The requisite connections between my notes can be produced only gradually, over a long period of time, and only externally, by means of addenda. The atoms of this simple structure, the notes and the addenda, are in most respects identical in nature, as they are intended to perform their functions with a great degree of autonomy. Some incongruities will therefore remain in the structure of the entire text, no matter how successful my attempts, voluntary or involuntary, to unify it. It is consequently that much more important to obtain a synoptic picture of the weaving of my Residua, by concentrating upon the dynamic part of the whole, that is, the flow of addenda over time. An accounting system satisfying this objective is presented and briefly analyzed here (see Table I).

The unit of accounting is a discrete text written at a particular date, that is, a note or an addendum. Their lengths neither differ nor vary significantly over time. The accounting period is one year. This corresponds to the fact that Residua appear in the form of yearbooks. Table I below shows the inter-temporal flow of addenda between accounting periods. The flow of addenda is represented by backward linkages, as forward linkages cannot occur in this framework. For this reason the inter-temporal table is lower-triangular in structure. Of course, the number of rows and columns increases by one at the end of each accounting period. The total number of rows and columns therefore cannot be increased beyond thirty to forty. Each row of the inter-temporal table shows the flow of addenda from a particular year to all the preceding years, including that year, together with the total number of addenda and notes written that year, as well as their total (the sum of addenda and notes). Each column, on the other hand, shows the flow of addenda into a particular year from all the superceding years, including that year, together with the total number of notes written that year and addenda to these notes written subsequently, as well as their total (the sum of addenda and notes). Each column thus corresponds to the current content of a particular yearbook of the Residua. Rows and columns can be aggregated into periods covering any number of years, as long as both rows and columns are aggregated in a consistent fashion. This double-entry accounting framework also shows the total number of notes and addenda, as well as the overall total written over the entire period, that is, from February 14, 1976, to this day. It should be noted that row totals do not change over time, except for the current year in the last row, while the column totals may increase over time, except for the number of notes written in preceding years. Put differently, the share of addenda in total output does not change, whereas the share of addenda in total input either remains the same or increases over time.

Although the objective of this note is primarily to introduce the above accounting framework, as a methodological device that may enlighten and guide my work, it is useful to provide an illustration of the analytical questions that can be entertained with thus assembled and arranged information. For instance, the share of addenda in total input and output can be easily obtained from Table I in order to compare different years in terms of the backward linkages provided (see Table II). It transpires that 1980 was especially productive in securing the requisite connections between the notes (the share of addenda in total output for this year is 33 percent). The same conclusion can be drawn from the absolute number of addenda written in this year. Moreover, 46 percent of these addenda extended the notes written in 1978. On the other hand, the share of addenda in total input is obviously a decreasing function of time, but it is nevertheless interesting to notice that the notes written in 1976 have received a very large share of my attention in subsequent years (the share of addenda in total input for this year is 41 percent). However, the absolute number of addenda extending the notes written in 1978 dominates all the others, which fact is not reflected in relative terms. Also, 41 percent of all the addenda extending the notes from 1978 were written in 1980. And finally, it should be mentioned that the share of addenda in total input or output for the entire period since 1976 is 20 percent. This percentage can serve as a benchmark for comparison of all the figures presented in Table II.

A periodic review of this information, which must be updated at least once a year, will hopefully expose significant trends and patterns, which may be of value in orienting my future work. The process of weaving of my notes by means of addenda is consequently rendered sufficiently transparent to allow me, and perhaps the most diligent among my readers, a certain degree of control over the whole. Of course, all the substantive matters have been left out of this deliberation. They are, in this context, if not in general, quite irrelevant.

To François Quesnay

Table I: Intertemporal Flow of Addenda, 1976 - May 1983
Y	‘76	‘77	‘78	‘79	‘80	‘81	‘82	‘83		A	N	O
‘76	0									0	19	19
‘77	3	5								8	32	40
‘78	0	5	4							9	58	67
‘79	2	2	5	4						13	44	57
‘80	4	2	11	6	1					24	49	73
‘81	0	0	2	4	3	0				9	57	66
‘82	4	0	4	2	0	5	5			20	88	108
‘83	0	1	1	1	2	1	0	0		6	12	18
A	13	15	27	17	6	6	5	0		89
N	19	32	58	44	49	57	88	12			359
I	32	47	85	61	55	63	93	12				448
Legend: Y = Year; A = Addenda; N = Notes; O = Output; I = Input

Table II: Share of Addenda in Annual Input and Output (%)
Y	‘76	‘77	‘78	‘79	‘80	‘81	‘82	‘83		‘76-’83
OS	0	20	13	23	33	14	19	33		20
IS	41	32	32	28	11	10	5	0		20
Legend: Y = Year; OS = Output Share; IS = Input Share

Addendum I (May 11, 1983)

Science begins where madness ends. But, what is science without a proper foundation? What is there to hope for if that hope is not already in question?

