SOURCE: Wessel, Karl. “Worlds of Chance and Counterfeit: Dick, Lem, and the Preestablished Cacophony.” In Philip K. Dick: Contemporary Critical Interpretations, edited by Samuel J. Umland, pp. 43-59. Westport, Conn.: Greenwood Press, 1995.

[In the following essay, Wessel explores the themes of forgery, conspiratorial “reality,” and paranoia in Dick's work and the writing of Stanislaw Lem, especially in the former's story “Shell Game” and the latter's novel Solaris.]

In his 1975 essay “Artifice as Refuge and Worldview: Philip K. Dick's ‘Foci,’” Darko Suvin lamented Dick's increasing preoccupation in his later works with private anxieties and opaque metaphysical riddles, at the expense of his earlier social and political concerns (Greenberg and Olander 73-95). It is true, for example, that many of Dick's early works dealt either explicitly or implicitly with the cold war and the self-destructive paranoia it engendered within the American body politic; one need only recall novels such as Eye in the Sky (1957) and Time Out of Joint (1959) or short stories like “The Defenders” (1953) and “Foster, You're Dead” (1955) to see Suvin's point. After his Faith of Our Fathers (1967), a story perhaps best described as Maoism turned metaphysical nightmare, it is also true that Dick's overt interest in political themes began to recede, though it can still be seen in an example such as the posthumously published Radio Free Albemuth (written 1976, published 1985).

It is curious, however, that Suvin should have imagined that sociopolitical concerns are not also ontological ones, to which they stand in the relationship of species to genus. Dick's trademark question, “What is real?”, is too easily taken for granted. “Esse est percipi,” Bishop Berkeley said: To be is to be perceived. It was one of Dick's earliest insights that politicians and their handlers understand the Bishop's meaning all too well, for the manipulation of representations of the real is the everyday bread and butter of political life. He also knew that due to the proliferation of mass media, the opportunities for dissimulation have increased exponentially in recent times. It so happened that the beginning of Dick's own writing career coincided with both the advent of television and the early days of the cold war and the red scare. A close look at this confluence of events should prove revealing.

The theme of forgery is ubiquitous in Dick's writing, taking both a concrete and a metaphysical form in a novel such as The Man in the High Castle (1962). In the first case we see the counterfeiting of familiar physical objects such as cigarette lighters and old magazines, in the second the falsification of memories and of reality itself, a possibility whose meaning occupied Dick ceaselessly. Why should this have been so? In particular, what constituted the political connection for Dick between the ontic and ontological, the being of particular things and Being itself, that such falsifications should be possible?

Unremarked by the critics, Dick's forgery theme seems to have possessed short, direct links to the American political environment of the late 1940s and 1950s, the era of the witch hunt.

The defense strategy used by Alger Hiss in his perjury trial, for example, was predicated on the thesis that certain damning documents ascribed to Hiss were in fact forged on a typewriter that had been tampered with to mimic his own¹; similar forgeries had become a cottage industry during World War II intelligence operations. In his novel Martian Time-Slip (1964), Dick erected a statue to Alger Hiss, “the first U. N. martyr,” on the Martian plain (§1.14). It seems clear that Dick was familiar with the facts of the Hiss case, which launched the career of Richard Nixon, later on Dick's bête noire.

Was the idea behind the Hiss defense farfetched? Apparently not. Senator Joseph McCarthy's staff, for example, soon began producing such forgeries as a matter of course. The initial test of McCarthy's favorite means of manipulating perceptions came in the fall of 1950 when his staff assisted in the defeat of McCarthy's political adversary, Senator Millard Tydings of Maryland. This campaign featured a technique that was to become a McCarthy staple, the doctoring of documents. McCarthy's underlings produced a fraudulent composite photograph showing the conservative Tydings engaged in pleasant conversation with the U.S. communist leader, Earl Browder. What this fake insinuated was not only false but absurd: The reactionary Tydings had himself once cited Browder for contempt of congress.² Had life imitated art? In George Orwell's 1984, published two years earlier (1948), exactly the converse of the Tydings affair occurs—a historically accurate photograph of Big Brother (Stalin) holding conversation with Goldstein (Trotsky), the only one still in existence, is taken from Winston Smith by O'Brien and destroyed. The revision of history, it seems, can be accomplished either by the creation of false memories (paramnesias) or the excision of true ones (amnesias). Inevitably the theme of the dissociation of memory, like that of the generation of paranoia, became one of the central concerns of the fiction of Philip K. Dick.

There can be no doubt that Dick was profoundly affected by the conspiratorial “reality” that McCarthy had constructed during his period of ascendancy. In 1953 Dick himself was repeatedly interviewed by FBI agents, whose object it was to convince him to inform on leftist radicals.³ When without his knowledge a story of his later appeared in the Soviet propaganda magazine Ogonek, Dick feared he might get subpoenaed to appear before the House Committee on Un-American Activities (HUAC), as had so many other writers before him. McCarthy's attack on the International Information Administration in 1953 found a reaction in a letter Dick wrote to its successor, the United States Information Agency (USIA), as late as 1974, during the most disturbed period of his life (Selected Letters: 1974 39). It is clear that Dick felt trapped within a false reality from which he couldn't seem to extricate himself, and began to explore its meaning through a series of science fiction short stories in which he took care never to name the actual object of his dread.

