Edward H. Hagen

Telephone: 805-893-2236

Fax: 805-893-8707

e-mail: hagen@sscf.ucsb.edu

Portions of this paper were presented at the Human Behavior and Evolution Society Annual Meeting, Santa Barbara, CA, June 1995

Keywords: human, evolution, adaptation, delusions, psychosis, mimesis, parasitism, ostracism

Non-bizarre psychotic delusions are hypothesized to be psychological adaptations which evolved to mitigate the dangerous consequences of social exclusion and ostracism. When we lived in small, kin-based groups, delusions would have functioned to combat social exclusion by closely mimicking conditions, such as external threats or illness, where fellow group members were likely to cooperate and provide assistance. If delusions are adaptations to social exclusion, then they should onset when an individual faces a serious social threat, they should function (in ancestral type environments) to prevent exclusion-at least in the short term-and they should cease when the social threat ceases, an hypothesis which is examined in the context of numerous published studies of Delusional Disorder (DD).

Studies of the role of life events in DD, studies of prison psychoses, and studies of the relationship between both sensory impairment and immigrant status with DD all indicate that social deficits play a significant etiological role. Cross-cultural data collected in small, kin-based societies show that individuals with delusions often receive social benefits. Finally, studies show that positive social variables are the most important predictors of good outcome for DD. A strong case can therefore be made that delusions are an adaptation to social exclusion. Proof of the hypothesis will require large, controlled studies showing that social threats cause delusions, that individuals with delusions living in small, kin-based groups frequently receive social benefits-at least in the short term-and that the delusions subside when the social threat subsides.

Humans have evolved to rely very heavily on cooperation with conspecifics. Evolutionary theories of cooperation indicate that individuals will not cooperate indiscriminately, but rather will weigh decisions to cooperate carefully based on the potential benefits to themselves (Alexander 1979; Alexander 1987; Axelrod and Hamilton 1984; Hamilton 1964; Trivers 1971; Trivers 1985; Williams 1966). These decisions, in turn, will be based in part on information communicated by the prospective cooperative partner in a variety of ways including not only speech, but indicators of affect, actions, and a host of other obvious and subtle signals.

In ancestral human environments, individuals who were unable to find cooperative partners, those who faced social exclusion and ostracism, also faced very serious negative fitness consequences. In such dangerous circumstances, I argue that individuals experienced delusional psychoses, somatoform disorders, and other psychiatric symptoms and syndromes which mimic the honest signals upon which cooperative decisions are based. By mimicking honest signals of fear or distress for example, individuals became eligible for cooperative benefits which they had previously been denied. Delusional and Somatoform disorders may be evolved deceptive strategies designed to extract social benefits from others.

My hypothesis requires two assumptions: first, human cooperation is based on discrete psychological mechanisms, and second, such mechanisms rely on signals communicated by other individuals, especially thosse that appear to be largely beyond conscious control, e.g. obvious signs of fear and distress. It is this second assumption which makes individuals vulnerable to exploitation, and which is the key to this paper. If individuals receive false information that they have good reason to believe to be true, they may cooperate when it is not in their interest to do so.

I examine this proposal for the case of non-bizarre delusions, although it may apply to somatoform and other disorders as well. Because delusions occur in a large number of syndromes, usually in concert with other cognitive and affective symptoms that would make it difficult to determine the etiology of the delusions themselves, I will focus primarily on Delusional Disorder (DD), often referred to as Paranoid Disorder, as a preliminary test of the hypothesis. Individuals with DD are cognitively, emotionally, and physically unimpaired, and their only symptom is a delusional framework. Although DD is rare, delusional symptomology is not: the prevalence of persecutory delusions with auditory hallucinations alone is approximately 0.7% (Robins and Regier 1991). Delusional systems appear to take a very similar form in many different societies (Westermeyer 1988) consistent with a universal adaptation, and common delusional themes like persecution and grandiosity have a striking social significance.

Just as nausea and fever are physiological adaptations to toxins and infections, delusions may be an evolved psychological adaptation designed to mitigate the deadly consequences of social isolation and exclusion, experiences that appear to play a significant etiological role in this disorder. In the human Environment of Evolutionary Adaptedness (EEA)-the environment in which we evolved and to which we are adapted (Bowlby 1969; Tooby and Cosmides 1990)-our ancestors lived in small, somewhat isolated kin-based groups with occasionally hostile neighbors (e.g. Lee and DeVore 1968). In this (as opposed to a modern) social environment, delusional individuals may have been able to unconsciously deceive others in order to receive badly needed social benefits like food and protection.

Individuals who experienced intense delusions of an external threat, for example, would have acted consistently with this belief (Kennedy et al. 1992; Wessely et al. 1993). By doing so, they may have been able to convince others to cooperate with them in defense against this external threat, mitigating the consequences of their previous exclusion from the group. Delusions therefore represent an example of exploitative mimicry. If this hypothesis is correct, delusions are not a disorder at all, but rather an innate and functional aspect of adult psychology.

The ability of individuals employing an exploitative mimetic adaptation (EMA) to successfully mimic aspects of cooperative communication in order to acquire benefits may be the result of an asymmetrical intraspecific arms race (Dawkins and Krebs 1979), since the benefits to a socially excluded individual of successfully exploiting the cooperative mechanisms of others are much higher than the costs to the exploited individual(s): they are potentially saving their own lives while imposing a less than deadly cost on others.

Looking at the EMA hypothesis another way, when the costs of deception are small and the potential benefits are large, deceptive strategies will be strongly selected for. The costs of deception include exclusion from future cooperation and physical injury. Because ostracized individuals are already excluded from future cooperation, and because their reproductive opportunities are close to nil anyway, the costs of deception will be very low. Any cooperative opportunities made possible by deception will, of course, be extremely beneficial. This argument suggests that socially excluded individuals should be expected to use deceptive strategies. As this paper will show, DD has many of the features one would expect of a deceptive strategy, and it has a close association with social deficits.

The EMA hypothesis is also consistent with the hypothesis that self-deception functions to facilitate the receipt of cooperative benefits (Alexander 1979; Nesse 1990; Slavin 1985; Trivers 1985). In modern contexts, individuals still face isolation and exclusion, but the delusions that may have helped them during our evolutionary history now often have the maladaptive effect of increasing rather than decreasing social isolation.

This last point bears brief elaboration. While paranoid and grandiose individuals seem to repel rather than attract cooperation, fear and exaggeration are potent strategies for manipulating small groups of vulnerable individuals (witness Waco, Heaven's Gate, and the occasional shaman). The small hunter-gather band with real enemies in a unpredictable environment were, at times, profoundly vulnerable. EMA's should flourish under these conditions.