Addendum II (May 20, 1983)

A careful inspection of the relationship between the notes and the addenda reveals a pattern contrary to the presupposition of continuity through accretion: even the most modest results, that is, thoughts digested only to the extent that they can be recorded with sufficient clarity for later reference, are subsequently defused and immobilized by an avalanche of equally undigested doubts that barely permit further elucidation and critical analysis. The notes and the addenda are indeed too similar to warrant any hope for a unifying structure, or at least a unifying method. The distance between all the unintegrable pieces flung outward at random is in fact increasing with time, and at an increasing rate. This explosion, or rapid disassembly—to use that jewel of euphemistic imagination concocted by nuclear engineers to denote reactor explosion, an indubitably horrible event—may involve an underlying law of motion, but it cannot but express a form of madness, a form of unreflected breakdown of all the internal connections intrinsic to a socially competent member of the species. No matter how complete the record, no matter how meticulous the notation, the trajectories of all the fragments of the dismembered body will keep diverging. The coordinates of the center, the core long vanished, might be intelligible from this record, but there is no reason to believe that the center could and should be reconstructed or reassembled. At best, the coordinates would represent a deserted place, a point without physical dimensions, and a point of no return.

Addendum III (October 12, 1984)

A point! It is surprising that a trace of optimism remains even in the bleakest premonitions of one’s future. A point: the geometric metaphor of unity!

Addendum IV (December 28, 1985)

A reader with little or no background in economics is likely to get lost among the many references buried in this note. I thus feel inclined toward providing some help on this score, although I am aware of the danger that this intervention may shed too much light on my method. This is a calculated risk, though.

First, the accounting framework proposed here in jest is associated with the work of Wassily Leontief, who was awarded a Nobel Prize in Economics in 1973 for his work on the so-called input-output analysis. Parenthetically, I started working in this field that very year, and have meanwhile become a significant contributor to one of its branches. Second, François Quesnay (1694-1774) is nowadays recognized as one of Leontief’s precursors, for he introduced the notion of the circulation of commodities based on the circulation of the blood. His Tableau Economique in some respects resembles an input-output table. And third, the title of the note paraphrases the title of Piero Sraffa’s famous book, Production of Commodities by Means of Commodities. He explicitly credits Quesnay with the fundamental insight that the process of production and consumption is circular. Quantum sufficit.

The only thing that remains to be added is that the title of this note, by implication, contains a humorous paradox. It states quite plainly for everyone to see that the ultimate product of the laborious process of writing and reading, reminiscent of production and consumption, will not be different in kind, but merely in bulk. Oh, what a horrible confession!

Addendum V (December 31, 1986)

An update of the information presented in Tables I and II, for the period from 1976 to mid-1983, is presented in Tables III and IV (see below), respectively, for the period from 1976 to the end of 1986. Several features of the new tables are worth noting. First, both total output and input declined from 1982 to 1985, and picked up slightly in 1986. My production reached the absolute peak in 1982. This was the last year when I taught exclusively at Northeastern. Second, during the period from 1984 to 1986 but a single addendum extended the notes written in the period from 1976 to 1982. The two periods are thus structurally disconnected. Nota bene, since 1984 I have taught exclusively at MIT, while in 1983 I taught both at Northeastern and MIT. Third, the share of addenda in annual input and output has increased from 20 percent in the period from 1976 to 1983 to 23 percent in the period from 1976 to 1986. This indicates that addenda are of an increasing importance in my Residua.

