolitics archive that tells search engines not to look there, because otherwise you get takedown notices or worse problems from the copyright holders. Is this the way you are thinking?

>>comment armin: I think you are right that if we publish many images and also full pdf's we should 1) do that only behind password, which is what we have already, and which already works - just see the homepage in a logged out state, but 2) you are right that I should probably also stop crawlers from coming to the protected area, I have to learn how that works.

Concerning the dating of informationalism, or whether it even should be called informationalism, I find the whole thing interestingly complex. I have spent the day reading about Toyotism, aka Just In Time production, which ultimately gets called "lean" in America. This is quite fascinating, because it mainly involves both a reorganization of labor relations and of information flow on the factory floor (that's what Kanban is: a simple, handwritten info system to ensure continuous flow of parts and avoid any stoppage of the production line for lack of parts). At the same time the Japanese introduce new relations with their suppliers, and in this way start up the whole networked region to which I referred in earlier notes. The point being to cut down on waste while increasing the versatility of production, the number of models, customizing etc. All that is pretty well ignored by Europe and America until in the 80s Japanese cars start selling tremendously well. Then begins a big scramble to start implementing lean, which becomes the new model for the organization of factory production only in 1990 with the publication of the book, "The Machine that Changed the World." The point is that there are not too many computers involved in all that (though, I dunno, we would have to check out exactly what the so-called "autonomated" machines are like)

>>comment armin: well, I have looked into that - the last question - a bit and there are hints, both in Braverman, A&S and P&S and F&S which I have put into condensed form into the labour text, as a transition from NC (numerical control) to computerised numerical control CNC; have got a book lieing around unread by Pollock on Automation; generally it seems this is something to study more, also the just-in-time system. I have got an e-text on that which I will upload.

and yet, all of that will really take off when you get the Internet. The beautiful thing is that Toyoda (with a "d" at the time) actually started out in the 20s in textiles, and the engineer Ohno who developed the Toyota Production System claims he was inspired by the looms! Which, as you know, are historically the first programmable machines....

So it appears there are good arguments for claiming a continuity between Just In Time, arising in the late 60s-70s during the crisis of American Fordism, and full-blown informationalism with PC computers and Internet. Moreover, something very interesting is that there was the beginnings of a possible hegemonic rivalry at this point in the 80s, which was widely written about and not just by the World Systems people (I think there was a book on the Japanese Challenge, and another one on American decline by a fellow named Kennedy, yup, Rise and Fall of the Great Powers, 1987). However, in the course of the 90s the US would successfully fend off this challenge, both by developing tremendous IT advances fueled in part by research for Reagan's Star Wars program from the mid-80s, and also by the arm of transnational finance, where the US could always come out on top of any monetary crisis. For all these reasons I don't like to begin the period at the high point, in the 90s - it seems to distract too much from the way the full paradigm develops, which is quite complex and can't be reduced just to the fall of the wall or the rise of the internet.

Actually I think that all the main components of the globalized economy that is now in crisis developed in the period of Fordism's crisis: late sixties to lates seventies. JIT, the effective rise container transport, the beginnings of the internet, the industrial fabrication of medium- and large-scale integrated circuits, all that stuff actually happened during the years of bear markets and industrial decline in the core countries. It seems much more intuitive to portray that as the start of a new cycle. But of course, there are always those who will say no, the world changed in 1989...

>>comment armin: we can do different timelines with different rationales: 68 and 89 are surely important dates where something changed

The other thing is, the 90s are a high point. Whereas I must say, what interests me as a critic are the crises. I really see, not just two halves of a period, but four parts: take-off, regulation crisis, maturity phase, decline crisis. The two crises I have known in my life are the 70s and now. It seems they are also the most interesting moments too... although, I would not say that if I had grown up in Eastern Europe!