DICK'S PARANOIDS

One of these stories was “Shell Game” (3:189-202), a profound little exploration of the nature of collective paranoia. While “Shell Game” is neither one of Dick's best nor best known stories, like several of his efforts of the time (1954) showing signs of hasty composition and overutilization of stock SF devices, it is nonetheless a fascinating example because its central argument can be formulated almost in mathematical terms, rendering its implications unusually clear cut. Dick's gambit, “How might one prove to a paranoid that he's deluded?” is simply an operational variant of his trademark “What is real?”. One can only wonder how Dick's failure to answer his own question positively—a false failure, as it turns out—might have influenced the later course of his career.

The situation depicted in “Shell Game” is typically byzantine. A group of “colonists” has crash landed on an alien world. Five years later they have come to believe that they've been repeatedly attacked with poison gas by unknown forces, though hard evidence for this hypothesis is oddly lacking. Mutual suspicion runs rampant in the camp, driven by the belief that the camp has been infiltrated by a spy. Meanwhile, a disturbing fact has come to light that suggests to several among the group that they may all be suffering a paranoid delusion, that the attacks may exist only in their own minds. Their damaged spaceship has just been recovered from a swamp and its layout suggests that it had not carried colonists at all, but rather psychiatric patients bound for an insane asylum light-years away. None of the colonists can remember this journey any longer. Thus Dick sets up one of his most characteristic premises, that of the link between amnesia and paranoia.

Are the colonists mad or not? Can the mind ever know itself in its madness? One of the colony leaders, Fisher, anguishes, “Look, we're going around in circles. We're trying to measure ourselves. You can't take a ruler … and ask it to measure itself. No instrument can test its own accuracy” (197). Rather than resign himself to this kind of paralytic paradox, a second colonist, Portbane, suggests a surprising way out of the trap, a means to establish, once and for all, the truth or falsity of the group's consensual belief system.⁴ “I can put together a valid, objective test” he claims. “And inside of a week, I'll have it set up” (197).

At the time of the next supposed gas attack, a sample of the poison is collected from the camp's defensive perimeter. A second, control sample is then taken from the center of the camp. Both samples are exposed to litmus paper, the point being that if gas is present it will change the color of the litmus.

Crucially, the two samples are then randomized; none of the colonists, including the test's designer, know which sample was taken at the perimeter, which at the base (thus the shell game metaphor). The two samples are then presented to each of the nine colony leaders. To avoid any possibility of influence by one of them upon another, the question asked of each is answered by means of a secret ballot.

The question is, “Which of the two samples was taken at the perimeter?” It is understood that no one will answer that both samples are positive, since if the base sample were positive the test taker would be dead and thus unable to take the test. On the other hand, it is possible that both samples will be perceived to be negative, a symptom of false alarm that might be resolved later on by repeating the test. In fact, this isn't what happens: Each subject perceives that one and only one of the litmus samples is positive.

The logic of the situation, which resembles a double-blind psychology experiment, is therefore well established. The crux of the matter is that what is being tested for is not the occurrence or nonoccurrence of a particular physical event, but a set of perceptions about its occurrence or nonoccurrence. The point of contention is the nature of agreement itself: A hypothesized delusion has been made contingent upon a hypothesized hallucination.⁵ How then, Dick asks, might one sever the mechanism of unconscious collusion from the shared belief system of a group of experimental subjects, in order to determine whether their beliefs are being generated by suggestion rather than by something objectively real?

The answer is to randomize the test samples and isolate the subjects. In the event that it is actually found, a result which remains intersubjectively invariant though obtained in the absence of influence must mark the presence of something objectively real.⁶ If, following Wittgenstein's metaphor of the language game, each colonist's belief about the events transpiring at the camp's perimeter is dependent on his understanding of the application of a socially learned rule of perception, then his experimental isolation from his fellows must sever the rule from its means of implementation. What is real is whatever constant residue remains.

All of this is but the necessary prologue to what follows. Now the results of the experiment must actually be checked. If, despite the statistical barriers of randomization and isolation all experimental subjects agree on one result (e.g., “Sample A is positive”), the hypothesis that all are suffering from a paranoid delusion must stand disconfirmed. (The odds against all of them agreeing by accident is 1 in 256, the same as that of flipping nine consecutive heads or tails of a fair coin.)

If, however, a mixed result is obtained, a different picture will emerge. In particular, the more nearly it should approximate the mean result of the binomial distribution for this experiment—a 5:4 split opinion—the more likely it would appear that the colonists are delusional. In fact, this is what happens. Though each colonist would prefer to believe that his choice has been determined by a real event, the disagreement within the group is best modeled by pure chance. Each of them might as well have based his choice on the flip of a coin. Dick's colonists have proven themselves paranoid.

Or have they? Unfortunately for the peace of the colony, no such simple solution to the riddle is possible. Once this fact is understood, the violent disintegration of the group immediately commences. Why? Because a crucial feature of the situation has not been reflected in the form of the experiment; moreover, the performance of the test itself crystallizes the nature of the omission, with consequences that are dire.

Either the litmus has turned positive or it has not; a belief about that fact either is or is not delusional. What has been overlooked is that any statement a test taker makes purporting to express his belief about the litmus may or may not be dissimulated. Whether or not he lies is independent of whether or not he is deluded. The paranoid imagination is by nature prone to assume that any opinion in conflict with its own must be a lie.

There are at least two different possibilities of deception in this situation: The hypothetical deceiver may either state an opinion opposite to the one he actually believes, or he can give a prerehearsed response, one that may or may not turn out to agree with what he has observed. In this case his purpose must be to disarm his adversaries by convincing them that they are insane; the motivation for such deceit is clear from Dick's plot, since he has already set the premise that a spy (or spies) may have infiltrated the colony. It is possible that the test has been corrupted by influence after all, in the form of prerehearsal, despite the safeguards taken to avoid it. The test designer may himself be a spy; in the kingdom of the mad, the least mad mind must become suspect.