The structural disconnectedness of the period ending in 1983 and that starting in 1984 is most disconcerting. A conscious attempt to link the two periods should be made in the future. This link apparently hinges on the concept of planning, that is, the conscious and purposive human activity directed toward social reconstruction. Although I have abandoned many of my views on planning, I should at least address them in a critical vein. My emerging interest in the Austrian school of economics, which is primarily focused on the Austrian capital theory, in fact demands a sui generis squaring of my own accounts with respect to planning. It goes without saying that the Austrian school has been in sharp opposition to planning as such since its inception. Therefore, planning as the bête noire of Austrian economics should provide a fascinating subject on both theoretical and personal levels.

In this context it is finally worth noting that input-output analysis, which provides the foundation for the accounting framework discussed here, is one of the key building blocks of planning systems utilized today. This applies to the entire spectrum of economic phenomena extending from the firm to the national economy. The Austrians are opposed only to planning outside the firm, while they consider it to be an essential feature of human action as such in the framework of a market economy. The planning of my Residua is therefore not in contradiction with the Austrian teachings. Although the weaving of my texts falls well outside the economic domain, the above conclusion gives me special comfort. For the time being I cannot ask for more.

Table III: Intertemporal Flow of Addenda, 1976 - 1986
Y	‘76	‘77	‘78	‘79	‘80	‘81	‘82	‘83	‘84	‘85	‘86		A	N	O
‘76	0												0	19	19
‘77	3	5											8	32	40
‘78	0	5	4										9	58	67
‘79	2	2	5	4									13	44	57
‘80	4	2	11	6	1								24	49	73
‘81	0	0	2	4	3	0							9	57	66
‘82	4	0	4	2	0	5	5						20	88	108
‘83	0	3	1	2	3	1	2	11					23	42	65
‘84	0	0	0	0	0	0	0	6	4				10	37	47
‘85	0	0	0	0	0	0	0	7	4	1			12	34	46
‘86	0	0	0	0	1	0	0	6	2	1	10		20	33	53
A	13	17	27	18	8	6	7	30	10	2	10		148
N	19	32	58	44	49	57	88	42	37	34	33			493
I	32	49	85	62	57	63	95	72	47	36	43				641
Legend: Y = Year; A = Addenda; N = Notes; O = Output; I = Input

Table IV: Share of Addenda in Annual Input and Output (%)
Y	‘76	‘77	‘78	‘79	‘80	‘81	‘82	‘83	‘84	‘85	‘86		‘76-’86
OS	0	20	13	23	33	14	19	35	21	26	38		23
IS	41	35	32	29	14	10	7	42	21	6	23		23
Legend: Y = Year; OS = Output Share; IS = Input Share

Addendum VI (January 11, 1990)

Tables V and VI (see below) present yet another update, covering the period from 1976 to the end of 1989. Of particular interest is the information concerning the period beginning with 1987.

Several things are worth noting about this three-year period. First, my annual output has increased considerably. In fact, the last year is the record year in terms of total output—the number of notes and addenda. The most likely explanation for this is that my productivity had improved during the period of disengagement from MIT, which was associated with the protracted tenure process. Second, the structural disconnectedness discussed previously had been reduced, although it is still rather palpable. Third, the share of addenda in annual output had also reached a record value. The fact that addenda are of ever greater importance in my Residua is reflected by the increase of their share in total input and output from 23 to 28 percent.

It is fair to say that the accounting framework established in 1983 has played an important role in guiding my writing. Most of the problems identified in 1986 have been rectified. However, the accounting framework has thereby been rendered somewhat obsolete, in addition to having become rather unwieldy. For these two reasons, this will be the final update of the information presented in Tables I and II.

Table V: Intertemporal Flow of Addenda, 1976 - 1989
Y	‘76	‘77	‘78	‘79	‘80	‘81	‘82	‘83	‘84	‘85	‘86	‘87	‘88	‘89		A	N	O
‘76	0															0	19	19
‘77	3	5														8	32	40
‘78	0	5	4													9	58	67
‘79	2	2	5	4												13	44	57
‘80	4	2	11	6	1											24	49	73
‘81	0	0	2	4	3	0										9	57	66
‘82	4	0	4	2	0	5	5									20	88	108
‘83	0	3	1	2	3	1	2	11								23	42	65
‘84	0	0	0	0	0	0	0	6	4							10	37	47
‘85	0	0	0	0	0	0	0	7	4	1						12	34	46
‘86	0	0	0	0	1	0	0	6	2	1	10					20	33	53
‘87	1	0	0	0	0	0	0	0	0	1	1	1				4	19	23
‘88	2	2	2	1	2	2	5	4	1	0	2	2	4			29	38	67
‘89	1	1	1	1	0	0	0	1	1	0	0	2	8	38		54	64	118
A	17	20	30	20	10	8	12	35	12	3	13	5	12	38		235
N	19	32	58	44	49	57	88	42	37	34	33	19	38	64			614
I	36	52	88	64	59	65	100	77	49	37	46	24	50	102				849
Legend: Y = Year; A = Addenda; N = Notes; O = Output; I = Input

