November 27, 1979

There is a need, as much for the “City” form as for the “State” form, to appropriate, to capture a war machine. And why do I say that? Because … this is one of the reasons – and I am not saying the only one – but specifically that the “City” form is not a good instrument for appropriating a war machine. … Once again, the “City” form … needs rapid wars based on mercenary action. It’s obvious that war or appropriation of the war machine can be oriented to many different paths, beginning with national conscription, from the viewpoint of men, through material investment, from the viewpoint of capital, and this is perhaps one of the principal reasons that capitalism has operated via the “State” form, and hasn’t gotten involved with the “City” form. … Hence, in fact, the extraordinary distrust both of cities as regards the “State” form, and of the “State” as regards cities.

Seminar Introduction

Following publication of Anti-Oedipus in 1972, Deleuze continues to develop the proliferation of concepts that his collaboration with Guattari had yielded. As part of this process of expanding concepts in order to produce the sequel of Capitalism & Schizophrenia, A Thousand Plateaus, this series of 13 lectures on “The State Apparatus and War Machines” constitutes the major seminar of 1979-80 and Deleuze’s penultimate consideration of these concepts. Deleuze first considers material begun during the previous year’s seminar, material corresponding to plateaus 12 (1227: Treatise on Nomadology – The War Machine), 13 (7000 B.C.: Apparatus of Capture), and 14 (1440: The Smooth and the Striated).

Deleuze then returns to the history of philosophy with five sessions on Leibniz, after which he concludes the academic year, at his students’ request, with two final sessions of reflections on Anti-Oedipus.

English Translation

Edited

Building from the previous session, on the possibility of coexistence within a social field of extremely diverse social formations and of defining these through the commodity of “machinic processes,” Deleuze emphasizes the importance of Braudel’s research on relations of city and State form. He recalls the different processes coming into play (e.g., warding off-anticipation), and continuing to explore the possibilities of exchange between primitive groups, he cautions against including certain elements (e.g., stockpiling, currency, markets, regime of labor) not yet available to these groups, proper only to the State form. Deleuze proposes to consider the development of the concept of labor and labor-time, distinct from the primitive formations of exchange, in the State formation of exchange. In this process, he develops the concept of “the utility of the last object” from economic marginalist theory, with distinctions between the “limit” and “thresholds”, and the concept of “the last” provides a means of understanding the movement within social groups to maintain or be required to transform the group “assemblage” in relation to the “last object” obtained in the exchange. Considering how objects exchanged may be compared under the theory of labor-value, Deleuze discovers in the classroom a student who can speak about these issues from the perspective of accountancy. While the student prepares to develop some comments, Deleuze considers the marginalists’ theory of evaluation and trial and error in contrast to advocates of labor-value. In short, the “last” in the sense of marginal does not mean “ultimate”, but really “second to last”, since the “last” as ultimate would be the first in the other assemblage, leading Deleuze to conclude that while the limit is what is anticipated, the threshold is what is warded off. Seeking additional proof of this series of exchange, Deleuze turns to the student in accountancy, and the recording ends as the student’s inaudible presentation begins.

 

Gilles Deleuze

Seminar on the Apparatuses of Capture and War Machines, 1979-1980

Lecture 04, 27 November 1979

Transcribed by Annabelle Dufourcq; augmented transcription, Charles J. Stivale

Translated by Charles J. Stivale

 

Part 1

Concerning our next meeting, I am saying, eh, next time, that at the request of many of those who were working with me last year, who wanted us to do a restricted session, not at all for the purpose of excluding anyone, on the contrary, quite the contrary, uh … but because we will come back to certain notions that we had developed the next year… uh, last year and, as we will assume they are familiar, this will be like a kind of session that really cannot be of interest, I believe, which can only be followed by those who did the work last year. So, the others, if they want, I’ll meet them the week after, a way of returning the favor, on the other hand, for “precursor” sessions which will be only for those … [Laughter] … There you go. I don’t see how, but … Fine.

Some students: [Inaudible]

Deleuze: About that, the last time, we had made some progress, not much, regarding …, following the traces of a topic that we needed, namely the possibility of conceiving of  a coexistence in a social field, a coexistence and an encroachment of very diverse social formations. And we assumed as a hypothesis the possibility of defining the various social formations by processes which we conveniently called “machinic processes”, and we had identified the coexistence and the interpenetration of these processes. So, no attempt had been made at all now to speak of societies, for example, primitive insofar as being social formations, but we had identified a process which, no doubt, appeared in so-called primitive societies, but was also appearing elsewhere and where this process was called “anticipation-conjuration”.

We spoke more exactly of the State apparatus, no longer as a social formation or state society; we spoke of the apparatus of capture, another process. We spoke of war machines, we spoke of … etc. [Deleuze does not finish] And we had outlined this social field where, then, when social formations are no longer related to modes of production that they would be supposed…. er… that they would be assumed to suppose, but are related to the processes that they envelop, the very idea of ​​a coexistence of all the formations in the social field seemed to us to be… to be verified, to be established. This is where I am asking if there are any points … or else if I might continue …, but I believe there are some points.

A student: [Inaudible]

Deleuze: What? Yes, yes. Go ahead, talk. Yes.

A woman student: It’s about the city and the State.

Deleuze: Ah, on the city and the State. We had indeed seen, I recall, that in these social formations, we had felt the need to distinguish the “city” form and the “State” form because it seemed very important to us historically to understand, for example, why capitalism had gone through the “State” form while the “city” form, in its beginnings, had given it so many chances and so many opportunities. [Pause] What’s going on? And we had tried to distinguish the “city” form and the “State” form, whatever the competition between the two, whatever the way there was in which the “State” form would later overwhelm the “city” form. We tried to distinguish, here as well, two very different processes by defining the “city” form through the constitution of instruments of polarization within a network, any city referring to other cities, while the “State”  form seemed to us to refer to some very, very different systems, systems of resonance that we had tried to… to distinguish precisely from instruments of polarization.[1] So, on this topic, did you want to add something?

The same student: Yes. Anyway, there is someone who has done work on Paris in particular and, from the 10th-11th century onward, and has been interested in seeing how Paris has resisted… er… well, the State, studying this up to 1980. So, I will just read you a page where he talks about Paris currently:

“The French villages did not want a train station, so they had to be built 2-3 kilometers from the church in the open fields. For a long time, there was no continuity in the construction (tissu) built between the old town center and its station. Then, little by little, around each station, the central zero began to secrete a rectilinear coagulation of foreign activities. This hiatus is still often noticeable; this is the story of the large number of Sainte-Emilie-Center and Sainte-Emilie-Gare that dot the French countryside. Paris has shown a similar reaction. The battle between the city and the State raged for fifty years and, in other forms, it has persisted until today ending and shutting itself down in the metropolitan rail node which is the process of being completed.”

That is, in fact, he explains, afterwards, that the city of Paris had widths for the rails, finally the gauge was at 1 meter 30 and that the rail network gauge was at 1 meter 40, and that it’s on this point that all the resistance occurred and up to the present. And he says that it is only now, that is, currently [late 1979], that the network of Les Halles has just connected the train system to the metro system, with the resistance ending; that is, what had started several centuries ago, but, by and large, particularly during the time of the French Revolution, that is, the struggle between Paris and the State, has now ended with the junction of the RER in Les Halles with the metro.

