glossen 26


Who's in control? Computer Subjects and Human Objects in Friedrich Dürrenmatt's "Elektronische Hirne"
Paul A. Youngman

 

 „Elektronische Hirne"

Noch sind sie unsere Knechte
Noch führen sie aus
Was wir ihnen vorschreiben
Dumm, stur, emsig

Aber schon sind die Resultate,
Die sie liefern
Nicht mehr zu kontrollieren
Nur durch ihresgleichen

Doch bald
Werden sie weiter rechnen
Ohne uns
Formeln finden,
die nicht mehr zu interpretieren sind

Bis sie endlich Gott erkennen,
ohne ihn zu verstehen
Schuld- und erbarmungslos
Straf- und rostfrei
Gefallene Engel
      Friedrich Dürrenmatt (1958)

 

The July 2006 issue of the journal Nature contained the details of a ground breaking experiment with troubling implications for the relationship between the biological and the technological, two realms humans prefer to think of as separate and distinct.  To wit, a team of scientists and physicians led by John Donoghue of Brown University drilled holes in the head of a tetraplegic subject who had been in that state for three years.  Through the holes, they implanted sophisticated neuromotor prosthetic devices into his cerebral cortex that connected him by wires to a personal computer.  Through this brain-computer interface (BCI), the patient was then able to control a computer cursor to open and read email simply by imagining the movement of his hand (Scott 141-42).  What this experiment has in common with Friedrich Dürrenmatt's 1958 poem "Elektronische Hirne" is the focus of this paper.  The 2006 neuromotor implantation experiment is one technological realization of the unease expressed in Dürrenmatt's work.  Both events are, at bottom, about borders, as are most discussions of the disquietude humans feel toward computing technologies.  The primary borders in question were drawn over the past three hundred years with the construct of the liberal humanist subject - a creature of modern Western thought that conceives of itself as "coherent" and "rational" with a right "to autonomy and freedom" coupled with a strong "sense of agency" (Hayles 85-86).  Western philosophers, with their acknowledged penchant for duality, went a long way toward establishing the perception of a clearly divided relationship between the human and the rest of the world, a perception that carries on in modern popular consciousness.  Indeed, the liberal humanist subject has become a construct with boundaries theoretically drawn in clear, bold strokes, but, as Dürrenmatt's poem presciently implies and the BCI experiment expressly bears out, computing technologies have encroached upon these supposedly bold strokes.  Looking through the lens of Dürrenmatt's "Elektronische Hirne," this paper traces the perceived assault on the putative border dividing the liberal humanist subject from external objects like computers as the primary cause of our unease with so-called electronic brains.  

The border humans wish to see remain firmly in place when it comes to their relationship with technology is the border dividing subject and object.  Humans conceive of themselves as the subject of the metaphorical sentence of their own existences, and the computer is one of many direct objects that lie in the predicate humans command with their potent verbs.  The sentence "I control the computer," for example, is much more comfortable than "The computer controls me."  Any suggestion of the computer as subject, as Dürrenmatt's poem makes, drives a very human sense of disquietude.  As Friedrich Rapp puts it, people worry, "daß der Mensch nicht mehr Subjekt des technischen Handelns ist, sondern zu einem beliebig manipulierbaren Objekt der jeweiligen technischen Aktionsmöglichkeiten wird" (Rapp 169).  The fear of this biological / technological role reversal lies at the heart of Dürrenmatt's work.  In the computer it is as if engineers, mathematicians, and scientists have developed an electronic subject that is destined to rise to the level of the human subject and ultimately surpass it.  In this manner, according to Rapp, the computer has engendered an apocalyptic sense of biological impotence (Rapp 169).  I would not go as far as calling the sense of the subject-object reversal "apocalyptic," but I suggest that the computer has engendered and continues to engender a great deal of unease.