Table VI: Share of Addenda in Annual Input and Output (%)
Y	‘76	‘77	‘78	‘79	‘80	‘81	‘82	‘83	‘84	‘85	‘86	‘87	‘88	‘89		‘76-’89
OS	0	20	13	23	33	14	19	35	21	26	38	17	43	46		28
IS	47	38	34	31	17	12	12	45	24	8	28	21	24	37		28
Legend: Y = Year; OS = Output Share; IS = Input Share

Addendum VII (November 15, 1997)

I am happy to report that today’s count of my notes and addenda is 1,768 and 646, respectively, yielding a total of 2,414 pieces. In other words, the share of addenda in total input and output is 27 percent, that is, roughly the same as the last time I computed it. Regardless of the remainder of the accounting framework—which I have updated, but which is indeed too cumbersome to be reported here ever again—this key statistic is most encouraging. The weaving of my Residua continues unabated.

Addendum VIII (August 28, 1998)

Unwieldy as it may have become, I still maintain the accounting framework initiated fifteen years ago, and I still analyze the accumulated information from time to time. Today I looked into clusters of years interconnected by addenda, and I was surprised how well defined they turned out to be: 1976-82, 1983-87, 1988-92, and 1993 onward. There are relatively few addenda connecting these periods. Of course, the first period is about my life with Elise, the second about my life alone after our separation, and the third and fourth are about the two phases of my life with Lauren. The particular years that have received a great deal of my attention to date are 1976, 1977, 1983, 1986, and 1989. Again, Elise and I split up for the first time in 1977 and for the second and last time in 1983; in 1986 I met Claudia; and in 1989 I met Lauren. It goes without saying that 1976 has received many addenda simply because it is the first year of the project. I have been working very hard on connections straddling the entire book only in the last period, and especially since 1994, but its basic structure is difficult to reshape because it appears to be so dependent on my relationships with women. This is something I have long suspected to be the case, but today’s analysis definitely confirms it. Funny enough, only a few days ago I felt Lauren was misguided when she thought my book was about fucking.

Addendum IX (December 4, 2002)

The model presented in Addendum II above resembles the model of the universe offered by Robert Kirshner in The Extravagant Universe: Exploding Stars, Dark Energy, and the Accelerating Cosmos.[1] Cosmologists had long believed that the universe would either end up in a Big Crunch, or that the expansion after the Big Bang would go on forever, albeit at an ever-slower rate. Kirshner and his team have shown that the universe is, in fact, expanding at an ever-faster rate. In other words, this is not only a lonely place, but an ever-lonelier one, as well. Just as I hypothesized about my Residua. To borrow from my own account, the universe must be mad, too.

Addendum X (November 16, 2006)

Heretofore I have thought of my Residua in terms of economists such as François Quesnay and Wassily Leontief. In other words, I have thought about it in terms of a rather mechanistic application of input-output analysis, which is perhaps useful merely for notational purposes. As of today I will think of it in terms of biologists such as Charles Darwin and Richard Dawkins, instead. My book evolves by means of addenda extending previous writings. What a momentous discovery! What a leap of imagination! What a vindication of my writing method!

To Terrence Deacon

Addendum XI (November 30, 2014)

I am happy to report that today’s count of my notes and addenda is 15,809 and 2,588, respectively, yielding a total of 18,397 pieces. In other words, the share of addenda in total input and output is 14 percent, which is about a half of what it was in 1997 when I last reported it here. Regardless of the remainder of the accounting framework—which I have updated for thirty-nine years, but which is way too cumbersome to be reported here ever again—this key statistic is not very encouraging. Although the weaving of my Residua continues unabated, it does so at a declining rate. And this worries me no end.

Footnote

1. Princeton, New Jersey: Princeton University Press, 2002.