It is important to analyze this situation carefully, the better to understand it. If in such an experiment all subjects are delusional but truthful, then the result of the test should lie reasonably close to the mean of the chance distribution. (The 5:4 split will occur in 49٪ of such instances, the 6:3 split in 33٪, the 7:2 split in 14٪, the 8:1 in 3.5٪, the 9:0 split in less than 0.5٪). If half of the colonists are delusional but truthful and half delusional but lying—in the sense of reporting the opposite of what they believe—then exactly the same result can be expected. Randomized lies will not affect the distribution.

Suppose the second type of case: that a group of conspirators among the test subjects has prearranged to answer in a particular way before the question is asked (“the four were plotting in unison”). On the assumption that the nonconspirators have all answered randomly, then any strategy chosen by the conspirators that does not itself mimic randomness will likely skew the distribution of the final tally away from the chance mean. But this is not what happens in “Shell Game”; in fact, the mean result—the 5:4 split—is obtained.

At this point, any subject contemplating the test results must confront two possibilities. Either he and his comrades are deluded and the results the product of sheer chance, or his own choice was correct and determined by the data, in which case a conspiracy must exist that includes roughly half of the test takers, each of whom has chosen to pretend the opposite of what he has seen. In Dick's scenario, a genuine conspiracy involving half of the population will precisely mimic the effects of a collective paranoid delusion. The experiment has achieved nothing, except to crystallize within each paranoid's mind the possibility that he has been subject to an organized conspiracy.

Things get worse. On the face of the evidence alone, there is nothing to choose between the two possibilites. All may be deluded or half may be liars. However, the utilities (as opposed to the probabilities) implicit in Dick's story must clearly favor the paranoid's conclusion that he is not delusional but has in fact been the victim of a conspiracy. He must act as if the conspiracy exists.

If a colonist acts as though he is under attack and really is under attack (implying the reality of the conspiracy), he will try to kill his enemies, who will also try to kill him. Ceteris paribus, his chance of survival will be 50٪. If he acts as though he is under attack but is not, he will wind up killing innocent people. If he acts as though he is not under attack and is not under attack, nothing will happen. Unfortunately, if he acts as though he is not under attack but is under attack, he will surely die.

The “minimax” solution to this game—that solution which minimizes the player's maximum possible loss—is to act as if he is under attack, since the worst possible outcome for him, certain death, occurs only in the event that he does not so act.⁷

The possibility of collusional prearrangement, of an a priori conspiracy, therefore undermines Dick's example of a test for objectivity; in fact, it has the opposite effect, increasing the level of collective paranoia. Although the possibility of collusional influence being exercised during the course of the experiment is eliminatable by secret balloting, no such assumption is safe with respect to arrangements made prior to the experiment. In general, it is almost impossible to prove the negative, that something could not have happened, and this fact is eternal grist to the paranoid's mill. The most extreme variations of the disorder would rule out the possibility of isolating influences at any time, hence the paranoid schizophrenic's telepathic obsession, his dread of the possibility of thought insertion. Given the hermetic assumption that there is no end to influence, the very meaning of objectivity is unthinkable.⁸

Even when he appears to do so, in reality the paranoid is never prepared to accept any ground rules; it is not the rules of the game that must remain fixed for him, but the outcome of the game. His world cannot be framed; beyond the frame's edge always lurk invisible monsters. In order to maintain his belief system, which embodies the utility structure of a man who eternally imagines himself facing imminent extinction, he is willing to dispute that the rules are really the rules, or that anyone has in fact been following them and not other, hidden ones. This is a world without foundation, an infinite regress, an archaeological excavation into the ad hoc in which no bottom is ever reached. “We are digging the pit of Babel” Kafka once wrote, and for the paranoid mind it is so.⁹

Dick has caught the essence of this delusional pattern in “Shell Game” and yet has made an error, one that must change our evaluation of almost every aspect of his story, especially its conclusion.

Suppose that instead of a single control sample of the litmus there had been many. The nature of a shell game must change as the number of shells under which one looks for the pea increases. It is true, of course, that no matter the number of such controls the paranoid can always imagine that a conspiracy of liars has acted to imitate chance.¹⁰ Crucially, however, the proportion of perceived conspirators relative to the total number of test subjects must increase along with the number of controls to make this feasible. As the number of these controls itself increases relative to the number of test subjects, agreement between any two of them on the identity of the positive litmus must eventually become unlikely given that, despite what they may think, all of them are only guessing.¹¹

Given sufficiently many controls in Dick's kind of experiment, though it would be possible to prove to a paranoid that he has no enemies, it would be easy to prove to him that he has no friends. The collective paranoid delusion, driven as it is by malignant suggestion, is therefore reducible to a private psychosis just in case it is possible to sever the relevant lines of influence. In such instances, it is clear that the logical force of the argument must also amount to psychological force, since it would follow the paranoid's line of least resistance, his tendency to maximize mistrust. Interestingly, the sort of Hobbesian madhouse that would result is not unlike the one which confronts us on the first pages of “Shell Game,” before the experiment has succeeded in crystallizing the objects of most intense suspicion. In a war of all against all, no one trusts anyone. If Dick had been aware of the effect of increasing the number of controls in his experiment, the arc of his story would have been circular, in the end looping back to its beginning. In fact, some of Dick's uncanniest narratives, such as those of Ubik (1969) and A Maze of Death (1970), follow just such a trajectory. More importantly, far from constituting an invincible army, in this altered scenario Dick's crippled paranoids would have been easy pickings for any real adversary. Historical examples of the kind are plentiful; for example, exactly this sort of vulnerability was demonstrated by the CIA during James J. Angleton's notorious HONETOL mole hunt, leading to the paralysis for years of that agency's counter-Soviet operations.¹²