This manuscript is organized in five sections, beginning with this introduction. Section two very briefly outlines the theoretical underpinnings of human sociality and the rich array of social resources available in ancestral environments. Section three presents arguments that social resource deficits (e.g. exclusion and ostracism) are to be expected when the receipt of social benefits relies on exchange. Section four provides the theoretical foundation for the evolution of strategies to exploit social resources when exchange is not possible. The theoretical tools developed in sections two through four are put to the test in section five, where the case is made that DD is an exploitative mimetic adaptation that functions to extract social resources when other means have failed.

Humans expend considerable effort extracting benefits from conspecifics through both cooperation and exploitation, and many authors have commented that such activities appear to constitute a significant fraction of both conscious and unconscious mental processes (e.g. Alexander 1987, Slavin and Kriegman 1992) as well as perhaps having provided a selection pressure for the increase in cranial capacity during hominid evolution (Trivers 1971). It is to be expected that humans have numerous psychological adaptations for exploring the possibilities and avoiding the pitfalls of cooperation (Trivers 1971). A precise specification of these adaptations requires a comprehensive model of human sociality however, which is currently lacking. Nevertheless, the vast bulk of human social behavior is probably based on inclusive fitness (Hamilton 1964) and some form of reciprocal altruism (Trivers 1971). The process of social exclusion and the evolution of strategies to mitigate or avoid its consequences must be understood in the context of these two theories.

According to inclusive fitness theory, individuals will increase their reproductive fitness when they provide benefits to genetic relatives. As long as the cost of providing the benefits is less than rB, where B is the fitness benefit realized by the relative and r is the degree of relatedness, the individuals will "profit" reproductively even if the relative does not reciprocate. While a child's status as a group member was based on inclusive fitness, kinship dilution and sexual competition between even closely related adults implies that adult social relations must have relied heavily on reciprocity in the social environments of evolutionary adaptedness, hereafter referred to as the SEEA. Parents and children are only related by r=1/2 (as are siblings), and this ratio decreases rapidly with each genealogical remove. In order for the parent to increase their fitness by providing benefits to their offspring, the fitness benefits realized by the offspring must be more than twice the fitness costs of providing them. This differential is easy to achieve when children are young, but is more difficult when offspring are adults and able to provide for themselves. Furthermore, adult siblings are competitors for mates and it is unlikely that a brother or sister will continue to provide benefits to a sexual competitor if those benefits are unreciprocated.

These arguments suggest that inclusive fitness alone cannot explain human sociality and that reciprocal altruism is fundamental to understanding adult social relations. While it is beyond the scope of this paper to summarize the considerable literature on reciprocal altruism and cooperation (see, e.g., Axelrod and Dion 1988), all models share the following features:

1. Individuals attempt to receive important benefits from other individuals by providing benefits.

2. The net value of the benefits an individual hopes to receive over a finite number of transactions is greater then the cost of the benefits provided.

The exchange of benefits need not be simultaneous however; in fact, a considerable period of time may pass before a reciprocal "debt" is paid. The transfer of a benefit may in this case be viewed as an investment in a future return. These arguments suggest that most beneficial social relations among adult humans are based, at least in part, on psychological mechanisms which evolved to extract benefits from conspecifics by providing benefits "at a profit." These benefits will be increased by choosing to reciprocate with kin over non-kin. Adults who failed to provide valuable benefits would have been at increased risk for social exclusion in the SEEA.

After approximately 100,000 generations, the sole surviving member of the genus Homo made a rapid transition from foraging to domesticating food, a mode of subsistence it has practiced for about 500 generations. Roughly coincident with this subsistence shift were equally dramatic demographic transitions (Cohen 1977), including a three orders of magnitude increase in overall population, huge increases in population densities, and a shift to a relatively non-kin based social organization in several subpopulations. Because few adaptive genetic traits would both arise and reach fixation in 500 generations, most researchers assume that complex human psychological adaptations to the social environment must be adaptations to social environments associated with foraging rather than domestication. The SEEA was the social environment of gatherers and hunters.

Less is known about SEEA than is commonly presumed, as virtually all models of the SEEA are based on ethnographic analogy with groups like the Dobe Ju/'hoansi, a practice which has recently faced blistering criticism (Schrire 1980; Wilmsen and Denbow 1990; cf. Lee 1993; Solway and Lee 1990). Nevertheless, it is safe to assume a simple model that has the following features: humans lived in small, kin based groups that cooperated in the acquisition of food and mates, in defense against occasionally hostile neighbors, that exchanged important information, and that provided care for offspring and injured or ill members (Brown 1991; Cohen 1977; Keeley 1996; Lee and DeVore 1968; Sahlins 1972; van den Berghe 1990). These are crucial benefits, and the fitness potential of individuals without access to each of them was very likely zero. Humans in the SEEA did not survive, let alone reproduce, without the care and assistance of others.

For organisms which rely heavily on cooperation, cooperative deficits represent a significant risk. There are many ways that cooperation can fall short or even fail, a few of which I list:

1. Failure to find a cooperative partner.

2. Terms of cooperation are poor.

3. Cooperative partners can provide few benefits.

4. Reciprocal investments yield few or no returns.

5. Renegotiation of a cooperative relationship on less favorable terms.

6. Termination of a relationship.

7. Simultaneous termination of several relationships.

8. A hostile individual seeks to impede one's attempt to cooperate with others.

9. Several individuals cooperate to impede or sever one's relationships, or drive one from the group (ostracism).

There are theoretical reasons for suspecting that cooperative deficits are an intrinsic aspect of cooperation. It has recently been recognized that reciprocal partner choice is potentially important for organisms which rely on cooperation (Bull and Rice 1991; Nöe 1992; Nöe and Hammerstein 1994; Nöe et al. 1991; Tooby and Cosmides 1996). Clearly, we should expect organisms to attempt to cooperate with individuals who can provide them with the greatest benefits at the lowest cost. Since cooperation with one individual may necessarily preclude cooperation with another, individuals may have to compete for cooperative partners, with the result that there will be a market for cooperators. Competition for cooperative partners has many implications, the most important for this paper being the constant risk faced by each individual that important relations will be "renegotiated" at a lower profit margin, or even terminated should a better partner be found.

Anthropology has tended to reify the group; groups and group membership are taken for granted. If cooperative partner competition is important in human sociality, as it appears to be, then it is unlikely that individual human psychology takes group membership for granted. Cooperative relations must be constantly maintained by providing benefits.