Deleuze: Yeah. Indeed, it would be very, very important …

Student: It was published …

Deleuze: … and I think they must be from the school of [Fernand] Braudel, right?… I was saying: it is really Braudel who, it seems to me, at least in France, launched this extremely deep study into the at once polemical and competitive relationships between the city and the State which really constitutes something fundamental in the history of Europe, right? Braudel is surely correct in saying, in Asia, it did not happen like that because, in Asia, there is a subordination …, in the history of Asia, there is a much more direct and fundamental subordination of the city to the “State” form. But, in Europe, there was a kind of wavering, right, and really, I believe for the birth of capitalism, it’s a huge problem to wonder about, in fact, with all kinds of data, who outlined the formation of capitalism according to the “city” form.[2]

And, once again, I recalled this sentence by Braudel, because it seems to me very good, that he states: each time, there are two runners … each time, there are two runners, and he says: it is the hare and the tortoise. And, of course, it’s this is the city that’s the hare, right? And it is the “State” form… the hare, it’s the “city” form, that is, it was at the highest speed and, for us, we can even add that we gives the word “speed” a very precise meaning. It is not at all a metaphor; it is really the speed of deterritorialization. There is a power of deterritorialization of the form… in the “city” form much greater than in the “State” form… uh… I was trying to say this the last time… because in the city, deterritorialization is truly dynamic, while State deterritorialization may not be less so, but it is static.

And we tell ourselves: but, in fact, what would capitalism have been and how did it take shape when it, and to the extent that it did, passes through the “city” form? What made it need the “State” form? Or what occurred to make it choose the “State” form? This will be very important for us to try to clarify this question. Here, I am saying immediately one of the hypotheses that I see, which seems to me to be one of the most important — I do not know if I had stated it — it is closely connected to a related topic, which we will also encounter, which we encountered last year, namely that, in any case, there is a need — both for the “city” form and for the “State” form — there is a need to appropriate, to capture a war machine.

And why am I saying that? Because at that point, maybe, everything just becomes a little bit clearer. It seems to me that this is one of the reasons — I am not saying at all the only one — but that the “city” form is not a good instrument for appropriating the war machine. That’s the “State” form, because it is the “State” form that can make the investments in so-called warfare “materiel”. The “city” form has much less of this possibility. Once again, the “city” form, it’s not that it does not generate and does not have…, that it does not appropriate its war machine, but it essentially needs rapid wars, it needs rapid wars based on mercenaries. It is obvious that war or the appropriation of the war machine will be oriented toward entirely different paths, starting with national conscription, from the point of view of men, by material investment, from the point of view of capital, and that it is… that is perhaps one of the main reasons that capitalism went through the “State” form and did not limit itself to the “city” form, which implied that States… well, as Braudel says, that the hare… that, uh the… the tortoise catches up with the hare. The “State” form had to assert its size, its power apparatus — which is not the same as that of the power of cities – the State apparatus had to wield this power apparatus over the cities. Hence, in fact, what the text you have just read confirms, the extraordinary mistrust both of cities in relation to the “State” form and of the “State” form in relation to cities.

The same student: Yes, finally, but that is especially related to circulation precisely. Because … they …

Deleuze: Yes. Yes. Completely, since cities, insofar as … as they are instruments of polarization within a network, the city refers to another city or to other cities; it is essentially thought in terms of entrances and exits, essentially, while the State is not that. Of course, there is an entry and exit for the State; I am saying: this is not the main thing.

I want to say that, because, you understand, we will need to remember this point, that when we manage to define social formations by processes (processes of capture, processes of… anticipation-conjuration, etc.), we must see that these processes sometimes play out in a pure state … I can say a primitive society, okay, or a gang — it is already not the same thing — but these are social formations in which really the anticipation process… uh… the anticipation-conjuration mechanism trumps all others. But that is a first point of view.

A second point of view, we must see how each formation not only envelops the preeminent process that corresponds to it, but adapts, adapts in its own way, the other processes. For example, you will also find anticipation-conjuration processes in cities, in the “city” formation, to ward off the state. You will also find anticipation-conjuration processes in capitalism to ward off the limits of capitalism.[3] So, if you will, each process plays a preferential role in a certain type of formation, but it can very well be taken up in a completely different formation. That’s why it creates a social field where everything necessarily coexists. You follow? Well. Good. So, we’ll continue.

Another student: I have a question about last week.

Deleuze: Yeah!

The student: When we were talking about the last word, the last drink, the last love …

Deleuze: Yeah!

The student: Maybe we could add: last assemblage in Beckett for …

Deleuze: The last assemblage in Beckett, yes, why not? We’ll see where we can put “the last” in Beckett. Yes, there is a last love in … which he weirdly calls first love, yes, there is all of that, yes. Good.[4]

The student: And then, for [D.H.] Lawrence, I think you said once that you even have to [Inaudible] love. You see?

Deleuze: Ah yes, this is quite simple, yes. What he means… It’s not all that simple, right, but what Lawrence means when he says that, fine. Well, we’ll talk about all that.

So good, this is a new topic we are tackling today. I’m just asking you to sense that these topics are very much a continuation of each other. Here, we find ourselves in the following situation: we tell ourselves that our task, today, should be to build even a very abstract model, an abstract model that we could call a “model of the primitive exchange”. In the background, we are concerned that, nonetheless, among all these mechanisms that we spoke about the last time, the mechanism that we saw fit to call “anticipation-conjuration”, both interests me, but it was still not left all that clear: what are these collective anticipations-conjurations? So, would a model of primitive exchange allow us to move forward in this direction and, above all, would that allow us to strengthen our hypothesis of a coexistence of the most diverse social formations in a given field? And the basic conditions … So, we’re going to do it like that, uh … if you will: we are proposing, let’s say … uh, I guess we are proposing to build this abstract model of primitive exchange.

At the point we find ourselves, we know what we cannot allow ourselves. What can’t we allow ourselves? Uh! And what must we allow ourselves? Well then, by definition, we have to allow ourselves groups that are in a certain communication with each other and, in fact, — I’m not going back over this — we do not believe, it does not seem necessary for us to presuppose what is classically called the autarky of small primitive societies or their independence, or their incommunicability. One assumes that primitive groups can be entirely in relation with one another. Moreover, I assume that it’s inevitable. It’s inevitable because primitive formations already coexist with States, with apparatuses of Empire, that these apparatuses of Empire imply and put the primitive groups into communication with each other, so we have every reason to think that there are expressions of primitive exchange and not independent groups.

Moreover, I tell myself: there is communication between foreign groups as soon as there is, not even writing (writing, that would refer to the State apparatus, to the apparatus of Empire), but as soon as there is speech, as soon as there is language… uh… Something makes us anticipate that, in the end, language, I understand this to include oral language; it does not exist simply for those who speak the same language. I would even say more that language only exists to the extent that there are already contacts between people who do not speak the same language. I mean: language is inseparable from a function of translating languages ​​and not from a function of communication within one and the same language. Anyway, all that…, we offer ourselves this minimum: primitive groups in relation to each other.

But there are things you cannot allow yourself. What is it? If we assume that there are primitive exchanges, we cannot allow ourselves, first, the coexistence of a stockpile. I am stating the conditions of the problem to me, right? I would have no problem with some of you saying: ah well no, the problem for me is otherwise. Me, I am stating how it occurs for me. I cannot allow myself the pre-existence of a stockpile for a very simple reason, which is what we saw the last few times — I won’t go back over this — that, far from the stockpile presupposing a surplus, it was the stockpile which constituted the surplus, and the stock was an act of the State apparatus.