The power of computing technologies as represented in contemporary fiction by German-speaking authors has led literature and technology critic Georg Ruppelt to identify "zwei wesentliche Grundhaltungen in der Literatur gegenüber Automaten, Denkmaschinen und Computern."  On one side he posits there is fear and on the other "Bewunderung, Verehrung, ja sogar Anbetung" (Ruppelt 22).  I argue that in contemporary works of literature by German-speaking authors featuring computing technology there is indeed very little of Ruppelt's speculated worship or adoration.  From Heinrich Hauser's Gigant Hirn (1948) in which a giant brain-like computer uses its wireless network to cut a swath of destruction in America so wide that even the President of the United States comes under assault, to Hanns Kneifel's Der Traum der Maschine (1965) in which a computing machine creates and controls the universe conflating space and time and wreaking havoc on what the characters understand as "reality," to Dürrenmatt's posthumously published "Der Versuch" (1992) a satirical narrative set in the year 12,000 in which the very computers Dürrenmatt feared in 1958 have indeed been in control for so long that they believe they created humankind and have forgotten why, it is safe to say that the contemporary literary reception of computing  technologies, while not necessarily always fearful, is one of great unease and apprehension.  Dürrenmatt's "Elektronische Hirne" is no exception.

When he penned his work in the 1950's, these machines were not the relatively inconspicuous instruments one sees now in the form of lap- and hand-held devices.  They were monstrous new contraptions that took up entire rooms with their various apparatuses and stood at the threshold of their service as the fulcrum for contemporary technological progress.  Most importantly, according to "Elektronische Hirne," they still worked for humankind; they followed instructions as diligently as intended when humans constructed the first computers.  They appeared to be very clearly the object of the human subject.  Already in the second of four stanzas, however, Dürrenmatt's poem turns from the era of human regulation to a time in the future when computer output will only be able to be controlled by other computers, and then finally to a time when the machines will be able to compute without us.  They will outstrip their creators, and the creators ultimately will become mere objects in a world dominated by computers.  Dürrenmatt goes so far as to hypothesize that computers will discover the existence of God, but they will not be able to understand the concept due to one primary flaw that they share with humankind -- the hubris of forgetting their creator.  The machines, fallen angels that they will have become, will take to consciously exalting themselves over God.  There will, however, always be one key difference between humans and machines.  The computing machines will remain innocent and rust-free, inasmuch as their sin, committed as it is against its human creators, is only derivative of the original sin transgressed by humans against the true Creator. 

Had this poem enjoyed wide distribution in 1958, it would have been received as an absurd piece of fantasy or, at a minimum, as the work of an alarmist.  After all, despite the fact that the computers depicted in Dürrenmatt's work carry out the normal tasks of data calculation and control, they are also very much personified -- a poetic device that is itself often used to blur boundaries.  A closer analysis of "Elektronische Hirne" bears this conclusion out.  The two dominant images in the poem come from the first line and the last.  In the beginning, computers still serve as "Knechte" to humankind, and in the end they are "Gefallene Engel," albeit guilt and rust free.  Dürrenmatt's personification of computers is the key to understanding this work.  The development or, more appropriately, evolution of computers mirrors that of humankind and places humans in the exalted, albeit hubristic role of God the Creator.  Just as humans started out as God's servants, as God's objects, so too do computers vis-à-vis humans.  The tradition of original sin tells us that humans acquired more than the simple knowledge that God would have ascribed to them, and were thus ejected from Eden.  Humans began to calculate on their own, out of God's control, and, like the machines, humans were "Nicht mehr zu kontrollieren / Nur durch ihresgleichen."  Humans thus established worldly governance and laws outside of the divine realm, and, in the beginning of the seventeenth century, managed to develop intelligent creatures, just as God did.[1]  Thinking machines, as Dürrenmatt predicts, will follow the same course as human beings, and only when it is too late, will they discover that there is a God.  They will, in one sense, remain on the same level as humans because even though they recognize that there is a God, they will no more be able to understand Him than humans do.  Both humans and machines, given their mutual hubris, will remain damned with humans as fallen angels of the first order and machines as second order derivatives.