>>comment armin: I have just come across an OECD statistics on mobile phone subscribers and one on ICT exports. The first shows how low mobile phone uptake was in early 1990s, the second how dominant the US are as IT exporter in 2007. That relativates abit that picture of that waning economic superpower US. Maybe it is just very uneven, that the US are still very very strong in key areas.

Feel free to edit this comment and reply inline if you like.

best, Brian

no preference

Hi Brian

it is not the case that I have a strong preference for the Freeman and Soete timeline. Theirs and Perez' diagrams overlap, they have the same technologies and keywords, more or less, the only difference being that Perez dates the start with the invention of the new leading technology, whereas F&S put the beginning at a time when the paradigm has become fully established in their eyes. So there is a phase of overlap, where you have a mature but declining paradigm and a yound one developing inside it. So, if we want to settle for an agreed timeline, we need to decide for instance if we include a date in the early 1990s - 1993 I have put for now - where the information society proper starts or we skip that and decide for 1999-2000 as the paradigm defining moment when, according to Perez, the bursting of the new economy bubble would signal the half-time regulation crisis which a new paradigm leading technology causes, according to Perez. I only try to present the options with no strong argument for either or. It is a bit tricky in this phase, I think, as in a way we have both lived through this period and can anecdotically confirm that the information society did take off in the early 1990s, in the U.S. a bit earlier, maybe here at around 1995. This period is also important because it gave the U.S. a new image, as the country was seen as leading technologically and ideologically. Here also the whole knowledge society thing and the fetishising of innovation started fully to take off. I propose also to consider always the two halves of a paradigm, which means to consider both the 20-30 year cycle and the 40-60 (or more orthodox or simplified 50/25) and the relation to the profit rate and overall market development.

Great to talk like this now, not on email. have you tried? you should also be able the document which I started above? You can edit and write into the document, not just comment on it. Please try and improve the timeline ...

Have you seen the hypertext 'book' which I've started to grow? For now I have started a tree with your four meta-categories, which largely make sense to me (only that Cultural Horizins could be called Superstructure).

We could also, once we have an agreed timeline, make a separate tree with the chronology as make book-chapter headings and then discuss the meaning of the periods in those chapters. Do you think that would make sense?

OK, maybe I'm about to get it...

12/02/2009 - 14:16.

This is getting clearer now. I have not yet read the Freeman & Soete, and unfortunately left it in the USA, but I was able to snag this table off of g-books. So we can compare the two periodizations. What I wonder is, what do you prefer about the Freeman-Soete version? I am sure you have some good reason and more power to you, but since I haven't read the book I don't have a clue.... Also if there is a better table than this one, please post it.

As far as overlaying the cycle of hegemonic change from Arrighi and company, I think it is essential. What they show, if memory serves, is this quite amazing spatial system where hegemons and contenders face off. What's interesting is that each time, the contenders are more widely separated. So, for the periods that concern us, first you have the Napoleonic Wars and the victory of Britain, which becomes the hegemonic power. Then you have a first Britain-Germany struggle in WWI, followed by what turns out to be a USA-Germany struggle in WWII. Then you have a USA-USSR struggle in the Cold War. what this indicates, for our purposes (beyond the millions of useless deaths which I hate to just skip over in silence) is that the focus of economic innovation shifts from Britain, to a possible focus in Germany, then definitively to the US. This also correlates with something I found very interesting in Braverman's account of the development of technoscience, namely that it developed first in Germany, and that the US only really caught up after WWII (or so he says, I would say, during the war). I actually suspect that the periodization of economic development in the world system has to take its benchmark from the geographical sites of major growth. Indeed, Perez does this on the graph above. Thus WWI and the 20s may be an era of relative stagnation in most of Europe (we should verify that) but those are years of tremendous growth in the US, as the assembly-line process and consequent economies of scale really kick in and the continent-sized market is exploited. It seems to me that by establishing the chronology on the basis of this changing focus, it then becomes possible to map out the geographical differentials between core, semiperiphery and periphery, which are about lags, on the one hand, and then opposing models of society, when you get to the mid-twentieth century.