So, if I’m looking for an expression for primitive exchanges that coexist — let’s understand this well, which coexist with State apparatuses, but which do not constitute State apparatuses, since they will both anticipate and ward off, you remember, State apparatuses — I cannot assume that primitive exchange implies a stockpile. At the most, I can say that it implies an elasticity of supply and demand. That’s enough for me, yes. There is a certain elasticity of supply and demand, that is, sometimes they eat more, sometimes they eat less, good, so they can exchange food, etc., but I’m not demanding any … any pre-existing stockpile. I’m just asking for some evidence (données) of… say — I’m using a common political economy term — elasticity of supply and demand. So, about primitive exchange, I am saying: impossible to allow oneself a stockpile for [inaudible word] expresses it.

Second thing: impossible to allow yourself a balance or a monetary equivalent (or of another type) for the same reason. Once again, these primitive formations coexist with the State apparatus but involve other processes. Now, we have seen that, no less than the stockpile was an act of the State apparatus, the market, currency, are acts of the State apparatus. And, in this sense, we have already questioned without yet sufficiently justifying it, we have questioned the idea that money can find an origin in trade, that is, in generalized forms of exchange, in order to say: no, in any case, if there is an origin of money, it is on the side of taxation, that is, of a fundamental State act, which we have to research. So, no way to provide yourself: balance, equivalent or market, or the existence of a market.

Finally, third: no question either of allowing oneself the hypothesis of an intervention, in the primitive exchange, of a labor or of a labor-time, of a necessary form of labor or of a socially necessary labor-time for the production of the objects exchanged. Why? Because, there too, last year we had already developed this, but we will have to discover it again under another aspect, this idea, for a very simple reason: it’s that these so-called primitive formations do not work under the regime of labor which is a very special regime of activity. Last year, we tried to clarify this. For lack of anything better, we had found the expression of “continuous action” or “activity in continuous variation” which we opposed to the activity of the labor type.

But, in fact, it seemed to us and it will be better … it will be better to explain it, perhaps today, it seemed to us that, in any case, that labor was not a naturally determined activity, but was a very special determination of activity, a model to which one subjected activity, and this “one” seemed to us once again the State apparatus, namely, it is the State apparatus that subjects activity to the “labor” model. So, there is no question of invoking a labor-time that would serve as a possible comparison criterion between exchange … between objects exchanged at the level of a so-called primitive exchange.

So, you see, I am continuing my completely abstract hypothesis. And so there arose, to help us, a hypothesis which I must say right away, it is curious that it occurs to us here. I was saying, well yes, there are economists who have said: value cannot be explained either by labor or the labor-time necessary for the production of the object, nor by the utility of the object… — so you see they are opposed both to use-value and labor-value — and they state this mysterious expression: value refers to the utility of the last object.

So, I say right away: what is curious is that this thesis — the last object being the marginal object — is well known in political economy under the name of “marginalism” or “neoclassicism”. and that this theory had and still retains a fundamental importance.[5] I am saying that it was obviously designed to take account of capitalism and capitalist market. But theories have adventures… like… like everything. It would not be improbable that, for example, a theory invented for a particular sector to account for certain phenomena of the capitalist market, notably of the balance of prices under a capitalist regime, shifts direction and is discovered to have a field of application in non-capitalist formations. So, we can always ask this question: does the marginal object, the idea of ​​the last object, not find a … a very curious application in so-called primitive formations? Won’t that help us? That’s my question, in our hypothesis.

I’m quoting a … text from a textbook, not at all by a great economist, from a textbook that explains marginalism. I will read slowly, ok? It’s a question … this time, it’s a question of labor and productivity, not about the product; it is a question of the productivity of labor and the worker. And this is what the author says: “Let it be a sheep farmer …” — see, it’s quite simple, the example must be very clear, right? — “Let it be a sheep farmer who wonders if his team of shepherds is sufficient. He can see that if he hired one more shepherd (without making any change in his tools or in his constructions)” — he puts that in parentheses; I’m emphasizing it, right, emphasizing it for the future, for my future. – Take a close look at the conditions: you have the entrepreneur, right, the entrepreneur-breeder who tells himself: do I have enough shepherds? This is the question. And he tells himself: maybe I could hire, that is, pay one more shepherd without making any change in my tools or in my constructions. We should add: neither in my lands — you are aware that a sheep is so many… I don’t know, a cow is an acre, a sheep is…, I don’t know … uh … but the land area is also supposed to stay the same. – So, without making any changes, can he take on one more shepherd? This is a question that all … To understand, you have to put yourself … uh … all that we will have to say, later, about capitalism, you always have to put yourself in the place of a boss. So … there you go, the sheep boss, ok? Do I hire one more person, once it’s said that this is a matter of not changing anything in your business?

Sense that we already have something. I would really like to go slowly today and have you follow me… very closely and… uh…. A notion of threshold begins to emerge. All bosses know that; all bosses say that. There is even a famous rule which is something like the rule of fifty. There is always a threshold in a business. Bosses know very well that, beyond a certain threshold, the structure of the company must be changed. For example, there is a threshold at which if you hire one more person, you, as the boss, have a labor council on your back. You will tell me: this does not matter, it’s okay … Yes, no, uh … Fine … uh … There are also thresholds in which the accounting can no longer be done the same way. You have to change, I would say… [He does not complete this] There are thresholds beyond which the whole assemblage has to change. So, notice that my sheep boss, there, he is asking himself: can I hire one more shepherd without changing my assemblage, that is, without increasing my property, without changing the constructions, etc.? See the problem?

I’ll continue. “He can see that if he hired one more shepherd without making any change, the flock would be better looked after, the number of lambs could be increased, and they would be able” – that is, he would — “he would in this way be able to send twenty more sheep to market every year.” Let us suppose he realizes this: without changing the assemblage of my business, if I hire one more guy, I can have twenty more little lambs to send to the market. You see, right? You follow me? If you don’t understand, you won’t understand the other examples … it’s not … So up to this point, it’s okay.

“The net product”, the net product “of this additional shepherd’s labor will therefore be in the quantity of twenty sheep”. And, right, since with the additional shepherd, the guy, without changing anything, can have twenty more little sheep, the net product of this shepherd’s labor will therefore be, in quantity, twenty sheep, and in value, and in value, what the price these twenty sheep will be worth at market. This is all crystal clear. “If the breeder can hire the additional shepherd for a salary that is somewhat lower” — that is, a somewhat lower than the price of twenty, the price of twenty sheep minus x — “If the breeder can hire the additional shepherd for a lower price, a salary even slightly lower than the price of the twenty sheep, he will. Otherwise, he will refrain from hiring him.” These are… these are basic rules for a business operation.

“This shepherd we’re about to hire …” — You see, right? — “This shepherd we’re about to hire …” It starts … the text starts to get interesting here, right? There, in the boss’s head, he is evaluating. He’s evaluating: can I hire a guy? Is he going … uh … is he going to earn twenty sheep for me? Can I pay him a little less or a lot less than the price of the twenty sheep at the market rate? — “This shepherd that we are about to hire is the marginal shepherd, he is the borderline (limite) shepherd.” Notice why he is marginal or borderline: because, if this last shepherd is hired, and if we hired another shepherd … another shepherd after the first one, then the assemblage would have to change. This is why the parenthesis is essential. This is the borderline shepherd, given the assemblage being considered. The guy can always hire more shepherds; at that point, he will need new land, new constructions; he will have to change the nature of his operating assemblage, the nature of the business. So, understand, here we are at a last shepherd in the sense of: the last one before the assemblage is forced to change, the last one before the assemblage is forced to change. This will be called “the marginal” or “the borderline object” or “the borderline character”. — Fine, is this OK? You stop me, huh, if this isn’t working — He will receive a salary approximately equal to the net value he adds to the total product, that is, less than the twenty sheep he brings in, less to a determined extent. [Pause] Uuuhh, there you go.