By the time Dürrenmatt is writing his poem in the late 1950's, humans had already conceded the fact that they needed help with thinking -- a concession with roots reaching back as far as the seventeenth century.  The acknowledged founder of the 'philosophy of technology' as a field of study, Ernst Kapp (1808-96), understood technology as that which results from the human desire to relieve a physiologically imposed limitation.  The role played by the artifact developed to assist in this relief is what Kapp refers to as "Organprojektion" (Kapp xvi.).  In this sense, the computer can be seen as playing the role of brain projection.  In the twentieth century, Jürgen Habermas, in describing the march of technology, takes the idea of organ projection a step further: "Zuerst sind die Funktionen des Bewegungsapparats (Hände und Beine) verstärkt und ersetzt worden, dann die Energieerzeugung (des menschlichen Körpers), dann die Funktion des Sinnesapparats (Augen, Ohren, Haut), und schließlich die Funktionen des steuernden Zentrums (des Gehirns)" (Habermas 56).  Humans arrived in this latter Habermasian stage, that of the technological strengthening of the brain in the era depicted in the first stanza of Dürrenmatt's poem.  In the early twenty-first century, we still await that which Habermas and Dürrenmatt predict -- the possible replacement of the brain, a development that may not be far off in light of Donoghue's 2006 BCI experiment.  Whether that time comes or not, humans have developed in the computer a technology that involves the projection or strengthening of our brains in an effort to assist us in thinking and, in some instances, even to think for us. 

The admission of the need for brain projection is not as simple as it is with other technologies; it is much more troubling.  For example, the invention of the spear was an acknowledgment on the part of humans that their limbs were not the most efficient tools with which to hunt.  The spear helped humans to become more effective and efficient hunters.  Although one may convincingly argue that the development of this early rudimentary weapon was a first step toward the manufacture of weapons possessing the capability to wipe human existence from the face of the earth, it still does not carry the same implications of the decision, conscious or not, that the problems humans face on this earth are so complex that in order to solve them they require the help of "elektronische Hirne."  The difference lies in the fact that there was never any fear of the spear one day surpassing humans, or, in other words, there was little concern that humans would someday become the object of the spear.

In his work, Hypertext: zur Kritik eines digitalen Mythos (2000), Stephan Porombka begins to unravel this question of technological and human boundaries specifically regarding computing technologies and in so doing supports the hypothesis that the human unease vis-à-vis computers is about those very boundaries.  He posits that from the beginnings of the Denkmaschine, the era of Dürrenmatt's first stanza, people have fantasized about what is going on in its insides, what potential slumbers within and which salvational, transformational, or destructive powers one can ascribe to it (Porombka 13).  Because the computer was quite literally a black box, "Dumm, stur" and "emsig" as it was, it was easy to wonder and worry about the mystical processes taking place within the machine without being overly concerned with its impact on the human subject/user save one matter that made it significantly different from the radio, the telegraph, or any previous technological innovation for that matter -- it could calculate, admittedly it could only calculate as Dürrenmatt describes "Was wir ihnen vorschreiben," but the average user had (and still has) no idea how it was working. 

In fact, it was this concern with what was actually going on inside the computer that forced J. Presper Eckert and John Mauchly, the developers of the first commercial UNIVAC, to add lights to the console to indicate to the user when the machine was busy calculating.  Working from the assumption that the eyes are the gateway to the soul, adding the flashing lights at least made the users feel that they had some access to the "soul" of the computing machine (Porombka 189).  The addition artificially assuaged some of the worry, but the fact remained that this machine could think, and in that respect there was an inkling of an incursion into a decidedly human realm.  One sees the anxiety over the potential power of these machines in the first stanza of Dürrenmatt's poem in which he creates a sense of foreboding with the repetitive use of the word "noch."  With this repetition, Dürrenmatt implies that machines still work for us now of course, but an ominous change is likely in the offing.  The unease associated with the opaque, black-box nature of the computer at its inception, however, was not as extensive as that which was to come.

Somewhere between Dürrenmatt's first and second stanzas, a great change in perspective takes place that is only alluded to in his poem.  With the further development of the computer into the 1960's, people began to understand it as more than a stand-alone system.  German computer engineers like Konrad Zuse and cybernetic theorists like Hermann Schmidt and Norbert Wiener (German by birth, American by citizenship) began to consider the reflexive nature of the relationship between user and computer.  For the purposes of this paper, reflexivity is defined as "the movement whereby that which has been used to generate a system is made, through a changed perspective, to become part of the system it generates" (Hayles 8).  A reflexive understanding of computing includes the system known as the human and the system known as the computer as subsystems of the larger grouping known as the human-computer system.  The idea of reflexivity muddies the water, so to speak, when it comes to borders.  If the human has indeed become part of the computer, then there is necessarily a breach in the subject-object borders.  Thus the unease is no longer solely associated with that which was taking place inside the black box, the unease also became associated with how the black box impacts its external environment to include its human user -- a subtlety directly implied in Dürrenmatt's poem. 