During the SEEA, the consequences for an individual of cooperative deficits or failures would have been severe. As noted above, in all known traditional societies individuals receive very important benefits from their reciprocal relations with kin and allies. At a minimum, these include food, protection, and spouses. Reduction of cooperative relationships would have put access to each of these benefits at risk; loss of such relationships likely meant death. An individual who has failed to form, or lost, all cooperative relations has very few options. From an evolutionary standpoint, consequences must be judged by their impact on inclusive fitness, and in this regard, social failures would have greatly increased the difficulties in finding or keeping a mate, immature offspring would have received less care and investment, and close kin may have suffered as well. In short, cooperative deficits must have represented a significant selection pressure on the evolution of human psychology. For a serious threat of ostracism, an extraordinary response is called for.

Both shunning and social exclusion appear to be cultural universals (Brown 1991), and many cases are documented ethnographically in small, kin based societies (Hoebel 1954):

A Labrador girl was banished in the dead of winter because she persisted in eating caribou meat and seal together. Because of the imputed separatistic sentiments of the denizens of land and sea, this is a basic Eskimo tabu. Her act endangered the whole community, since the angered animals would shun the local territory: frustration of the efforts of hunters and starvation for the community threatened.

Among the several circumstances that affected the well-being of the [sacred] Arrows, and thereby of the tribe, was the killing of a Cheyenne by a Cheyenne....A curse fell on the whole nation. While the blood yet remained on the feathers "bad luck dogged the tribe." No war party could dare to hope for success; hunting parties would return empty-handed. All the people suffered for the act of one man....The solution was exile....

Occasionally a member of a [Carib] group acquires a reputation among his fellows as an undesirable character. He may repeatedly pilfer from others' fields; he may trouble the women, be lazy, show himself ungenerous, constantly pick quarrels, or make himself obnoxious in other ways. The men of the settlement will talk to him, but if he does nothing to improve his position in their eyes, he will be advised to leave on pain of having life made very unpleasant for him. If he persists in remaining, he will find that he and his family are social outcasts: they are not invited to drinking parties; he will be unable to borrow anything; he will get no help in hunting, fishing, field cutting, canoe building or other activities in which the men assist on another, nor will his wife receive aid in her occupations; his household will be excluded from the water-hole and bathing place. In short, he will lose all advantages of group life. In aggravated cases, the other men may beat him or even kill him....(Gillin 1934; cited in Hoebel 1954)

There are also examples of ostracism among non-human primates. Chimpanzees live in cooperative social groups, and Goodall (1986) has identified examples of social exclusion among free-ranging chimpanzees at the Gombe National Park, Tanzania, usually involving the rejection of an individual due to the aggression of a few males.

Cooperative deficits in general, and ostracism in particular, were very likely recurrent and dangerous problems faced by hominids during their evolution, and it is equally likely that humans have psychological adaptations designed to mitigate their effects. In the sections that follow, I argue that certain psychiatric "disorders" are just such adaptations.

From an evolutionary perspective, the intentional communication of information or sending of a signal by an organism should be viewed as a strategy intended to manipulate the receiver of the signal to the benefit of the sender (Dawkins and Krebs 1978). The information may or may not benefit the receiver. Organisms may communicate true information when it is in their interests to do so, and they may communicate false information when it is in their interest to do so. Correspondingly, a receiver organism may acquire information about a sender organism that benefits the receiver, whether or not it benefits the sender.

Markl (1985) argues that outright receiver or sender exploitation should be rare in cooperative dyadic relationships (as opposed to competitive interactions), since both are attempting to extract the maximum benefit from the cooperation. If communication enhances the effectiveness of cooperation, then bluff and deceit should be the exception rather than the rule among animals living in stable social groups where future interactions are likely. He notes, however, that the receiver in the dyadic relationship is susceptible to third-party parasites that mimic the sender's signal, since discrimination against the parasite's signals may jeopardize the benefits obtained by communicating and cooperating with the sender. "[P]ersistently effective third party exploitation can take advantage of a dyadic communication system which is highly important for the dyad's fitness. Since this is also the condition most favoring selection against intradyadic exploitation and for intradyadic cooperation, we should expect to find such exploitative third party intrusion most spectacularly combined with very cooperative and mutually beneficial social relationships. Ant mimics and social parasites yield particularly impressive examples (Hölldobler 1977)." Social systems that rely on "cheap" signals for the exchange of substantial benefits are very susceptible to exploitation by parasites that can mimic these signals. For humans, these "parasites" are most likely to be individuals excluded from social exchange since they have nothing to lose by exploiting others-excluded individuals faced imminent death in the SEEA.

Markl lists characteristics of cooperative communication systems

which enhance the efficiency of communication and make it as resistant as possible to uncooperative parasitic cheaters from within the group and against exploitation by intruders from outside....Mutualistic communication should be deictic, i.e. directed at specific addresses (Green and Marler 1979) in the most economical way (optimizing the benefit/cost ratio), informing partners as accurately as necessary to achieve the communicative purpose (e.g. by indexing external referents), controlling (e.g. by frequent role change) for fair reciprocation and against in-group exploitation, keeping the whole exchange as private as possible so that unwelcome eavesdroppers and freeriders stay excluded. We therefore would expect such communication to be narrow-cast rather than broadcast, unobtrusive, difficult to discover and recognize by outsiders and unaddressed receivers (including observers). This means, the very reasons that cause us to understand communication as manipulation force us to conclude that the most developed forms of cooperative communication should not be characterized by spectacular showiness but rather by finely adjusted subtlety (see Markl 1983).

Humans clearly rely on a large set of intentional and unintentional signals and cues in their relationships in order to maximize cooperative benefits and minimize exploitation.

I hypothesize that Delusional Disorder is an example of exploitative mimicry of the type predicted by Markl, employed by individuals unable to participate in dyadic cooperation. Individuals with this disorder are completely convinced that they face a life threatening external threat or illness, for example. I argue that the powerful and unshakable belief systems are a form of mimesis, since these fears can be communicated to others in a very convincing fashion. Socially threatened individuals employ this mimesis in an attempt to shift the cost/benefit calculations of others in their favor.

Mimicry and crypsis may constitute the largest single class of adaptations among vertebrates, arthropods and opisthobranch gastropods (Starrett 1993), and we should not be surprised if humans have also evolved highly elaborate mimetic adaptations. Adaptations are identified based on an argument from design: if a biological structure has a set of characteristics, each of which corresponds to a unique aspect of a particular problem, then there is a high probability that the structure represents an adaptation to this problem. It is important to note that the behaviors of complex organisms are not in themselves adaptations; rather it is the underlying neurophysiological structures that produce this behavior which constitute the adaptations (Symons 1979). Identical behaviors can be generated by different adaptations, and a single adaptation can generate many distinct behaviors, complicating the identification of a neurophysiological adaptation.