But if we assume that all the shepherds in the operation, this one, the marginal shepherd — this is a beautiful notion, the marginal shepherd, the morning shepherd, it’s curious, that … [He does not finish this] Okay then, uh, this must already, you sense that we are going to come, that we are going to come back to our problems that I mentioned the last time, more concrete ones there: the café, the last drink. But I tell myself, the worker of the last hour, we will … we will have to go and see, will have to return to this parable, there, of the worker of the last hour and see if, not at all if this is marginalism — that’s not how we work — but if there is a way to make, not at all an assimilation between a text uh … of the New Testament and a text of political economy, that would be of no interest, but if we cannot jump from one to the other, to create ruptures that will enrich this notion of “last” … —

But finally, there you have it. If we assume that all the shepherds of the operation, this one and those who had been hired previously — that is, the whole series, the whole series of successively hired shepherds all the way to the marginal shepherd — if we assume that all the shepherds of the operation, that one and those who had been hired previously, are interchangeable, that is, if we do not allow ourselves the hypothesis of a magic shepherd (berger génie), if they are interchangeable, we must think that all will necessarily receive the same salary. Notice what this means: if you’ve understood this phrase “everyone will necessarily get the same salary”, that’s great. He’s telling us … And by what right can he tell us that? He is telling us: The salary of previously hired shepherds depends on the evaluation of the salary of the marginal shepherd.

So, you will tell me: but the marginal shepherd is not yet hired. Obviously, he’s not yet hired. Across the entire series, the entire series is determined by the businessman’s evaluation of the marginal shepherd’s salary. And the salary of the shepherds, of the previous shepherds, across the whole series, will be determined by the salary of the last of the possible shepherds. Once again, how can I say [Deleuze coughs and chokes] “the last possible object”? I can say this because I am calling “the last possible object” …, uh, no, “the last possible shepherd”, the last shepherd before the assemblage is forced to change. One more shepherd, besides the last shepherd, the assemblage is no longer possible; another assemblage is required. Therefore, it is the evaluation of the marginal shepherd’s salary that determines the salary for all existing shepherds. Do you understand? Fine.

It’s no wonder that it was the English who had these ideas; it’s very, very … it’s very … it’s very funny. And, therefore, this author who only summarizes the marginalist theses can conclude, this is what interests me: “the productivity of the marginal worker …” — namely his power over twenty little sheep — “the productivity of the marginal worker thus determines not only that worker’s wages” — quite simply – “but that of all the others.” Fine. There, we chose the problem of productivity and wages. And he adds in brackets: “In the same way that, when it was a question of commodities …” — and undoubtedly, at the level of the commodities, this is even clearer; I therefore chose a more difficult case, so, if you understand it, you understand this all the more for commodities. “In the same way that, when it was a question of commodities, the utility of the last bucket of water …”, “the utility of the last bucket of water or the last sack of wheat governed the value not only of that bucket or that sack but of all the other buckets or all the other sacks making up the stockpile.” We’ll cross out the last bit: “making up the stockpile” [inaudible word], since it doesn’t fit, it doesn’t change a thing. “The value of all buckets of water and of all sacks of wheat is governed by the marginal object, that is, by the determining value of the last bucket of water and the last sack of wheat.”[6]

You will tell me: in that case, we vaguely understood what “the last” was in the case of the shepherd, but we understand more, perhaps, in the case of the bucket of water and … [Deleuze does not finish] Well, if we have already understood about the shepherd and we see that the marginalists began by undertaking their analysis not at the level of productivity and labor, but at the level of the commodity, so it’s the utility of the last object that determines the value of the whole series — that’s it, if I sum up, right? It’s the utility of the last object that will determine the value of the whole series and of each term in the series. If we say that, we tell ourselves: well, we’ve got hold, we’ve got hold of a start.

What can…? And, in fact, we have a break, and so, we tell ourselves, we have to catch our … breath; let’s forget, let’s forget about marginalism, or let’s pretend to forget it. And we say: the topic, this marginal thing, is really odd. The last one determines the value of the whole series. What can I say about that? That, in fact, it needs to be corrected: it’s the idea of the last one. The idea of ​​the last one determines the value of the whole real series, of the whole series of real terms. Why? Because we saw it quite well, and there it was clear in the example of the shepherd. It is not when the last shepherd, the marginal shepherd, is actually hired that he will become a determining factor of the salary, it is the salary of all the previous shepherds actually hired which is determined by the idea of ​​the marginal shepherd’s productivity, that is, of the last shepherd that can be hired without causing the assemblage to change.

In other words, there we already have something that matters to me enormously, it is… I will then formulate my hypothesis more precisely: there would be collective evaluations… — we are in the middle of the social field, in the middle of… in the midst of collective formation — there would be collective evaluations which would be of an anticipatory nature. What would they anticipate? They would anticipate the limit. — There we… we… we progress suddenly; we progress in leaps and bounds, even if it’s abstract. — They anticipate the limit. They anticipate the number of terms necessary to reach this limit, and they anticipate the time taken to reach this limit. Fine. [Pause]

Why is this collective evaluation anticipatory? After all, it is a philosophical subject that matters to you; well, even from the point of view of the history of philosophy, I am quoting for the record a very beautiful text by Kant that some of you know, on the anticipations of perception. And, in this text, Kant tries to show that perception has a structure such that it comprises at least … – perception being entirely related to experience, one perceives only something which is given in experience, otherwise we imagine (on conçoit), we do not perceive — well, that there is nevertheless a given that Kant calls a priori, that is, independent of experience, which intervenes in perception, only one. And it is this datum that constitutes the anticipation of perception. And this datum is that whatever it is, perception necessarily has a quantity, an intensive magnitude. Fine.

I am saying: maybe we encounter, on a whole different level, but maybe that… that, there too, the intersection will be created between this problem of anticipation in judgment… and, there, we grasp onto, really under a completely different aspect, we grasp onto a problem of anticipation in collective evaluation. And we say: collective evaluation is… — and at that point, we will call it, even if it means then saying “oh no, we were wrong, that was only one case; there are other collective evaluations …” — but for the moment, I can say: I will call “collective evaluation” an evaluation which bears on the idea of ​​a last object, of the marginal object, and therefore, on the number of terms in the series in order to arrive at this last object and the time necessary to arrive there. You see?

Why does that interest me? That interests me enormously since the other years, uh… I would find something like a confirmation there, but which, in previous years, I could not think through since… [Deleuze does not finish] In previous years, for example, we have not ceased to return constantly to the notion of assemblage, very often. And I was trying to say: the assemblage, an assemblage, which seems to me better than the notion of behavior – in the end, we assemble (on agence), we do not behave –, and well… it is very different, as a notion, at the same time, and well, I was trying to say: an assemblage always has two sides. It has like, very roughly, it has one side, and these two sides, one does not depend on the other; they are in reciprocal presupposition. Fine, every assemblage has a physical aspect or even, one should say, to harden the word, physicalist aspect, and it has a semiotic or semiological aspect. That is, it is both a machinic assemblage and an assemblage of enunciation. Why am I going through this parenthesis? Because, there, I have a very … [He does not finish] And these two aspects are not at all symmetrical, they do not correspond at all term to term. In any assemblage, you will find a system of wording and statements, and you will find a system of things and mixing of things.[7]

And there, in the primitive assemblage, where … that I’m dealing with right now, what do I find? I indeed find the aspect, how to say, a physical, physicalist “thing” (‘chose’ physique, physicale) namely the series of men, shepherds, or objects … [Interruption of the recording] [46: 20]

Part 2

In this case, what would I call the “speaking system (système d’énonciation) of the assemblage”? Precisely the aggregate of collective evaluations bearing on the … idea of ​​the last object. And don’t wait for the last object to be there. You do not wait…. The boss does not wait to hire the last possible shepherd so that his evaluation of the last shepherd’s productivity determines the actual wages of all the actual shepherds. Right? Fine. Well, this is it, this is the collective evaluation. [Pause]

And there, I am indeed bringing up a problem. What problem am I bringing up? Because, compared to the outline of the theory of labor-value, here I … this will become more concrete in a little while, ok? … You have to put up with me … You have to … Compared to the theory of labor-value, notice that there is a common problem with which political economy — and this is even what makes it interesting in my opinion from the point of view of a theory of statements or a theory of utterance — political economy inevitably encounters this problem, whatever its conception of value might be, namely: how is the evaluation of the means of exchange or the criterion of exchange established? There has to be a collective assessment. I would say that a fundamental and insufficiently outlined chapter of political economy is: whatever school to which one refers, it is that it [the school] necessarily includes a theory of collective evaluations which I would also call “a theory of the anticipation of social perceptions”.