In fact, the maladaptive nature of paranoia as a survival strategy has been the subject of intense research in recent years by game theorists who have analyzed the so-called “Prisoner's Dilemma.”¹³ Greatly simplified, the point is that in a mixed population of paranoids and nonparanoids, the disadvantages that accrue to nonparanoids in repeated encounters with paranoids are usually more than compensated for by the advantages they accrue in encounters with other nonparanoids, relative to the disadvantages accrued by paranoids in their encounters with other paranoids. It seems that the evolution of human cooperation has been deeply dependent on this principle, which overturns the naive perception that Darwinian selection must inevitably lead to virulent competition.

One is still left to wonder what would happen if Dick's colonists were able to foresee all of the possible outcomes of a test taken with extra controls. Would such a test be permitted to take place at all? What would his fellow colonists conclude about the person who had suggested it? Paranoia and highly developed simulative powers from an especially combustible mixture.

LEM'S DREAMER

If Dick was SF's prophet of dissimulation, Stanislaw Lem is its prophet of chance. Since the two themes are deeply connected, like two subterranean rivers, it is no accident that we should find Lem's most typical theme inscribed within Dick's work, and Dick's within Lem's.

In his discussion of Dick's Ubik in his critical essay “Science Fiction: A Hopeless Case—with Exceptions,” Lem claims,

One possible way to build a synthetic reality is to “encapsulate” the consciousness by connecting the brain of the person in question to a computerlike apparatus. … The most interesting puzzle is whether a “phantomatically imprisoned” man can divine the real state of things, whether he can distinguish the machine-simulated environment from the real one, by means of any one experiment.

(Microworlds 77)

In fact, an extensive literature on this theme exists, based on the work of the American philosopher Hilary Putnam.¹⁴ The advent of “virtual reality” systems may one day bring us to the point where the types of experiments envisioned by Lem and Putnam become possible. The questions raised are quintessentially Dickian; they are examples of the “What is real?” constellation, which includes the key query of “Shell Game.”

Another subtly different variant of the question occurs in Chapter 4 of Lem's novel Solaris (1961), in the form of an astronaut's attempt to determine whether or not he is dreaming. In trying to solve the puzzle, Lem makes the opposite epistemological error of the one Dick makes in “Shell Game,” attributing too much power to rational methods rather than too little. Like Dick, however, Lem must offer an unusually imaginative argument in order to fail in an illuminating way.

Soon after arriving at the Solaris space station, a bizarre and frightening chain of events causes Kris Kelvin to wonder whether he may be suffering a nightmare, some creation of his own hallucinating brain. Like the colonists of “Shell Game,” he searches feverishly for a way out of his ontological labyrinth. He monologizes, “If only I could think up some experiment in logic—a key experiment—which would reveal whether I had really gone mad and was a helpless prey to the figments of my imagination, or whether, in spite of their ludicrous improbability I had been experiencing real events” (Solaris 49).

Kelvin's solution is to obtain a series of celestial position measurements from a satellite orbiting overhead, taking care not to look at them, and then to compute the same measurements himself with the help of the space station's large computer. If the figures obtained from the satellite are only figments of his imagination, he reasons, it is impossible that they should coincide with those calculated at the station, since his brain would have been unable to fabricate such a matching series ahead of time.

In fact, the two series coincide. Kelvin concludes that he is not dreaming and not mad. From this point, Lem's elegant narrative proceeds on as if nothing were amiss.

Kelvin's argument is fallacious. Perhaps this is best seen with the help of a metaphor. Suppose that the scene Kelvin has just enacted were on film. Now imagine a film editor, one able to cut frames from the film, splice them together, alter content by inserting frames from other films, and rearrange them in any way he sees fit. If one imagines the satellite readout and Kelvin's computer computation to occupy different frames of such a film, the editor, himself nowhere on the film, would be able arbitrarily to alter them and place them in any order, whether improbable or not, even, perhaps, mimicking ordinary physical causality. But all of this is simply an analogy for the activity of the unconscious; tautologies only lead in circles, and a tautology is just what Lem has produced. It is possible that Kelvin is at once Kelvin on celluloid and Kelvin the film editor; the first Kelvin is simply unaware of the activities of the second. Perhaps what has misled Lem is the assumption that all that is unconscious for the waking mind is revealed to consciousness in the dream; but it is not so. The dreamer's consciousness and the waking man's together form but a subset of the mind, a set that is complementary to a larger set of mental operations of which neither is aware.

Lem has made the category error of treating dream data as if they were public data and therefore amenable to ordinary empirical tests; in the public world it makes no sense to talk about an inaccessible common cause—in this case the film editor—because no one can observe it. However, once one has already granted the possibility of the dream, the very premise of Lem's thought experiment, then one cannot abolish by fiat that unconsciousness which is not the dream, for neither the one nor the other is accessible to public inspection. It is an empty gesture—a vacuous attempt at self-legitimation—to conduct a public style test upon the private contents of one's own mind.¹⁵

Interestingly, Lem himself discusses the effects of such a hidden common cause, in a somewhat different context, in his philosophical novel His Master's Voice (90):

If two people not communicating with each other analogously translate an unknown text, one tends to think they have truly got to its “invariants,” that what they have obtained is objectively inherent in the text and does not merely reflect their personal preconceptions. … If people's preprogramming is identical [however], the results of their investigations may coincide, even though they have not consulted each other. … Therefore we should not exaggerate the cognitive importance of such coincidences.