Mimetic adaptations are often easy to identify. If a non-toxic butterfly has the same very distinctive appearance as an unrelated species of toxic butterfly, it is straightforward to conclude that the resemblance constitutes a mimetic adaptation to avoid predation. Demonstrating that DD is an example of intraspecific exploitative mimicry faces several obstacles, however. All humans have false beliefs, have emotional reactions to these beliefs, and communicate these to others. Furthermore, all communication can be interpreted as an attempt to manipulate others for one's own benefit. Without a better understanding of how humans obtain benefits through communication and cooperation, it will be difficult to specify the expected "design features" of exploitative mimetic adaptations (EMA's). Nevertheless, I offer the following hypotheses for an EMA:

1. The EMA is present in all individuals, but only activates in those who are facing the loss of important social relationships since extensive exploitation, if discovered, will almost assuredly lead to the loss of important cooperative benefits. This consequence of exploitation, or "cheating", is predicted by virtually all models of the evolution of cooperation (Axelrod and Dion 1988; Axelrod and Hamilton 1984). Only individuals who have little left to lose should attempt (even unconsciously) to deceive others by employing an EMA.

2. Individuals employing an EMA behave in ways that are difficult to consciously imitate, like displaying intense fear or excitement, because such displays are more likely to convince others.

3. The displays are designed to facilitate the forming of relationships and the receipt of social benefits. Convincing other group members of the existence of an external threat would be very effective in this regard. The social benefits should relate to defense, reciprocal investment, and reproduction.

4. Individuals can give reasons for their behavior that are difficult to independently verify, at least immediately. Examples include the claim that one possesses important information or has an intimate relationship with a high status individual.

5. With the exception of the condition being mimicked (fear, distress, etc.), an individual appears to be "normal" so that they are viewed as a potential reciprocal partner.

6. Mimicked honest signals are contextualized, that is, an individual can provide explanations or additional information which provide a plausible basis for the signals.

7. Individuals will find it very difficult NOT to provide contextualizing information; they will communicate with others.

8. Individuals mimicking honest signals will NOT be a priori ineligible for the receipt of the intended cooperative benefits.

9. The EMA de-activates when the social threat subsides.

There is evidence that the test case of DD, presented below, satisfies every hypothesis, and conversely, that these hypotheses account for most of the significant clinical, etiological, and demographic aspects of DD.

As noted above, during the SEEA social exclusion was a deadly threat. While we might intuitively expect individuals facing such a threat to respond by attempting to please members of the group and to conform to group values, there are at least three important situations where this strategy is unlikely to succeed. The first is if an individual has already substantially modified their behavior in an attempt to solidify their social position, yet still faces exclusion for whatever reason. Continuing to "suck up" will not benefit this individual. The second is if an individual violates a serious taboo, such as incest. Such a violation might be unforgivable. Third, if an individual angers a powerful person, perhaps by having an affair with their spouse or child, it is possible that no amount of submissive behavior will prevent their exclusion. As a last resort, individuals may (unconsciously) employ exploitative mimesis. In fact, because an EMA is a "cheating" strategy that might jeopardize future cooperative relationships, it should only be employed as a last resort.

There are three domains where humans receive substantial social benefits: defense, reciprocity (social exchange), and mating. Each of these should consequently be the target of individuals wishing to extract social benefits via mimesis.

Defense: Belief in an external threat provides a very strong impetus for cooperation among humans, and it has been argued that external threats by conspecifics were a significant selection pressure for the initial evolution of cooperation among hominids (Alexander 1987). Levine and Campbell , in their discussion of realistic group conflict theory, state that "Real threat causes ingroup solidarity" is "the most recurrent explicit proposition of the theory (1972, p. 31)." Although they note that they "can make no pretense to comprehensive citation of this most ubiquitous principle," they cite in support the following authors: White 1949, p. 133; Sumner 1906, p. 12; Sherif and Sherif 1953, p. 196; Sherif, et al. 1961, p. 21; Coser 1957, pp. 87-95; Simmel 1955; Lewis 1961; Murphy and Kasdan 1959; Boulding 1962, pp. 162-163; Mack and Synder 1957; Dahrendorf 1964, p. 58. The internal political divisions that cause certain individuals to be excluded vanish in the face of external dangers. Group members should readily cooperate against any possible threat because the costs of defending against a false threat are low compared to the costs of not defending against a real threat.

I make the obvious assumption that individuals will only cooperate against an external threat once they have received sufficient information for them to conclude that such a threat might exist, and that humans have evolved honest signals of fear and distress which will be used by others in evaluating the possible existence of a threat. In other words, I assume that honest signals of fear and distress have evolved to elicit some type of benefit, presumably cooperative and presumably for defense.

Reciprocal investment: I propose that humans will take opportunities to make reciprocal investments when they can provide large benefits to others at low cost to themselves (throwing a rope to a drowning man, for example). They are then eligible for a return on this investment when the benefited individuals reciprocate (Gouldner 1960; Trivers 1971). Humans have evolved honest signals to indicate when they are in distress and are eligible for receiving these kinds of cooperative investments. These include signs of physiological disorder (pain, discomfort) and fear of illness. In addition to purposeful signals of distress, humans give off unavoidable cues of distress as well; for example, humans don't flee from a predator to signal distress, but it is a very reliable cue of distress nonetheless.

Mating: Males provide resources and protection to females in exchange for mating opportunities. Conversely, females offer mating opportunities in exchange for resources and protection (Symons 1979). Given the huge reproductive advantage males receive by mating with multiple females, if they receive signals that a nubile female is willing to mate with them in exchange for social or material benefits, they will likely cooperate.

Several authors have suggested that certain psychiatric symptoms and syndromes may be adaptations (Badcock 1990; Chance 1988; Gardner 1982; Gilbert 1989; Glantz and Pearce 1989; Huxley et al. 1964; Nesse 1991; Price 1972; Thornhill and Thornhill 1990; Wilson 1993). Unpleasant experiences like nausea, vomiting, and fever are healthy, physiological adaptations in response to toxins and infections. Analogously, intense, negative psychological experiences like delusional psychosis and hypochondriasis may be normal, "healthy" adaptations to some of the considerable risks and dangers encountered by organisms that lived in small, cooperative social groups-shunning and social exclusion in particular.

Although delusions can accompany many psychiatric and medical disorders (Manschreck 1989), I restrict my argument to the distinct nosological entity Delusional Disorder, characterized by the following clinical, etiological, and demographic features (see Jorgensen and Jensen 1988; Kendler 1980; Kendler 1982; Kendler 1984; Kendler 1987; Kendler and Tsuang 1981; Koehler and Hornstein 1986; Opjordsmoen 1987; Winokur 1977, for recent work on the nosological validity of DD and related delusional psychoses):

Clinical (APA 1994):

presence of:

non-bizarre delusions of at least one month's duration

absence of:

hallucinations

disorganized speech

disorganized or catatonic behavior

flattening of affect

markedly impaired functioning

odd or bizarre behavior

underlying medical condition

physiological effects of a substance

Etiological: significance of life events in the genesis of the disorder (Kraepelin 1971 p. 300, cited in Kendler 1982; Kaffman 1981), particularly those appearing to involve social isolation or exclusion (Cameron 1943; Lemert 1962; Retterstöl 1966).