This emphasis on the divide between subject and object begs the question:  What if the duality upon which the liberal humanist subject is based, has been overstated?  Granted subject and object are two different things and therefore by definition dual, but what if the border is not as starkly drawn as we like to believe?  Both modern consciousness theory and contemporary cybernetic theory suggest that consciousness and physical objects are inextricably linked, that they are "fused to an irreducible duality" (Kurzweil 62).  In cybernetic terms, the relationship between human consciousness and an external, material object is really one of a reflexive informational feedback system.  For example, my consciousness as an informational entity gleans information from a tree, and from that process my consciousness projects back to the tree constructing its own understanding of it as a material object.  Thus, the boundaries heretofore considered reducible and concrete are blurred by the metaphor of the feedback loop.  To put this in terms of consciousness theory, the consciousness of the humanist subject is not a thing among other things.  It is no "mere epiphenomenon" lacking in "material consequences," but rather an "efficacious" process that can "cause things to happen" (Edelmann 3, 6).  Human consciousness is not a passive receiver of information from the external world.  It is both a receiver of information and a shaper of the world and its objects.[2] 

In this sense, there is no clear subject and object and there never was -- a conclusion reinforced by technology theorists like Norbert Bolz.  In his Theorien der neuen Medien (1990), Bolz suggests "es gibt keinen Unterschied mehr zwischen der mechanischen und der organischen Welt; die Technik hat die Selbstverständlichkeit von Gliedmaßen; der Mensch ist widerspruchslos mit seinem Werkzeug verschmolzen; Totes wird reibungslos in Lebendiges eingebaut" (Bolz 98).  The idea of an uncontradicted fusion of subject and object, the grasping of a spear by a human hand or brain-computer interface for example, goes further than a mere blurring of the boundary.  Bolz is suggesting that even if a presumed divide had existed, it has long since been eradicated by technology.  Thus the difficulties with the borders alluded to in Dürrenmatt's poem should not be that troublesome.  Things are as they always have been.  The liberal humanist subject exists in an irreducible dualistic feedback loop with the external world, a world that includes both the spear and the computer.  It is not and never has been a simple question of subject and object; the human is part of the computer and the computer is part of the human.  And yet, the disquietude with technology carries on, and does so most intensely vis-à-vis the computer. There are many reasons for this, but two stand out in the context of this paper.  First, reducible duality is something humans in the Western tradition stubbornly embrace.  Humans prefer to think of body and soul, heaven and earth, real and virtual, and physical space and cyberspace.  Many prefer not to dwell on metaphors that meld two realms that we believe to be separate and distinct.  Therefore monikers like "electronic brain," "thinking machine," or "artificial intelligence" can further human discomfort over their relationship with computers given that these terms explicitly undermine a simplistic dual understanding of the place of the human being in nature. 

The second reason for the discomfort is that despite exhortations to the contrary, there is something very different about the computer as opposed to any previous technology.  The bottom line is that there is just something unnerving about the fact that a computer thinks, and is therefore, as Martin Burckhardt points out, "weniger ein Werkzeug als eine Werkstatt (Burckhardt's emphasis)" (Burckhardt 13).  Dürrenmatt's poem, written not twenty years after the development of the first modern computer, turns on this question of thinking, both biological and artificial.  More specifically, at what point does organ projection, in this instance brain projection, become brain subsumption and finally mark the subsumption of the entire being?  Do experiments like Donoghue's BCI research, for example, finally mark a tipping point in the question of who or what is in charge -- human or computer?  Dürrenmatt sees this subsumption coming, and his hypothesis parallels the famous revelation by the mathematician Stanislaw Ulam.  In a 1958 ode to John von Neumann, a mathematician and central figure in the development of the computer in the United States, Ulam writes that von Neumann speculated on the idea that "the ever accelerating progress of technology and changes in the mode of human life . . . [give] the appearance of approaching some essential singularity (my emphasis) in the history of the race beyond which human affairs, as we know them, could not continue" (Ulam 5).  From Ulam's brief paraphrasing of von Neuman's thoughts, technology theorists embraced the term Singularity to encompass that point in time or period of time when technological progress will be driven by machines more intelligent than humans and therefore machine development will advance at an even more rapid pace making, as von Neumann suggests, "human affairs" obsolete.  The Singularity, posited in the same year as "Elektronische Hirne" was written, is the idea around which Dürrenmatt's work revolves.  The machines of the first stanza characterized as "Dumm, stur, emsig" suddenly grow beyond human control in the second stanza.  Somewhere between the first and second stanzas, humans reached the stage at which the decisions necessary to keep a computing system running became so complex that they were incapable of making them intelligently.  By the time "Nur durch ihresgleichen" is uttered, humans no longer fully control the "elektronische Hirne" of Dürrenmatt's work and are therefore no longer involved in making the most complex decisions that have been left to these electronic brains.  At this singular point, the computing machines are in effective control.   