For “two people” read “two film frames”; for “analogous translation,” “matching number series,” and for “preprogramming,” read “invisible film editor.” The structural point, which is about the deceptive effects associated with hidden common causes, is identical in Lem's two fictional examples. The mind has its secrets of which the other part of the mind knows nothing, and in this sense a person is no different than a society.¹⁶

The “Shell Game” and Solaris scenarios are both based on experiments, the purpose in each case being to determine the existence and direction of flow of causal influence. If Kelvin's unconsciousness in Solaris consists of that which is unavailable to a part of his mind about another part of it, the corresponding reality in Dick's story is whatever information is unavailable to a part of a social group about the activities of another part of it. Each of the participants in Dick's experiment can see that all sources of influence—direct, causal links between test subjects—are absent during the course of the experiment because of the secret balloting mechanism; presumably, more sophisticated forms of signaling could be neutralized by more sophisticated mechanisms, up to the telepathic limit. Unfortunately, nothing of this applies to events that might have taken place before the experiment commenced. Since there can be no end of hypothetical common causes, logic can provide no antidote to paranoid projections.¹⁷

Paranoia, often treated as a simple symptom, is itself a syndrome composed of many symptoms. Among them are threat inflation, chronic ad hoc reasoning, and the invalid attribution of causality. In turn, paranoid causal fallacies can be analyzed into two distinct types of errors about hidden (occult) causes: in the first case attributing divergent phenomena to hidden singular causes, in the second attributing convergent phenomena to distinctly different hidden causes. The many may flow from the one, or the one from the many. One can imagine that everyone in one's town, from the high school janitor to the bank president, is part of the conspiracy from Alpha Centauri, or conversely that one's own twin brother has been replaced by an alien—or an android. Philip K. Dick was a past master of both possibilities; “Shell Game” is an example of the first, and the precession of simulacra to be found in his books, linking him to postmodernist thinkers like Baudrillard,¹⁸ provides sufficiently many examples of the second. The threat posed by the simulacrum, of course, is that it may diverge from its original in its future behavior, leading to effects that the paranoid is only too willing to imagine. It engenders suspicion because of its occult, mimetic, alien origin. It may eventually revert to type: What was once different and is now the same could one day be different again. Repeatedly, obsessively, Dick depicts this temporal dialectic of identity and alterity in his stories and novels, as he faithfully tracks the breakdown in the immune system of postmodern society.

But he even goes beyond this. To the extent that paranoia is a symptom of the inaccessibility of crucial information—that is, information about causes rather than effects, of which we always possess a choking surfeit—it can only be aggravated by the presence of amnesic gaps in the memories of persons or groups of persons. One can assess the degree of causal independence of two events only by continuously observing them as one moves through time. Here we arrive at what may be the central theme of Dick's work, which has inexplicably received little critical attention: that of the dissociation of memory, consciousness, and identity.¹⁹ Inevitably, the presence of amnesia breeds suspicion of the occult common cause because the amnesiac cannot see what is happening beyond the curtain of time.

Similarly, creatures of celluloid cannot see what the film editor is up to. Dick made explicit use of this metaphor in his paranoid-dissociative opus, A Scanner Darkly (1977). Elsewhere in his writing the metaphor seems to have evoked for him a venerable philosophical idea, that of the preestablished harmony, a notion favored by seventeenth-century thinkers such as Leibniz, Malebranche, and Geulincx. For these philosophers ordinary, linear causality was a misleading appearance: the counterfeit of an occult harmony established by God outside of time.²⁰ Leibniz's God was an invisible clock synchronizer; by the same token, Dick's invisible film editor was a Leibnizian editor. What distinguished Dick from the doyens of seventeenth-century idealism, however, was that for him the preestablished harmony looked more like a cacophony, the work of some sinister, gnostic demiurge. Dick had no use for Dr. Pangloss's deduction of theodicy, that this is the best of all possible worlds,²¹ but instead forever brooded like Ivan Karamazov over the problem of evil.²²

Empirical means can never satisfactorily address such questions, or even some of the undecidable riddles of science itself, as Stanislaw Lem's provocative explorations of the chain of chance have shown us. And yet they are sometimes more powerful than we imagine. Randomization, isolation, redundant control, and other statistical methods can help us distinguish the ideological from the ontological, the delusional from the real, the relative from the absolute. Where such means are unavailing, it is likely that there can be no theoretical knowledge of the human being at all, but only rhetorical self-deception.

Lem's translation thesis is answered by the fact that test subjects, like test items, can be randomized—if need be, over the whole human population. Theses of the indeterminacy of translation are contradicted in practice by the fact that people actually do learn and become proficient in new languages. How is this supposed to happen? Truth is social, but not in the constructive sense that the sociologists of knowledge would have it. The objectively real is whatever it is that perfect strangers already agree about before they first encounter each other. What this is cannot be established by Cartesian introspection or phenomenological navel gazing; that is sheer hubris.²³

Without such a hidden, species-universal foundation, learning, communication, and cooperative action would be impossible. The infinite predicability of all objects guarantees this conclusion, as can be rigorously demonstrated.²⁴ The real is that which remains intersubjectively invariant in the absence of influence; it is the hidden consensus gentium, the unknown koinos kosmos. We dwell in it as a fish dwells in water, and we are as little aware of our medium as the fish is of his. If things were otherwise, we would no more be able to understand each other than the scholars of Lem's His Master's Voice project are able to understand their message from the stars. This principle of consensible, reproducible results provides us our only means of testing the reality of our beliefs, of surviving the postmodern labyrinth of influence. It is the zero point of knowledge.