Demographic: association with immigrant status (Carpenter and Brockington 1980; Chiu and Rimon 1987; Ettinger 1959; Ettinger 1960; Kendler 1982; Ødegaard 1932; Westermeyer 1989); association with low socio-economic status (Kendler 1982); late age-of-onset and the extreme rarity of DD among pre-adolescents (Kendler 1982).

The presence of non-bizarre delusions is the essential clinical feature of DD. Meissner (1987) defines delusional conditions as

a relatively permanent and unshakable delusional system accompanied by the preservation of clear and orderly thinking. The delusions are typically persecutory in nature or take the form of delusions of jealousy. The delusional system may form the basis for anger and resentment. Ideas of reference are common, and are often associated with a life pattern which involves considerable social isolation, seclusiveness, or eccentricity (emphasis added).

Although easy to recognize, non-bizarre delusions have been difficult to define, for they are not simply false beliefs. The following are considered important aspects of delusional systems:

Non-bizarre. The delusions could, in principle, be true. They do not violate fundamental cultural tenets.

Systematized. The delusional system forms a logical and coherent whole. Recent events may be incorporated into the system, or used as supporting evidence.

Oltmanns (1988) lists the following defining characteristics, with none considered to be either necessary or sufficient:

The belief is held with firm conviction. The person's statements or behaviors are unresponsive to the presentation of evidence contrary to the belief.

The individual is preoccupied with (emotionally committed to) the belief, and finds it difficult to avoid thinking or talking about it.

The belief involves personal reference, rather than unconventional religious, scientific, or political conviction.

The belief is a source of subjective distress or interferes with the person's occupational or social functioning.

The individual does not report subjective efforts to resist the belief (in contrast to patients with obsessional ideas).

Delusional Disorder was previously called Paranoid Disorder (PD, DSM III). PD is identical to DD except that delusions other than persecutory or jealous are excluded (see below). The term "paranoia" usually refers to PD. Paranoid Schizophrenia (DSM IV) is similar to DD, except that prominent auditory or visual hallucinations are present in addition to delusions. A diagnosis of Schizophrenia requires significant social/occupational dysfunction and continuous signs of the disturbance must persist for at least 6 months. It is not the purpose of this paper to propose an adaptive function for Paranoid, Catatonic, or any other type of Schizophrenia. Unfortunately, studies of delusions often include individuals who might be diagnosed as Schizophrenic or for whom a diagnosis of DD is excluded due the presence of prominent hallucinations or other psychotic symptoms. Besides delusions and hallucinations, psychotic symptoms include disorganized speech, and grossly disorganized or catatonic behavior. The use of data including any such individuals will be noted.

Where possible, findings for DD will be contrasted with those for Schizophrenia. Schizophrenia provides an excellent control case for DD since it also a psychotic disorder whose symptoms include both bizarre and non-bizarre delusions, as well as the more disabling psychotic symptoms.

As will be seen below, DD has a social and demographic "fingerprint" quite distinct from Schizophrenia. When delusions are separated from other psychotic symptoms, an etiology of social exclusion and isolation begins to emerge.

Paranoid Personality Disorder (PPD, DSM IV) is not considered to be a Psychotic disorder; individuals are not delusional-they do not cling tenaciously to an elaborated false belief-nor have they experienced other psychotic symptoms. They are, however, very distrustful and suspicious of others, whose motives are interpreted as malevolent. I suspect that PPD represents an adaptation to cooperative deficits based on vigilance rather than mimesis. Socially threatened individuals must be on the constant lookout for attempts to deprive them of material, social, or reproductive resources. They must also be more vigilant in defending against injury and disease. PPD, anxiety, obsessive-compulsive, and certain somatoform "disorders" may therefore be vigilance-type adaptations to social threats. Vigilance appears to be the more commonly employed set of strategies, and is probably used for less severe social threats than mimesis. Socially threatened individuals who fear members of their in-group may be increasing their vigilance towards likely internal adversaries rather than attempting to exploit them. Although PPD matches the folk conception of paranoia, the term paranoia or paranoid will NOT be used for individuals with PPD.

It is far beyond the scope of this paper to critique or even review the extensive theoretical literature on delusions. Briefly, following the review of Winters and Neale (1983), delusions have been attributed to disturbances in affect and thinking, deficits in perception, defects in the psyche, projections or externalizations of personal wishes, conflicts, or fears, altered views of the self, susceptible personality types, existential conflicts, avoidance responses, unsuccessful social interactions, and cybernetic regulation of the self and others (please see Winters and Neale 1983 for references and critique). Winters and Neale observe that most theories can be characterized by two major "themes": delusions are either motivational (individuals are motivated to explain unusual perceptions, or they are motivated to reduce or ameliorate uncomfortable emotional or psychic states), or delusions are a sign of an underlying cognitive defect. They conclude: "In sum, despite large numbers of explanation and theories on delusional thinking, there is no agreed upon conceptualization or general model concerning their nature and very few theories enjoy empirical support."

Despite the apparent pathological nature of DD, several lines of evidence support the hypothesis that it is in fact a suite of adaptations, a theoretical approach that has never, to my knowledge, been applied to delusions. Because, as I will argue, delusions are deceptive strategies, they are therefore high risk strategies; high risk, however, must not be confused with pathology. Every common category of delusion has a natural interpretation as a functional mechanism to elicit cooperation or mitigate the consequences of cooperative deficits. Cross culturally, the vast majority of delusions can be characterized by a relatively few themes-persecutory, grandiose, jealous, somatic, and erotic (Ndetei and Vadher 1984; Westermeyer 1988), an interesting feature of DD that is downplayed by the DSM IV emphasis on the presence of any delusion.

Delusional themes are not random or arbitrary, but rather reflect their close association with important aspects of cooperation. I argue that each delusional theme, or subtype, is a distinct mimetic adaptation:

Persecutory: As noted above, socially threatened individuals need to greatly increase their vigilance towards the social environment. Thus, paranoid fears of persecution by powerful individuals may have been adaptive during our evolutionary history. Fears that are highly elaborated and directed towards external threats may be better explained by the mimetic hypothesis, however. Individuals with these delusions can give very convincing accounts of the reputed threat, behave consistently with the delusion (Wessely et al. 1993), and give cues of genuine fear and distress (Kennedy et al. 1992). Belief in an external threat provides a very strong impetus for cooperation among humans, especially those living in small, autonomous bands with real enemies. This type of delusion exploited the defensive cooperative mechanisms of others.