Computer engineers have attempted to ward off this singular point.  The development of the visual display unit, which provided the "changed perspective" characteristic of reflexivity, was one such attempt.  It was an effort to gain control of this machine, but it was almost as effective as Eckert and Mauchley's flashing lights.  Porombka explains it best, "Doch wurde der Computerbildschirm nicht nur als dynamisches Medium zur Ausgabe von Daten entwickelt.  Man wollte mit ihm nicht nur den Weg von innen nach außen verkürzen, sondern zugleich den Weg von außen nach innen" (Porombka 190).  As a shortcut from the inside out or the outside in, the visual display unit provided the wary user with what she perceived as a view into the inner workings of the computer, as well as an idea of how the computer was impacting her as a user.  This interactivity provided a sense of control.  The user felt that she no longer needed to rely on the flashing lights of the early computer; she felt she now had the means necessary to quite literally monitor it.  Due in part to our inherent suspicions of thinking machines, the computer monitor or, in German, the Bildschirmgerät short for Bildschirmüberwachungsgerät, was developed, as the name implies, to regulate and control the computer.  Any sense of control, however, was chimerical at best.  What computer engineers succeeded in doing in their efforts to monitor the computer system, as Porombka implies, was to reconfigure any previous conception of borders between the user and the computer, instantiating the cybernetic understanding of the reflexive relationship between human and computer and thus engendering further unease.  The reconfiguration of the border meant the eradication of control; computers were "nicht mehr zu kontrollieren," and in some instances came to control humans.  With the reflexive relationship so drastically on display, we begin to understand that the monitor was capable of monitoring us as well -- think of computer based medical technology such as imaging devices like the Ultrasound machine.  Through this device, a pregnant woman literally invites the computer inside her womb so that fetal development may be monitored.  When the machine is off, the woman can no longer visually monitor the progress of her baby.  Far from easing our worries about this powerful new technology, the development of the visual display unit only served to obscure the presumably distinct boundary between human and machine that supposedly existed when it was only a mysterious black box.  Humans began to grasp that they are a part of the computer and the computer is a part of them, but with an added unforeseen twist.  Inasmuch as the monitor could monitor us, we began to recognize a certain degree of authority in computing technologies and, as Burckhardt hypothesizes, "von hier aus wird die Bereitwilligkeit plausibler, sich der Sache, und das heißt: der Maschine zu unterwerfen" (Burckhardt 18).  In sum, we feel as though we have allowed ourselves to be the subject of technology instead of the more comfortable other way around.

Humans still find themselves struggling to maintain some sense of sovereignty over their intelligent objects.   Despite hypotheses by technology theorists like Ray Kurzweil and Karlheinz and Angela Steinmüller suggesting that computers are metaphysical machines that will one day grasp the concept of God (Steinmüller 170, Kurzweil 152-53), early twenty-first-century humans and computers have not reached the fourth stanza of "Elektronische Hirne."  We exist in the third stanza with a great sense of unease firmly intact.  Donoghue's BCI experiment is an apt example of the source of the unease.  His subject's borders have been fully breached.  On a physical level, the computer is implanted in the man.  The biological and the technological coexist in the subject's brain -- "Totes wird reibungslos in Lebendiges eingebaut" -- an extreme example of Bolz's fusion theory.  On an informational level, the implant and the cerebral cortex partake of a feedback loop that raises legitimate concerns over who is the subject and who is the object.  The question therefore remains, if humans do exist with their environment in an irreducible informational feedback loop, that is, in a relationship requiring the understanding that the boundaries of the liberal humanist subject have always been somewhat hazy, are Dürrenmatt's concerns as expressed in "Elektronische Hirne" relevant?  The answer is "Yes, of course."  If contemporary literature is any measure of our insistence on reigning as the controlling subject, the feedback loop as described is not how humans conceive of themselves in the popular consciousness.  Even though computer engineers have provided a means of understanding how fluid those boundaries always have been, humans have not yet grasped this flexibility.  It is a matter of perspective, and our perspective has not changed enough to be comfortable with these Denkmaschinen.  We dwell on, "I control the computer."  We fear, "The computer controls me."  But perhaps we should embrace, "The computer and I control the computer and me."