Fine. Uh … So that would work just fine, because in fact, the advocates of labor-value, what are they telling us? They assume that the means of exchange for objects refer to the so-called socially necessary labor-time to produce objects. You see, this is an extremely clear thesis. Once again, there, to credit … to credit Marx with this thesis is nonsense, not that Marx does not accept it, but Marx does not claim to invent it; on the contrary, it is the most classic old theory of classical political economy. Marx’s innovation is quite profound, but, precisely, it is not there, right? Good.

We are told: if there is therefore an exchange of objects, it is because there is indeed a way to compare the objects exchanged, you see, in the theory of labor-value. And this way to compare the objects exchanged is to compare the socially necessary labor-time for the production of object A and object B. I assume that object A takes double the time; it will be worth two objects B. Okay, very simple. You see that the theory of labor-value nonetheless implies …, we do not assume that these societies have, for example, systems or even a state of labor which is mechanized labor. [Pause] Now, if we do not assume a so-called scientific, pseudo-scientific mechanization and quantification of labor, the socially necessary working time immediately implies and refers to a collective evaluation of both the worker and the entrepreneur and the collectivity itself. Ah yes, that takes … Uh … you use … with a steel ax … uh … they are not going to take measurements anyway … With a steel ax, you do twice as much … or three times more work than with an iron ax … This is a collective evaluation which relates to labor-time. It assumes the labor regime. Fine.

For reasons that we have seen and others that we will see – once again, I cannot say everything at the same time — we have deprived ourselves of this possibility in the case of so-called primitive exchanges, that we called by convention primitives, since we are saying: in this, there is no socially necessary labor-time since the activity there is in continuous variation, therefore there is absolutely nothing that corresponds to labor-time. However… However, here I am quoting for the record, so that…, out of a concern…, of course, I know that there are certain ethnologists who have tried to apply criteria, even very quantitative criteria, of labor-time to primitive societies. It’s very odd that even those say that it has no correspondence in the group consciousness, that it has no equivalent. We can always apply it, but … Fine, I am thinking of an Australian who developed much of this kind of research, saying: well yes, but… it does indeed work; it works, it is true, but there it is, it does not matter. For reasons that I have tried to state or suggest, we cannot think that collective evaluation has a bearing on labor-time.

Moreover, one of the most developed texts in this respect is found in Engels, in Engels, the preface that Engels wrote for Book III of Capital, Book III of Capital, which is not published by Marx himself, is published by Engels, and Engels appends a preface to it. And, in this preface, is, I believe, one of the most precise texts in which Engels says: once it’s given, once labor-value is stated, how in a very primitive society can the evaluation of labor-time occur since he is an advocate of a labor-value theory? And his answer states that, well yes, there is a kind of collective evaluation in the mode of anticipation… — this is a very curious text; well, those who are interested, you will look it up; I will pass it around at the end of the session — he adds: otherwise, those involved in exchange would not meet their expenses. Grant me… uh… I’m sure you’ve understood all this, why should that last sentence, “otherwise, those involved in exchange would not meet their expenses” put us into a state of joy that I am not adequately expressing? [Laughter] That’s because, maybe, you will surely feel this, but it comes close to something, specifically he is in the process of reintroducing a marginalist criterion.

If we try to comment on “otherwise, those involved in exchange would not meet their expenses”, how can we define the cost recovery independently from the reference to any other type of evaluation? Won’t the respective assessment of costs not allude …? [He does not complete this] In any case, notice in what state we find ourselves: we are saying, in certain cases, namely where there is no State apparatus, in so-called primitive groups, we made a big leap; we were right to talk about an anticipation mechanism, because it works, in fact, in the form of anticipation. There is an anticipatory collective evaluation. And we specify: what does this anticipatory collective evaluation consist of? And we answer: this anticipatory collective evaluation consists in this, that it anticipates the limiting idea of ​​the last object or of the last producer and fixes the value of all the terms of the series and the time necessary to exhaust the series — implied without the assemblage changing — and fixes it according to the idea of ​​the last object.

So, we answer, no! Not necessary! The collective assessment does not have to have a bearing on labor-time. Collective evaluation may well have a bearing on the idea of ​​the limit object or marginal object. You see? Confirmation, like the confirmation of rest. — I am sensing that, in a little while, we are going to take a break because … You tell me if you can take more because … I mean, that bothers me … For me, I find this fun… Fine, this, this puts me in a great state of joy, but for you, it is not necessarily the same thing. Uh … And sometimes we can talk about things that bore me fully and that are fun for you, you never know. That’s how we should distribute UVs [academic credits]; I’d ask “who’s having fun with this?” Anyone who would say that they’re having fun, well … [Laughter]

A student: [Inaudible] … accelerate growth without changing the structure.

Deleuze: That’s it, that’s it, that’s it. of course. I’ll summarize what he’s saying, ok?

The student: [Inaudible]

Deleuze: Yeah, yeah, yeah. Absolutely, absolutely, the bottom lines are completely … Yes, yes, he’s pointing out that, in fact, in any company, precisely what is called accounting operations and. in particular, the exercise of the balance sheet, the preparation of the balance sheet precisely relies on this whole system which implies and which is always created as a function of a threshold beyond which the entire company would have to change its structure, in particular the global payroll, I suppose, what we call the global payroll. Here I think there is someone here who would in any case be much more knowledgeable than me to talk about this, uh…. There we are.

The student: What’s important is that [Inaudible] in general accountancy [Inaudible] to accelerate growth without increasing costs.

Deleuze: Yes, of course! Without changing, it’s not so much to increase costs, it’s worse than to increasing costs …

The student: It’s time [Inaudible]

Deleuze: That’s it.

The student: [Inaudible]

Deleuze: You are right. You are all the more correct since the importance here, indeed, of this accounting, in fact, causes the factor of duration to intervene. There is the factor of numbers of elements, the numbers of elements in the series; what we are doing is … this is a theory of … of what we could call serial groups, but we would call that precisely, borrowing the expression from Sartre, but giving it a completely different meaning. These are serial groups and serial operations that cause a number of terms of the series to intervene, the limit object, idea of the ​​limit object, and time required to [do something]. And, in fact, he is right to insist on the temporal factor in accounting, to the point that one would imagine a thesis: the idea of ​​time in accounting, there yes, that would be a beautiful thesis … It would, right? I think so, right? Uh … Certainly …, but finally … what would have a lot to say would have nothing to say, but that, it’s true. Have you studied any accounting?

The student: [Inaudible]

Deleuze: You know what you’re going to do, right? [Laughter] Well, what? This is … Uh … Would you be ready [to present], here?