Grandiose: Individuals are convinced they have some great (but unrecognized) talent or insight, or have a special relationship with a very important person. Individuals are essentially presenting themselves as highly valuable cooperative partners. Because group members should prefer to socialize with those of high status, individuals who convincingly present themselves as having high status could have a considerable impact on others, at least in the short term. By doing so, they are attempting to exploit partner-choice and reciprocal investment mechanisms.

Erotomanic: Individuals believe that another person, usually of high status, is in love with them. Males with erotomanic delusions often attempt to rescue females from some imagined danger (APA 1994). They are claiming to offer defensive benefits at no real cost to themselves, apparently in an attempt to obtain both cooperative and reproductive benefits. Females, on the other hand, often believe that high status males are in love with them. Individuals who can claim an intimate relationship with a high status individual, a claim that could be difficult to verify, increase their own status. This will exploit partner-choice and reciprocal investment mechanisms. Females may also be attempting to sexually manipulate high status males, although erotomanic delusions would seem superfluous in this case.

Somatic: Individuals with somatic delusions, which are often difficult to distinguish from Hypochondriasis (APA 1994), are preoccupied with the fear or idea that they have a serious disease based on a misinterpretation of one or more bodily signs or symptoms. The fear persists despite medical reassurance. Again, socially threatened individuals need to be particularly concerned about falling ill because of the uncertainty that others will care for them. In addition, individuals displaying convincing fears or signs of physical distress or illness may encourage others to provide care as a social investment. Group members may be tricked into providing care under the mistaken assumption that they are going to receive a substantial payback for helping a seriously ill person. Delusions that an individual emits a foul odor from the skin, mouth, rectum, or vagina, or that certain parts of the body are ugly or misshapen may also be indicative of the vigilance of socially threatened individuals, although an adaptive function for these delusions is not obvious.

Jealous: Individuals believe their mate to be unfaithful. I hypothesize that jealous delusions represent a form of hyper-vigilance. In the SEEA, socially threatened individuals were likely at great risk for losing their mates. In contrast to other types of delusions, jealous delusions are not exploitative.

Delusions of different types often appear together in a complimentary fashion. An individual who is attempting to gain cooperative benefits based on defense profits by presenting themselves as a valuable and important person as well.

It is important to note that individuals act consistently with the delusional framework. Wessely et al (1993) report that among their sample of 83 subjects with delusions, 60% reported at least one action based on delusion. Of the 59 subjects for which information was available from informants, 52% were reported as having either probably or definitely acted on a delusion. Persecutory delusions were significantly more likely to be acted upon than other beliefs (p = 0.002). Kennedy, et al (1992) report on a series of 15 patients with DD who were being supervised by a forensic psychiatric service after violent or threatening acts. In all cases, the violent acts were congruent with moods of fear and anger, but in three cases the violent act was not related to the delusion. Other actions, such as fleeing or barricading to avoid delusional persecutors were also consistent with the congruence of mood and delusion.

If DD is a universal psychological adaptation, then it should be found in all cultures. Westermeyer (1988), relying on a review of the literature, four years of field work in Asia, 15 years at an International clinic at the University of Minnesota Hospitals and Clinics, and several studies of culture and psychopathology conducted in the United States, makes the following cross-cultural generalizations about DD:

1. The structure of delusions varies little, if any, across cultures, whereas the content may be influenced by culture.

2. Culture-bound (i.e. emic) and secular (i.e. etic) delusional content are not mutually exclusive, but may coexist in the same individual.

3. Delusional content can be quite etic, or secular, and yet still give rise to behaviors that are highly culture bound or emic (such as building a religious shrine or undertaking amok-type violence).

Delusional themes are remarkably constant across cultures. Westermeyer presents several studies (Carpenter et al. 1973; Leff 1974; Westermeyer and Sines 1979) on the universal structure of delusions, supporting the necessary requirement that adaptations are universal.

I argue that during the SEEA individuals facing severe social threats and isolation developed powerful delusional systems. These caused them to behave in such a way that their probability of receiving cooperative benefits increased significantly, mitigating the social threat, at least temporarily. For example, an individual experiencing a persecutory delusion-the Bongo-Bongo are trying to kill me-would display very convincing signs of fear and distress and be able to cite evidence of the truth of their claims. In a small, somewhat isolated band with occasionally hostile Bongo-Bongo neighbors, such a display could be convincing enough that others would cooperate with this individual against the Bongo-Bongo, a common enemy.

Proof of the hypothesis requires two basic demonstrations: first, that one or more types of social deficits causes DD, and second, that individuals experiencing DD in traditional contexts were likely to receive cooperative benefits which they otherwise would not have received. Unfortunately, in modern environments, this adaptation is no longer adaptive. Only a tiny fraction of the world's population currently lives in small, isolated communities with hostile neighbors. Citizens of industrial societies live in large communities with extensive police and military forces, and have access to many sources of information. Since external attacks are unlikely and exaggerated fears are easy to disprove in these contexts, delusional displays of persecution have very little chance of success, and in fact are almost always maladaptive-they tragically tend to intensify social isolation rather than mitigate it. Therefore, although social deficits should cause DD in all societies, DD will provide social benefits only in the now rare small, kin-based societies.

If DD is an adaptation to social deficits then social deficits should cause DD. Social deficits are strongly associated with DD in a number of ways. Whether these associations are best explained by social selection theory (individuals with DD tend to experience social deficits), social causation theory (social factors cause DD), confounding variables, or methodological problems, is currently an open question. Nevertheless, many theorists have argued for a causal role for social deficits in DD. Kraepelin noted that while in dementia precox (Schizophrenia) it was possible to observe an internal "morbid process" at work, paranoia could usually be understood as the response of certain persons to "the influence of the stimulus of life." (Kraepelin 1971 p. 300, cited in Kendler 1982). Kendler (1982) argues that the low rates of psychiatric illness among family members of patients with DD, and the fact that social variables like immigration and low socioeconomic and educational status appear to play a significant etiological role in DD, distinguish DD from Schizophrenia. Environmental factors look more important than genetic-constitutional ones. The hypothesis that these "environmental factors" are social deficts is supported by several lines of evidence. Principle among these are case control studies of DD vs. Schizophrenia. Because the symptoms of DD are less disabling than those of Schizophrenia, social selection theory would predict that DD will be associated with fewer social deficits than will Schizophrenia. Several studies described below show just the opposite: DD, the less severe syndrome, is associated with more social deficts than Schizophrenia, exactly what would be predicted by social causation theory (assuming a less social etiology for Schizophrenia-a plausible assumption given that Schizophrenia is known to have a significant genetic component).