In the end Philip Dick himself found no solace in any statistical conception of truth, but only in his own strange and solitary gnosis. In this particular shell game, Dick made the shaman's choice.

Notes

1. See, for example, the account given in John Chabot Smith, Alger Hiss: The True Story (355-377).

2. The account is given in Reinhard H. Luthin, “The Making of McCarthy,” from Earl Latham, Ed., The Meaning of McCarthyism (12). McCarthy was up to the same tricks again in the Fort Monmouth espionage episode at the time of the televised Army hearings in the spring of 1954. This time, however, he made the mistake of forging an FBI document and had to withdraw his “evidence” (Latham xiii). Dick's “Xerox missive” crisis twenty years later possessed a number of odd parallels to the incident, arguing for a degree of unconscious influence.

3. Gregg Rickman, To the High Castle (238-240). It is possible that the FBI concluded that Dick could be manipulated after he had gotten into an argument at a Communist Party meeting he had attended with a friend (see Rickman, In His Own Words 122). Such meetings were frequently also attended by FBI agents incognito.

4. Self-reference usually takes benign forms that possess no paradoxical implications. For example, if this were not the case the DNA code would not be able to replicate itself because it would not be able to specify itself without altering itself. Unfortunately, the self-referential paradox is one in a long list of technical ideas that have become distorted in their passage through the philosophy of criticism. It is important that Dick critics, in particular, begin to understand when and why these paradoxes become vicious, because some of Dick's best works incorporate them.

5. Dick's argument might have possessed greater generality and intuitive credibility, if less dramatic effect, if he had not made it attend upon a sensory hallucination induced by suggestion. However, such effects have been known to experimental psychologists for decades. See Ernest P. Hilgard, The Experience of Hypnosis (1965).

6. See Chapter 3 of John Ziman, Reliable Knowledge, or Chapter 11 of Henri Poincare, The Value of Science. For a variety of ideological reasons, Poincare's kind of definition of objectivity is seldom credited within the literary culture. At point is the irrelevance to science of traditional “correspondence” theories of truth, favored strawmen of the French poststructuralist philosophers. The only correspondence that matters is the one that links two or more independently originating descriptions, and not a single description with some mysterious thing-in-itself. What is the likelihood that two observers, not in communication with each other, should agree about something given the infinity of possible alternatives? Much of the current debate about reference and representation thus shifts its ground to the theory of probability. In this kind of environment, poststructuralist ideas about reference tend to evaporate into thin air.

7. The “minimax solution” is an idea deriving from the theory of games, which originated in Oscar Morgenstern and John von Neumann, The Theory of Games and Economic Behavior (1944). The frontispiece to Dick's first published novel, Solar Lottery (1955), refers to von Neumann's theory “bearing out my belief that Minimax is gaining on us all the time.”

8. In contemporary literary theory the cognate of “influence” is “intertextuality.” Here as elsewhere the antithetical instincts of C. P. Snow's two cultures are evident. See C. P. Snow, The Two Cultures and a Second Look (1964).

9. Franz Kafka, “The Pit of Babel” from Parables and Paradoxes (35).

10. From the paranoid point of view, as the number of controls increases, the strategy of the “conspirators” must change in order to maintain the “illusion” of chance; this would amount to randomizing answers over the increased set of choices minus one—the paranoid's own.

11. In general, for a fixed number, R, of randomly choosing test subjects, the minimum number, N, of test samples for which the likelihood is less than 50٪ that some two of the subjects will agree is yielded by the minimum value of the ratio between the numbers of permutations of N objects taken R at a time with and without replacement, for which that ratio exceeds two to one. In particular, for R = 9, N = 55. In Dick's experiment, in order to stand a better than 50-50 chance of proving to each of nine paranoids that he has no friends, you would need 54 control samples.

12. See the account in Tom Mangold, Cold Warrior (1991). Angleton was the CIA chief of counterintelligence operations from 1954 to 1974. HONETOL was an acronym derived from the last name of J. Edgar Hoover and the first name of Angleton's favorite KGB informer, Anatoliy Golitsyn. Angleton wanted to leave the impression that the FBI director was solidly behind his investigation, but this was not the case (245). HONETOL ruined the careers of a large number of CIA officers and eventually turned against Angleton himself.

13. See Robert Axelrod's important book, The Evolution of Cooperation (1984); see also the popular account in William Poundstone, Prisoner's Dilemma (1992).

14. See “Brains in a Vat,” from Putnam, Reason, Truth and History (1-21).

15. On the hypothesis that he had not dreamed the second set of measurements, Kelvin could not have dreamed the first set unless he were clairvoyant. But how could he know he had not dreamed the second set? This is the source of the circularity in Lem's argument.

    The argument's tautologous (uninformative) nature can, however, be removed by recasting it in decision-theoretical rather than strictly inferential terms. The point is that if it is the case Kelvin did not dream the second set of calculations, then the result of his experiment would provide him a basis for deciding actions that he would not otherwise possess. In particular, should both sets match—as happens in the story—Kelvin would be advised to act as if he had not dreamed the first set. Leaving the question open simply because he might have dreamed the first set could result in a bad conclusion. The test is asymmetrical in its force, just as is Dick's in “Shell Game”: In some cases Kelvin can prove he is dreaming, but he can never prove the contrary.