The student: Sure.

Deleuze: Would you be ready right away? That’s going to be great; I’m going to continue presenting my concrete examples. If you even like to go next door to think a little bit, how just to explain to us… — because, there, that would interest me greatly — to explain to us in about fifteen minutes, I’ll have someone find you… [Laughter] You could explain to us, precisely, the time factor in accountancy, in business accounting. Could you do that?

The student: Yes, I can …

Deleuze: Can you? So…

The student: Not right away; it would be better that way; I … [Inaudible]

Deleuze: In my opinion, this is where you are inspired, this is where you [Inaudible words] [Laughter] So precisely, or you cover your ears, you cover your ears, eh, you just don’t listen, you … re … Or you go outside and … uh … so … just to give yourself the time [to prepare].

So, we’ll forget that, ok, because it’s … it’s just a thing … I am saying, there’s a second point. Here, my first point was, you see, this collective evaluation which is, from the beginning, ah yes, I am emphasizing this, because … there is, for those who would like to extend this then in a philosophical direction… you see the extent to which it is… — this is good, we have an accounting direction thanks to him — I was thinking of a philosophical direction; we would have to start again. Those who know Kant, Kant’s text, in my opinion, is completely wrong [inaudible word] anticipation. In fact, if there is an intensive quantity of perception, that’s because social perception is fundamentally serial and works as a function of the limit object… uh… but… So, it is for another reason than the one that he believes … Uh … No, one shouldn’t say that: Kant is never wrong, he guessed all that [inaudible words] …

So, understand me … What did I mean …? Yes, there is… All marginalism has a completely crazy theory because they’re all crazy, really, all of them… uh… completely crazy; this is precisely the theory of evaluation and trial and error. And you see that, in a way, if we pursue this, it becomes quite funny since, here, the opposition between the proponents of labor-value and the marginalists takes on a very, very concrete meaning. They both agree, once again, in saying: there is a need …, nothing would happen in the social field, neither productivity, nor exchange of products, if there were no anticipatory mechanisms of collective evaluation.

Good, that interests me a lot; it’s not about planning, it’s not about … It’s much more about concrete things, much more general stuff in societies. Simply, the difference is that the advocates of labor-value tell us: collective evaluation has a bearing on socially necessary labor-time. The marginalists tell us: that’s not it at all! So, we are right to follow the marginalists on…, in certain cases. I am saying: if it is true that, in certain cases, in certain social formations, you find yourself faced with an activity regime of the “continuous variation” type and not at all of the “labor” type, there is no collective evaluation of labor-time. And yet there is a collective evaluation which will govern trading. Trading is still possible, because at that point, you have a collective evaluation having a bearing on the idea of ​​the limit object or the last object. But, once again, this idea is there from the start of the series. In other words, it is an anticipatory evaluation of the time required to reach the limit object of the series. But this evaluation occurs from the first term of the series, it occurs from the beginning. In other words, it is much faster than the time required to “get to,” to the last term of the series. Moreover, it is necessarily faster than the time required to go from term 1 to term 2 of the series.

Hence a very, very curious notion that some marginalists develop in what they call their theory of evaluation or of trial and error. They developed the notion of infinite speed of adjustment, the infinite speed of adjustment. So there, they conceive it…, they fight among themselves, because there are several ways of conceiving of the infinite speed of adjustment, either in a single operation, er…, it is a really differential operation in the sense of differential calculus. But it can already be conceived in the form of a kind of integration of differentials; that is, there would be several operations which would occur in an extremely rapid time, there would be an op… a summation of these operations; or there would be only one operation, well…, This is very important…, This would be very… And they develop diagrams, then, pseudo-mathematical ones very, very, very…, very creative, very interesting, very funny …

The previous student: [Inaudible]

Deleuze: That’s it, that’s it. That’s it. And now, so, no doubt, with uh … with everything we’ve done … what’s that called? Uh… the terminal, with the distances of the… factors… of the terminal factor, there would be equivalences of infinite speeds of adjustment. The stock market currently must function… well … he … he will be saying that later. See what horizons that opens up for us. [Laughter]

So, the time of … I am saying: a second point. And, there, we have a first point which, already, opens lots of research for us on … Second point, and well, it seems obvious to me that, in all this, it seems that we are playing … — and, if you have something that you did not quite understand, it is only for this reason, it is not your fault — we are playing on the word “last” and that, in fact, there are two of them, two “lasts”. The “last” has two very, very different meanings. There we are.

So, suddenly, I’m going back to my example, because it’s obvious … let’s forget about the marginalists before going back to it. In that way, we will have three points, right? These will be… We created, little a, we start, little b, we forget everything. And I go back to my seemingly simple examples like that, but not easier than anything else. I mean…, I was saying: let’s take the word “last” literally. What is the last drink? At the cafe, it’s an assemblage. The “men at the cafe” assemblage. What time is it?

A student: Twenty-five to twelve.

Deleuze: Twenty-five to twelve already? Uh … well, men at the cafe, this is an assemblage. Or else, I said: a domestic drama, at first glance, that seems an event, in that way. No, it is an assemblage, to see the way in which, in certain households, it is repeated, and it fulfills a very specific social function. We have a fight, right; if we don’t have our fight, something is wrong. So, I am saying, at that point, it’s even more of a habit, really, it’s truly a functional assemblage. I will stick to these two examples; we can multiply them. Either … Either I am quoting for the record, because I … I would very much need it later, the last violence. Okay, I’ve got: the last drink in the cafe assemblage. See, I am looking for examples, but you should find others, but in other areas. The last drink in the café assemblage, the last word in the domestic drama…

A student: [Inaudible]

Deleuze: The last bonus trick in belote, in a game, yes, there, that’s a very complex case, yes … And then … uh … the last violence, which we will leave aside, it’s later on that we’ll need it.

Well then, well then … I’ll say right away: what’s going on? When the men are at the cafe — I’m trying to recount … to tell the uh … common story, a common story — they’re at the cafe. Each of them asks uh… he orders, uh… a drink. There are four of them, let’s say. [Pause] Okay, I am saying, don’t you get a sense that there is something strange? I assume that this very marginalist, that, in fact, the number of terms, that is, the number of drinks and the length of time they stay at the cafe, of course, is subject to all kinds of factors: the need to get home not too late, right?… uh… economic constraints, money… fine, whatever you want… But among all these factors, is there not one series of marginalist type, namely that the value of all drinks is determined by the idea of ​​the marginal drink, the last drink, so much so that it’s in order to reach — like in a finalized series – it’s in order to reach the last one that all the others are drunk?

What does it mean? [Laughter] So, why don’t they start with the last one? [Laughter] That’s absurd, because not only can they not start with the last one, but they cannot even hurry along the time which is the object of a collective evaluation. And whoever wants to hurry along the time will be laughed at. Likewise — to say how good people are and are not alcoholics no matter what some say — the one who wants to lengthen their time excessively will be despised and treated like an alcoholic. There is a kind of collective evaluation, fine; there, we have a number of drinks which is often distributed according to the number of people, that is, each one has his round. And, I am saying: the value of each drink is determined by the idea of ​​the limit object, that is, the marginal glass.

And what does “the last drink” mean? What meaning do we give to: “Come on, the last drink” … all that? Here too, these are problems of collective evaluation. The last drink is the last one before [Pause] the assemblage is forced to change. If you go past that limit, it can no longer be the same assemblage. What does that mean? If we go past that limit, then there is still another last one then, there is a last one after the last one, again, there is a last one after the last one. The last one after the last one is the one that would involve a different assemblage.