Cameron (1943) was among the first to explicitly locate the genesis of delusional systems in the social arena. Cameron identifies the importance of social isolation and lack of social communication in the development of a delusional framework, and notes specifically that "[p]aranoic attitudes and actions...grow out of a breakdown in the machinery of social cooperation." Cameron, however, felt that isolation from the community was only the "final outcome" of a process that led the delusional individual to act detrimentally on his environment. Interestingly, he recognizes that delusional behavior may occasionally make an individual a "distinguished person and, though rarely, a leader of men," an important phenomenon that may have occurred more frequently in traditional social contexts.

Lemert (1962), contra Cameron, presents a strong case for the causal role of social exclusion in paranoia. He retrospectively studied eight cases of persons with "prominent paranoid characteristics." Four cases involved persons admitted to the state hospital at Napa, California, with diagnoses of Paranoid Schizophrenia. The lack of any history or evidence of hallucinations or intellectual impairment excludes Schizophrenia as a likely diagnosis for these cases, however. The others involved persons admitted to hospitals, involved with the law, or having chronic job difficulties. One case resembled Paranoid Personality Disorder.

Lemert spent as much as 200 hours per case collecting data from anyone who played a significant role in the life of the person involved, attempting to establish the order in which delusions and social exclusion occurred. He found that

[t]he paranoid process begins with persistent interpersonal difficulties between the individual and his family, or his work associates and superiors, or neighbors, or other persons in the community. These frequently or even typically arise out of bona fide or recognizable issues centering upon some actual or threatened loss of status for the individual. This is related to such things as the death of relatives, loss of a position, loss of professional certification, failure to be promoted, age and physiological life cycle changes, mutilations, and changes in family and marital relationships. The status changes are distinguished by the fact that they leave no alternative acceptable to the individual, from whence comes their "intolerable" or "unendurable" quality. For example: the man trained to be a teacher who loses his certificate, which means he can never teach; or the man of 50 years of age who is faced with loss of a promotion which is a regular order of upward mobility in an organization, who knows that he can't "start over"; or the wife undergoing hysterectomy, which mutilates her image as a woman (emphasis added).

Difficulties with the community often result in the formation of a coalition which attempts to exclude the individual, and Lemert concludes that it is this process of exclusion and isolation which leads to the development of the delusional framework and not the converse. He notes that paranoia emerges in situations where "the goals of the individual can be reached only through cooperation from particular others, and in which the ends held by others are realizable if cooperation is forthcoming from ego," precisely the conditions that pertained during the SEEA.

Retterstöl's retrospective/case-control study (1966) of 301 first admission patients to the Psychiatric Clinic at the University of Oslo who presented with paranoid and paranoic symptoms, also supports a social causation theory. Based on an analysis of case notes and follow-up interviews, he found that 100% of paranoid psychosis are caused by an event that "provokes the insecurity of the individual," i.e. those that tend to isolate the individual and make him feel an outsider, either by making him unpopular within his own group, or by transplanting him to new and strange surroundings. This was true of only 54% of cases diagnosed with Schizophrenia.

Kay, et al (1976) conducted a case-control study of 132 consecutive first admission or re-admission patients to a psychiatric hospital with diagnoses of either paranoid (n=54) or affective (n=57) psychosis. The paranoid patients were specifically diagnosed as belonging to the following ICD 8 categories: Schizophrenia, 15 cases; Paranoid States, including Paraphrenia, 38 cases; and Alcoholic Hallucinosis, 1 case. The affective diagnoses were: Hypomania, Mania, or Manic-depressive Psychosis, 11 cases; Endogenous Depression, 32 cases; Psychotic or Agitated or Involutional Depression, 10 cases; and Depression Unspecified, 4 cases. At the onset of illness, the following features were significantly associated with the paranoid group: low social class, having few or no surviving children, living alone, and social deafness-all social resource variables. By contrast, affective illness was significantly associated with the presence of (presumptive) precipitating factors of the illness and with a family history of affective disorder in first degree relatives-presumably a genetic variable. Five variables were significant in discriminating between the two diagnoses: low social class, few surviving children, and deafness predicted paranoid illness; precipitating factors and family history of affective disorder predicted affective illness. Paranoid patients were rated more highly by both themselves, relatives, and friends on items suggesting difficulty in forming and maintaining satisfactory interpersonal relationships before the onset of the illness, and had been more solitary, shy, reserved, and suspicious, and less able to display sympathy or emotion. Kay, et al conclude that their data support "a multifactorial hypothesis, according to which various adverse circumstances, especially when in combination, such as being unmarried, having few close relatives, belonging to lower social class groups, or becoming deaf, are expected to increase the chances of hardship, insecurity and loneliness in later life. The accumulated sense of deprivation and injustice might be supposed to conduce to paranoid illness." Because personality score did not correlate with these social variables-socially impaired personalities were not associated with low social position-they disfavor downward social drift as an explanation for the correlation of the social variables with paranoid illness.

Kaffman (1981), in a retrospective study group of 34 individuals with DD (DSM III Paranoid Disorder), found that in every case there was a clear and realistic connection between paranoid premises and facts and events in the patients' life. He also found that "authentic past and current interpersonal transactions play a dominant role in generating and activating the paranoid beliefs (emphasis added)." From the limited number of case studies presented, these transactions appear to have involved isolation and rejection.

In a study by Jørgensen (1995), 88 patients with delusional psychosis (first-admitted patients to a psychiatric hospital during the calendar year of 1984) were asked "If you got (sic) personal or emotional problems, do you think you could talk with any of your family, friends, fellow workers or neighbors?" Fifty percent answered no, compared with 10% of the general Danish population between 18-66 years of age. Only 35% reported that they met friends on an average of at least once a week (no comparable data is presented for the general population). Only one fourth of the patients seemed socially well adjusted and economically self-supporting.

If social deficits cause DD, then isolated individuals should be particularly at risk, especially those who are purposefully removed from society. The association of delusional systems with solitary confinement-an extreme form of social isolation-has long been recognized (Nitsche and Wilmanns 1912). These data also support a causal role for social isolation in the onset of DD.

Grassian (cited in Henderson 1995) evaluated inmates of California's maximum security prison, Pelican Bay, in particular those in the Security Housing Unit (SHU) who are isolated 22 1/2 hours a day, and are excluded from work and group exercise programs. Of 50 inmates interviewed, 58% suffered from paranoia largely or completely attributable to SHU conditions, and 34% were actively psychotic. Whether these individuals would be diagnosed with DD is unknown.

Grassian (1983) also reports on a study of 14 inmates subject to solitary confinement at Massachusetts Correctional Institution at Walpole. The population was not preselected by overt psychiatric status. All were male, with a mean age of 28 (range: 22-38). Median duration of confinement in isolation was 2 months (range: 11-300 days). In this study 43% of the inmates experienced persecutory fears attributable to isolation, although these paranoid ideas did not reach delusional intensity. Many of the other symptoms also reported by these inmates, hyperresponsitivity, perceptual distortions and hallucinations, anxiety, and cognitive difficulties, are documented effects of sensory deprivation studies (Heron et al. 1953; Lilly 1956).