    Under what circumstances should Kelvin be able to infer he's been dreaming? In the first place, he must assume that if he were awake at the times both calculations were made they would have matched, and that the satellite and space station computers are both accurate in reality. (But what is the epistemological basis of this premise? Was it determined in dream or in reality?)

    In the second place, Kelvin knows that he was either dreaming or awake at the time the two calculations were compared. If he were dreaming, then the case is already demonstrated. Suppose he was awake and that the two calculations did not match. At what point, and by what means, did he then pass from the dream to the wakeful state? If he were awake at the time of the second calculation, then how is it possible that the earlier dream calculation got placed in a real drawer so it might really be compared with the later one? It seems paradoxical to say that a fictional object and a real one can be compared in reality.

    There is no paradox here; we have simply confused the phenomenological and ontological aspects of the dream. Kelvin dreams he is receiving data from the satellite and dreams he's not looking at it, while at the same time in reality, quite unconsciously, writing down a nonsensical sequence of numbers (automatic writing). He then dreams he is transferring these figures to the drawer while simultaneously, in reality, transferring them to the drawer. In principle there is no reason a somnambulist cannot dream he is doing what in fact he is doing. If the dream surrounding the initial calculation were sufficiently vivid, it need not have been qualitatively distinguishable from the waking state. Kelvin might have then passed imperceptibly from the one state to the other, or else have simply been amnesic for the moment of transition. He could have then completed the experiment and concluded he was dreaming when he ordered the satellite computation.

    Perhaps the larger question is whether it makes sense to ask the present-tense question “Am I dreaming?” If one is dreaming, to whom is one addressing the question? One's dream self? Some personality in one's dream? But if it is not possible even in principle for the dream listener to answer the question, is it a question at all? Maybe this was the import of Wittgenstein's last words, written as he lay dying: “Someone who, dreaming, says ‘I am dreaming,’ even if he speaks audibly in doing so, is no more right than if he said ‘it is raining’ while it is in fact raining. Even if his dream were actually connected with the sound of rain” (On Certainty 89).

    It is arguable that the question only acquires meaning in retrospect. Even as the waking mind (sometimes) recollects the dream, it is possible that the dreamer recollects his waking life and thus anticipates that he will wake again. He may even aspire to an enhanced state of consciousness which recollects the dream that is ordinary wakefulness, as in wakefulness he recollects ordinary dreams: but there is no test he can use to reveal that he is dreaming. Phenomenological distinctions may not help distinguish reality from the dream. Their absence is the source of much of the uncanniness in Dick's fiction, where such boundaries are instead suggested by narrative discontinuities.

    In his best novels, such as Ubik and The Three Stigmata of Palmer Eldritch, Dick plays many perplexing variations on this theme. Several critics have remarked on the Chinese puzzle-box nature of the nested realities to be found in these books. If one grants the premise that it only makes sense to speak about the dream one has had, rather than the dream one is having—by analogy to the hierarchy of types Bertrand Russell once constructed to resolve his own famous logical paradox—then what is one to believe when the last dream in such a sequence is also the first? And yet that is the situation Joe Chip encounters at the end of Ubik: “This was just the beginning.” It is as if the smallest Chinese box were to loop back and consume the largest, reminding us that dreams have no size. Part of Dick's fascination is the way in which his books remind us of other of our period's oddest cultural artifacts, such as Gödel's theorem or Escher's drawings (e.g., “The Endlessly Rising Staircase”). Douglas Hofstadter baptized these “strange loops” in his book Gödel, Escher, Bach (1979), and Dick's are among the strangest.

16. The unitary natures of consciousness, volition, memory, and identity are among the lingering myths of Cartesianism. This psychological matter has nothing to do with “the disappearance of the subject” in postmodern criticism, which is a phenomenon of textuality rather than of lived life. Various aspects of the ways in which the mind comes apart are treated in Michael Gazzaniga, The Social Brain (1985), especially the author's treatment of the effects of cerebral commisurotomy surgery, and in Colin Ross, Multiple Personality Disorder: Diagnosis, Clinical Features and Treatment (1989), which discusses the functional dissociations of MPD (now referred to as dissociative identity disorder [DID]).

17. It might be argued, for example, that randomizing the values of R and N at the experiment's inception would make it impossible to prearrange an answering strategy. But this simply mistakes the relevant meaning of prearrangement. The best way to prearrange that the experimental outcome should look random is for the “conspirators” to in fact answer randomly over the reduced choice set, for any values of R and N. One effect of Dick's approach in “Shell Game” is to define paranoia in terms of the projection of simulations and chance. In nature the purpose of a predator's camouflage is to simulate the mean distribution of background objects so it will not be visible to its prey; one aspect of human paranoia derives from this hunter's psychology.

18. See Jean Baudrillard, “The Precession of Simulacra,” from Simulations (1983).

19. Gregg Rickman suggests, on what appears to be the basis of substantial evidence, that Dick himself suffered from dissociative identity disorder (DID). See his To the High Castle (56, 413); his reply to Robert Philmus, “Dick, Deception and Dissociation,” in Science-Fiction Studies 54, rpt. in Mullen 262-64; and also Chapter 8 of this volume.

20. Compare the following passage from Dick's Exegesis (Sutin, In Pursuit of VALIS 122-23):

    That the spatiotemporal universe of multiplicity (physical things in time and space governed by causation) is in fact subsumed by at least one higher level of volitionally-imposed organization—and that such a structure is aware of us whereas we are not only not aware of it but normally unable to be aware of it—if this can be made the subject of indubitable observations it would lie beyond any discovery in the prior history of man.