What other assemblage in this specific case? I can imagine several. Let’s start over. Let’s start again to be clear. So, the marginal last one, the one that marks the limit …, what does that mean, the limit? It means everyone has had enough. It means, we stop drinking, finished, right? This society, this social formation there, the guys in the cafe… well, that’s it for today. What does that mean? “See you tomorrow, see you guys tomorrow.” What does that mean? Literally, to go faster, I am saying: you have to recover, right, you have to stop drinking to be able to drink again. You have to stop drinking to be able to drink again, that marks what? It marks the necessary break between two sets of the same assemblage. It marks the necessary break between two exercises of assemblage. — I hope in accountancy, he has found us [inaudible words] to match. — The necessary break between two exercises of the same assemblage and marked by the marginal object.

Good. So up to there, it’s fine. I would add: this is why … this is why, in a way, you recognize alcoholism in that alcoholics are people who keep continuously stop drinking, right?[8] [Laughter] These are not people who drink all the time. Just as you only encounter drug addicts in the process of detoxification, you only meet alcoholics who are stopping their drinking. The “I’m quitting my drinking, quitting my drinking” is strictly a part of alcoholism. You will tell me: what isn’t a part of it? We’ll see. So, good. So, these are series that are like so many assemblage exercises, the assemblage remaining the same and, each time, the marginal object arrive at the end.

You will tell me: but the marginal object of each series can change. Yes, there are phenomena in the history of the speed of adjustment; there may be an intensification of the marginal object within the same assemblage, but see if, on the sly, already, the assemblage is not in the process of changing. From here, I turn to the other aspect of my problem: what would the change of assemblage be? I can say just anything. There is a threshold; and here as well, alcoholics are… they’re not idiots, eh, they’re very, very… very sensitive to thresholds. In this way, I am saying: fine, a threshold… where they only have a feeling that… they might not survive (tenir). We would have to jump into another assemblage. What then? Or else, to change the nature of the drinks, we also see that in the collective evaluation of drugs. A threshold is reached, and we say to ourselves: ah, we’re going to have to change the assemblage, the “drug” assemblage; we’re going to have to move on from grass to something else. It happens. It is indeed a problem of evaluating thresholds and limits. And then … or else, to change the nature of the drinks or change the assemblage, that is, the composition of the assemblage: not the nature of the drinks, but the people with whom you drink. They won’t be able to be the same anymore; we’ll have to change the assemblage, we’ll spend it with real alcoholics, we won’t spend it… it won’t be with friends after work, it will be something else. Fine. Here as well, there is a kind of evaluation.

Or else, we feel quite well: ah well, if… if I go too far, there, it’s… I still am risking something, it’s… this will be a particularly terrible assemblage, it will be the hospital assemblage, the “hospital” assemblage. This is an assemblage, the “hospital” assemblage, it is an assemblage. I am saying that, there, we will have reached the threshold because the limit will have been crossed. Having crossed the limit defined by the marginal object, it is necessary to change the assemblage in one way or another, by inventing another… another “alcohol” assemblage or else by entering into a “hospital” assemblage … — I don’t mean that this is necessarily a “hospital” assemblage … — or else, or else, or else… depending on the inventiveness of the assemblage, there are so many assemblages.[9]

Likewise, with my story of … I can say … so, I can go really fast here, because this is exactly the same in my story of a domestic drama. The domestic drama works, right, like an assemblage. It works through series. Each series or each exercise of assemblage, I would say, is strictly determined by the collective evaluation, that is, the oddly common, vaguely common evaluation that both partners make, if the assemblage is a couple — therefore, the collective evaluation is the assemblage of the two partners — that the two partners make concerning what? The idea of ​​the last word, the idea of ​​the last word, the marginal word, the limit word. It may not always be the same, but it would be defined by a certain weight or a certain color. Fine. There is also a collective evaluation of how long it takes to get to that word and what other words through which one has to pass. I would say: the value of the other words and the value of time … [Pause] and the amount of time it takes to reach, … etc., is determined by the collective evaluation of the marginal word. It is the limit object, in this case.

Suppose they go beyond it [the word]: here, something has gone wrong. All of a sudden, there’s one of the two saying something they shouldn’t. We can imagine it; this happens all the time in domestic dramas. Once again, when you’re the bewildered spectator, you’re like, “Well look out, they can’t go any farther than that”. [Laughter]. And then there is one of them who suddenly speaks a word that, to you, seems quite restrained, and that is the one that is not acceptable. It has gone past the limit. It has gone past the … Yeah. It went past …, precisely, it left behind the kind of agreement implied by collective evaluation. At that point, the whole assemblage changes, namely, we enter the “divorce” assemblage, the “separation” assemblage, this is another assemblage. It’s no longer the couple assemblage; it’s another assemblage, right?

Good, so what am I saying? I’m saying that “last” in the sense of marginal, does not mean “ultimate”; it actually means “second to last”, since in fact, the last in the sense of “ultimate” is the one starting from which the assemblage is forced to change. “Last” in the sense of “ultimate” or the one starting from which the assemblage is forced to change, this is what I will call “the threshold”. So why not reserve the word “last” for that one? We’re indeed forced to reserve the word “last” for that one since it is the first. It’s the first of the other assemblage. Good.

On the other hand, the last in the sense of limit and no longer threshold, no longer “threshold of new assemblage”, but “limit of the previous assemblage”, the last in the sense of limit, this is the marginal object. This is the second to last. French has a word that comes from Latin and which clearly distinguishes the second to last or the last in the sense of second to last, from the ultimate, it is the word “penultimate”. Penultimate is literally the almost last, or the last before the last. The marginal object is the penultimate or the limit. Beyond the marginal object, there is something else, the threshold once again. Beyond the limit object, there is the threshold starting from which another assemblage begins.

Hence, I can answer the question: in such a formation, what is anticipated, what is warded off? What are these anticipation-conjuration mechanisms that I attached so much importance to the last time? The answer now becomes, it seems to me, crystal clear, uh… water, uh… obvious: in the collective evaluation, what is anticipated is fundamentally the limit; what is warded off is fundamentally the threshold. [Pause] And the collective evaluation indissolubly unites anticipation and conjuration, and cannot anticipate the limit without conjuring the threshold, nor conjure the threshold without anticipating the limit.

Hence: how does it work … — I have only to apply this, and I have finished my problem anyway. Uh… — Let’s suppose, then, an abstract mode — I really insist on abstract – abstract mode of primitive exchange. This is what is happening. I take two groups, a group of pickers, who collect wild seeds. Do you remember? I am sticking to my conditions, right, I’m not giving myself a farmer’s group. For I don’t know if you sense this: farming is incapable of fitting into these series outlines. If you sense that, you’ve understood everything. Why can’t farming fit into these serial outlines, why does it necessarily refer to a State apparatus? We will see that much later, later, but you must already feel it.

So, in any case, I’m not reflecting on that; I’m telling myself: here is my little group of gatherers, they are picking wild seeds, this is not farming. The others are not metalworkers, but they make axes, they make axes. You have group A; I would say no more than A, pickers, and B, ax makers. The best axes were made from obsidian before metallurgy, for example, right? Obsidian, I think it’s a kind of volcanic rock with which you get very … very excellent edges. And then, that’s a nice word. Well, I am saying: [in] the primitive exchange, no need to compare the labor-times, no need to ask oneself what the labor-time is in order to gather, what the labor-time is in order to make the obsidian ax, and then compare both. How do you expect that to be possible?