 

Notes
1 The beginning of mechanized calculation and thus the computer can be traced back to Germany and the early part of the seventeenth century when the Tübingen theologian-cum-astronomer Wilhelm Schickard invented a calculating machine in 1623 for use in determining astronomical dates.  Letters and journal entries show that he shared his discovery with the likes of famed mathematician and astronomer Johannes Kepler and enjoyed the acclaim of fellow astronomers like Ismail Boulliau and Pierre Gassendi, suggesting Europe-wide acknowledge of his invention (Zuse 611).  Some twenty years later Blaise Pascal developed his own calculating machine that could do simple addition and subtraction.  Then in 1671, Gottfried Wilhelm Leibniz invented the first full-fledged, four-species calculating machine -- a machine that could add, subtract, multiply, and divide.  Despite the fact that this machine does not rise to what we would consider a computer in the modern age, Leibniz is credited with developing one of the first "thinking machines" (Ruppelt 7).

2 This is, in some ways, similar to Kant's idealism.  Indeed in his book Understanding Consciousness (London:  Routledge, 2000), Max Velmans suggests that, despite some flaws, Kant's construct comes the closest to representing how consciousness works (164-65). 

 

Works Cited
Bolz, Norbert.  Theorie der neuen Medien.  Munich:  Raben, 1990.

Burckhardt, Martin.  Vom Geist der Maschine:  Eine Geschichte kultureller Umbrüche. Frankfurt:  Campus Verlag, 1999

Dürrenmatt, Friedrich.  "Elekronische Gehirne." 1958.  Vol. 27 of  Philosophie und Naturwissenschaft.  Essays, Gedichte und Reden.  Zürich:  Diogenes, 1980.  542.  

Edelman, Gerald, M.  Wider Than the Sky.  The Phenomenal Gift of Consciousness. New Haven:  Yale UP, 2004.

Habermas, Jürgen.  Technik und Wissenschaft als 'Ideologie.'   Frankfurt:  Suhrkamp, 1968.

Hayles, N. Katherine.  How We Became Posthuman.  Virtual Bodies in Cybernetics, Literature, and Informatics.  Chicago:  University of Chicago Press, 1999.

Kapp, Ernst.  Grundlinien einer Philosophie der Technik:  Zur Entstehungsgeschichte der Cultur aus neuen Gesichtspunkten.  1877.  Braunschweig:  Westermann, 1978.

Kurzweil, Ray.  The Age of Spiritual Machines:  When Computers Exceed Human Intelligence.  New York:  Viking, 1999.

Porombka, Stephan.  Hypertext.  Zur Kritik eines digitalen Mythos.  Berlin:  Fink, 2000.

Rapp, Friedrich.  Analytische Technikphilosophie.  Munich:  Verlag Karl Alber, 1978.

Ruppelt, Georg.  "'Der große summende Gott':  Geschichten von Denkmaschinen Computern und Künstlicher Intelligenz."  Lesesaal 7 (2003):  4-22.

Scott, Stephen H.  "Converting Thoughts Into Action."  Nature 442 (13 July 2006): 141-42.

Steinmüller, Angela and Karlheinz Steinmüller.  Visionen:  1900, 2000, 2100.  Eine Chronik der Zukunft.  Hamburg:  Rogner&Bernhard, 1999.

Ulam, S. "John von Neumann (1903-1957)."  Bulletin of the American Mathematical  Society 64.3 (May 1958):  1-49.

Velmans, Max.  Understanding Consciousness.  London:  Routledge, 2000.

Wiener, Norbert.  Cybernetics.  1948.  New York:  MIT Press, 1961.

Zuse, Konrad.  "Some Remarks on the History of Computing in Germany." A History of Computing in the Twentieth Century. A Collection of Essays.  Eds. J. Howlett, et. al.  New York:  Academic Press, 1980.