21. Voltaire satirized Leibniz as Dr. Pangloss in Candide.

22. Dick: “One can almost—almost—view Satan's activity as a high technology in which the simulation of world order is achieved” (Exegesis 111).

23. In Lem's His Master's Voice, Peter Hogarth says (90),

    Somehow it has not occurred to any of our philosophers that to deduce from the pattern of one's own thoughts, laws that hold for the full set of people, from the eolithic until the day the sun burns out, might be, to put it mildly, imprudent. … When, in formulating their theses of categorical imperatives, of the relationship of thought to perception—when did they conscientiously undertake to question, first, a large number of human beings?

24. As the philosopher of science Marx W. Wartofsky puts it, “everything is like everything else in an infinite number of respects and different from everything else in an infinite number of respects” (Conceptual Foundations of Scientific Thought 129). A circle is simultaneously a closed curve, a convex curve, a curve of constant width, a curve possessing infinitely many axes of symmetry, a curve of algebraic degree 2, etc., ad infinitum. Two circles of different radii differ from each other in infinitely many different arc lengths subtended by infinitely many distinct central angles. How, then, are we supposed to imagine that the learning of “circularity” ever takes place? How do we come to understand the meanings of rules and categories, in the sense of standard practice?

    If the a priori probabilities are equal that each of infinitely many examples belongs to a given set of finite measure, then they are all equal to zero. Only if the infinite sum of such a sequence of probabilities is finitely covergent can learning take place; but this requires that the probabilities themselves be listable in some monotonically decreasing order, like the decreasing increments by which Achilles gains on the tortoise in Zeno's paradoxical fable. It is exactly on this reef that Wittgenstein's treatment of rule learning founders in his Philosophical Investigations; he is unable to explain how it is that different language games are ever learned, given that we are not born playing them. It is through learning theory—a subject greeted with blank stares by legions of postmodern relativists, conventionalists and sociologists of knowledge—that basic realism must inevitably return to the theory of knowledge.

Works Cited

Axelrod, Robert. The Evolution of Cooperation. NY: Basic Books, 1984.

Baudrillard, Jean. Simulations. NY: Semiotext(e), 1983.

Dick, Philip K. Martian Time-Slip. NY: Ballantine, 1964.

———. The Collected Stories of Philip K. Dick. 5 vols. Los Angeles: Underwood-Miller, 1987.

———. Solar Lottery. NY: Ace, 1955.

Gazzaniga, Michael S. The Social Brain. NY: Basic Books, 1985.

Hilgard, Ernest P. The Experience of Hypnosis. NY and London: Harcourt Brace, 1965.

Hofstadter, Dougles R. Gödel, Escher, Bach: An Eternal Golden Braid. NY: Basic Books, 1979.

Kafka, Franz. Parables and Paradoxes. NY: Schocken, 1973.

Latham, Earl, ed. The Meaning of McCarthyism. NY: Heath, 1973.

Lem, Stanislav. His Master's Voice. NY: Harcourt Brace, 1983.

———. Microworlds. NY: Harcourt Brace, 1984.

———. Solaris. 1961. Trans. Joanna Kilmartin and Steve Cox. NY: Berkley, 1971.

Mangold, Tom. Cold Warrior. NY: Simon and Schuster, 1991.

Morgenstern, Oscar, and John von Neumann. The Theory of Games and Economic Behavior. Princeton: Princeton UP, 1944.

Orwell, George. 1984. 1948. NY: NAL, 1961.

Poincare, Henri. The Value of Science. NY: Dover, 1958.

Poundstone, William. Prisoner's Dilemma. NY: Doubleday, 1992.

Putnam, Hilary. Reason, Truth and History. Cambridge: Cambridge UP, 1981.

Rickman, Gregg. “Dick, Deception and Dissociation: A Comment on ‘The Two Faces of Philip K. Dick,’” in Mullen, R. D. et al., Eds., On Philip K. Dick: 40 Articles from Science-Fiction Studies. Terre Haute: SF-TH Inc., 1992.

———. Philip K. Dick: In His Own Words. Long Beach, CA: Fragments West/The Valentine Press, 1984.

———. To the High Castle: Philip K. Dick, A Life: 1928-1962. Long Beach, CA: Fragments West, 1989.

Ross, Colin A. Multiple Personality Disorder: Diagnosis, Clinical Features and Treatment. NY: Wiley, 1989.

Smith, John Chabot. Alger Hiss: The True Story. NY: Holt, Rinehart, 1976.

Snow, C. P. The Two Cultures and a Second Look. NY: Cambridge, 1964.

Sutin, Lawrence. In Pursuit of VALIS: Selections from the Exegesis. Novato, CA: Underwood-Miller, 1991.

Suvin, Darko. “Artifice as Refuge and World View: Philip K. Dick's ‘Foci’.” Greenberg, Martin H., and J. D. Olander, Eds., Philip K. Dick. NY: Taplinger, 1983.

Wartofsky, Marx W. Conceptual Foundations of Scientific Thought. NY: Macmillan, 1968.

Williams, Paul, Ed. The Selected Letters of Philip K. Dick, 1974. Novato, CA: Underwood-Miller, 1991.

Wittgenstein, Ludwig. On Certainty. NY: Harper & Row, 1972.

———. Philosophical Investigations. NY: Macmillan, 1953.

Ziman, John. Reliable Knowledge. Cambridge: Cambridge UP, 1978.