Here, I select an example that fits perfectly. And these two groups don’t even speak the same language, well, what do you want them to compare? Moreover, gathering is a typical activity with continuous variation. For example, women go gathering, and then they sing, and then they chat, and then they gather some more. This is called an “activity with continuous variation”. There is no … there is no measurement of labor-time. For the ax maker, he’s not going to compare. It’s even an idea that would make the … the primitives laugh, comparing the time taken to gather wild seeds and the time taken to make an ax … Well, sense this then, how are they going to create the exchange?

On the other hand, they have a collective evaluation. We assume that it cannot be related to labor-time which would even presuppose a common language that they do not have. I assume that they’re even two rather distant tribes, not of the same cultural background, not sharing the same language. Fine. They exchange seeds for axes. I am saying: how can the exchange take place? Well, we know that now. We know it. We know it. Uh … [Pause] The question of whether anyone ever used this method to exchange anything is another question. But I think, yeah, I think people acted like that all the time.

Group A, you remember, eh, in your head, group A must be the one that offers seeds and receives axes. I say, in parenthesis, that it is too obvious after all I have said: marginalism is a primacy of the law of demand over the law of exchange. I’m not saying why, but you [inaudible words, Deleuze lowers his voice]. Well, group A that gets axes may develop — I say “may”, this is all abstract, eh; you cannot make the slight reflection to me about [inaudible word] — Uh… group A may develop an idea about ​​the last quantity of axes that would force them to change their assemblage. And the group may have [this idea] from the start, the collective evaluation of the last quantity of axes that would force it to modify the structure of its gathering assemblage.

What does that mean? So, I’m continuing, eh, I’m continuing in the abstract. Group B can make a collective assessment of the last quantity of seeds which would force it to modify its own assemblage. If you grant me that, you grant me everything. I would say, at that point, that the value of the traded items is strictly determined by the two limit objects for each group respectively, the marginal ax for the group that gives wild seeds, the marginal seed for the group that gives axes. And if the last ax … And then the law of exchange will be … [Pause] — Ah that’s very simple; it’s the lack of words — the law of exchange will be : [Pause] the last ax, that is, which forces the I-don’t-remember-which group to change their assemblage… which would force the group… beyond which, rather, beyond which the group would be forced to change its assemblage, equal to the quantity of seeds that would force…, beyond which the other group would be forced to change its own assemblage.

You see? This is a great market. You can’t go wrong with this system. Why? Because: what’s valuable here? There is never a direct comparison; this is an indirect exchange. Each group evaluates respectively on its own behalf the value of its marginal object, [Pause] and it is this respective evaluation by each group of the value for each of the marginal object relative to each one that will determine the exchange. In other words, the objective relation in the exchange arises from two subjective series; the direct relation arises from the indirect relation or, if you prefer, the equalization in the exchange arises from two unequal, non-symmetrical processes.

But I’m just adding … you’ll tell me: this is a little confusing, how do they evaluate? What does that mean, “they would be forced to change their assemblage”? Well, let’s reconsider the group — and I’ll finish up with this — let’s reconsider the group of people receiving axes. Beyond the marginal ax, either they have nothing to do with it, they keep the same assemblage, but they have no concerns about the last ax, they have no need to use it. So, the exchange loses all interest. This doesn’t happen anymore. This is the moment time for the break before another series begins again, the axes having worn out. This is exactly the moment for a break at the café. Or are they forced to change their assemblage, so what would that mean? Giving up their itinerance; giving up their gathering itinerance; creating with the threshold-axes – no longer the limit-ax that itself belonged to the “gathering” assemblage – but creating what with the ultimate ax, with the threshold ax? From brush clearing or worse, from stump removal, they become farmers. They could no longer follow the flow of gathering; they would switch into a completely different type of assemblage which, at least, would already involve agricultural elements. So, they ward off the agricultural threshold by anticipating the “gathering” limit. Same reasoning for axes, not difficult, namely that beyond the subsistence seeds necessary for their livelihood, the ax makers would have to change their assemblage.

All is fine; it is still the development for itself of each respective series that will fix the value of the exchange, the value of the object exchanged starting from what functions as the marginal object in a series and from which this functions as marginal object in the other series. Proof of which … Proof of which … it’s uh … it’s just the last effort that I ask of you, this gives us a confirmation, jumping from savages to things, to the most modern forms; this gives us a singular confirmation concerning the time, the margin, the limit and the threshold, the accounting exercise. [Deleuze indicates the student who he has asked to speak] You pick it up here … you pick it up here…

The previous student (who has just prepared a brief presentation): [Inaudible remarks]

Deleuze: Ah, ok, you aren’t picking it up here… And you should speak loudly… you should speak as loudly as possible.

The student: [Inaudible presentation; some students say “Louder!”]

Deleuze: Okay! I would translate that, in my own language, by saying: the assemblage exercises, the successive assemblage exercises, each, uh, primitive assemblage forms a series, and it recurs after a pause, right? Er… Each series is therefore an assemblage exercise. It is obvious that the assemblage exercises are variable. Right? No, what I want you to discuss very quickly, there, to emphasize nonetheless, is your story uh … the importance of time and of the evaluation of duration in accounting. If necessary, if you can’t hear, I’ll repeat what he says, if he doesn’t want to speak loudly, right? …

The student: [Continued inaudible presentation] [End of recording] [92:45]

Notes

[1] Besides the previous session, Deleuze discusses the relation city versus State throughout Plateau 13, “Apparatus of Capture,” in A Thousand Plateaus.

[2] On Braudel and these points, see A Thousand Plateaus, pp. 431-435.

[3] Deleuze considers the anticipation-conjuration process in plateau 13 (on the apparatus of capture), A Thousand Plateaus, pp. 435-437 (Mille plateaux, pp. 543-545).

[4] Beckett’s short story, “Premier amour” (1946) is translated “First Love”.

[5] Deleuze develops these same terms (e.g. utility, the “last”, marginalism, threshold) in section XII, “Capture”, in plateau 13, A Thousand Plateaus, pp. 437-442 (Mille plateaux, pp. 545-550).

[6] In footnote 24 of Plateau 13, A Thousand Plateaus, p. 566, Deleuze attributes this quote (and provides its text) not to an Englishman, but to Gaëton Picon, from Economie libérale et économie dirigée, vol. 1 (Paris: Sedes, 1946-47), p. 117. After the text of this citation, Deleuze adds parenthetically: “Marginalism seeks to quantify the assemblage, when in fact all kinds of qualitative factors are at work in the evaluation of the ‘last’.”

[7] Deleuze develops these two aspects of the assemblage in A Thousand Plateaus, plateau 5, “On Several Regimes of Signs,” notably, pp. 140-145.

[8] Deleuze discusses his own alcoholism in this perspective in “B as in Boire [Drinking]”, in the video interview with Claire Parnet from 1988-89, Gilles Deleuze, From A to Z [MIT/Semiotext(e), 2012).

[9] See also “D as in Desire” in Gilles Deleuze, From A to Z, for Deleuze’s moving account of being forced to deal with student at Vincennes who has experienced the extreme threshold experience and requiring hospitalization.

Notes

For archival purposes, the transcription of this seminar by Annabelle Dufourcq took place starting in 2011 with the support of a Purdue University College of Liberal Arts grant. The augmented transcription was prepared in August 2020 with reference to the revised text prepared in early 2020 by Florent Jonery at WebDeleuze. The translation was prepared in September 2020, with additional updates to the translation and transcription in June-July 2023, and a new description completed in September 2023.

Lectures in this Seminar

square right_ol
square right_ol
square right_ol
square right_ol
square right_ol
square right_ol
square right_ol
square right_ol
square right_ol
square right_ol
square right_ol
square right_ol
square right_ol