Auditory or visual impairment is often cited as a causal factor in psychoses, indirectly supporting a causal role for social deficits in the formation of delusional systems. Since human sociality relies heavily on verbal and visual communication, individuals suffering sensory impairment are arguably at higher risk for experiencing social deficits.

The association of hearing loss with persecutory delusions has been known since Kraepelin (1915). Cooper and Curry (1976), and Cooper et al. (1976) have found significant correlations between certain subcategories of deafness that they term social deafness, and paranoid illness, for the same patients studied by Kay, et al. (above). Rates of deafness among paranoid patients were over three times higher than expected, as compared to only 1.4 times higher for the affective patients. They conclude that prolonged social deafness is probably a factor in the etiology of paranoid psychoses occurring late in life. Other studies (Moore 1981; Watt 1985) have found no correlation between delusions and either deafness or blindness. Cooper et al. stress, however, that only in contexts where deafness leads to social isolation or exclusion does one find delusional symptoms.

In another study of these patients, Cooper and Porter (Cooper and Porter 1976) found that impaired visual acuity and ocular pathology are more common among the paranoid group, although the etiological significance is less clear than for social deafness. Ziskind (1958) argues for an etiological role for sensory deprivation in cataract "post-operative psychoses" (implied to be primarily of the paranoid type). There is a greater incidence of this complication than after other types of surgery where deprivation does not occur. The psychosis disappears in most cases when the unoperated eye is uncovered, and there was a progressive reduction in the incidence of psychosis with advances in operative technique which reduced the period when both eyes were covered.

A number of sensory deprivation experiments were conducted in the 50's and early 60's. Although these researchers mention that some subjects developed delusions (Lilly 1956), I have been unable to document this. Lilly also mentions that there are several cases of acute paranoia among pairs of ocean crossing sailors, although he provides no documentation.

Immigrants and refugees are quite likely to suffer social deficits since they have often left family, friends, and other important social ties behind. The successful formation of new social ties in the adopted country is far from assured. Tellingly, numerous studies have found extremely high rates of delusional and paranoid symptoms among immigrant and refugee populations (Carpenter and Brockington 1980; Chiu and Rimon 1987; Ettinger 1959; Ettinger 1960; Kendler 1982; Ødegaard 1932; Westermeyer 1989). Two studies show rates of DD among immigrants be 40-50 times that of the indigenous population (Ettinger 1960; Westermeyer 1989), compared to only a 3 1/2 fold increase for Schizophrenia (Ettinger 1960). Kendler (1982) found rates of DD among the foreign born to exceed rates of either Schizophrenia, or affective illness (p << 0.0001). Clearly, DD has a particular association with immigrant/refugee status.

Unresolved is whether these results are best explained by social selection theory, social causation theory, or other factors. In a careful study of paranoid symptoms and disorders among 100 Hmong refugees living in the United States, Westermeyer (1989) notes that in 3 cases (33%) pre-emigration personality or genetic characteristics may have been significant in subsequent paranoid disorders, but in the 6 other cases, no such pre-emigration factors could be found. His study indicates that successful acculturation, assessed in several ways, is associated with low paranoid symptoms. Chiu and Rimón (Chiu and Rimon 1987) report that 56% of the paranoid immigrants in their study had no history of psychiatric treatment prior to immigration (only 22% of these patients had a DSM III Paranoid Disorder while 61% were classified as Paranoid Schizophrenic). These studies indicate that social causation appears to contribute to the high prevalence of delusional symptoms among immigrants, although social selection is probably an important factor as well.

DD is associated with poor social and economic standing, as is mental illness in general (Robins et al. 1991). This association seems particularly strong in the case of DD, however. In Jørgensen's study (1987), over 50% of delusional patients were not self supporting. Kendler, in his review of the demographics of DD as compared to Schizophrenia and affective illness (1982), notes that patients with DD were more likely to come from poor economic backgrounds and to be more poorly educated than either patients with affective illness or (in most cases) Schizophrenia. He argues that since Schizophrenia should produce greater psychosocial disability than DD due to more disabling symptoms, downward "social drift" is an unlikely explanation. Kay, et al (Kay et al. 1976) in their comparative study of paranoid vs. affective illness also found the paranoid patients to be significantly associated with low social class as compared to the affective patients. They, too, disfavor the social selection hypothesis. Whether this kind of data supports the social deficit hypothesis depends on whether economically or educationally disadvantaged individuals in Western societies have increased odds of suffering social deficits of the types outlined above. This cannot be assumed.

Mirowsky and Ross (1983), however, present a simple model which links low socioeconomic status with mistrust and thus with vigilant paranoid tendencies. Using data on 463 individuals collected during a community mental health survey in El Paso, Texas, and Juarez, Mexico, they demonstrate that "paranoia" ("paranoia" being determined by responses to four questions similar to diagnostic criteria for DSM IV Paranoid Personality Disorder, but without a determination of whether a reasonable basis for these beliefs exists) is most strongly associated with "mistrust"-the feeling that it is safer to trust no one. Mistrust, in turn, is associated with belief in external control together with low socioeconomic status. Belief in an external locus of control-the expectation that "outcomes of situations are determined by forces external to one's self, such as powerful others, luck, fate, or chance"-is associated with low socioeconomic status, Mexican heritage, and being female.

Although the study did not examine whether delusions (as opposed to paranoid tendencies) were associated with low socioeconomic status, the association of mistrust with low socioeconomic status under certain circumstances suggests that economically disadvantaged individuals are more likely to suffer social deficits. Those who mistrust feel that others wish to cheat or exploit them rather than cooperate. Mirowsky and Ross similarly argue that powerlessness, victimization and exploitation are the causative factors of mistrust and "paranoia."

Because one major risk of deception is exclusion from future cooperation, young individuals should be less likely to employ a deceptive strategy since they are putting at risk more future opportunities to cooperate-young individuals should try harder to "suck-up." An EMA should also be less likely to activate in older individuals because they have fewer potential cooperative benefits to offer. Older individuals are riskier reciprocal investments since they are less likely to survive to provide a return on the investment. They may also be able to provide fewer defensive benefits due to physical frailty. Consistent with these predictions, DD is largely an affliction of middle age (figure 1).

Additionally, an EMA should not be activated in individuals who are a priori ineligible for, or do not require, the receipt of the intended benefits, that is, it should not be activated in individuals for whom the deceptive guise would be implausible. Although children faced serious social threats, in particular the loss of one or both parents, it would have been unlikely that they could mitigate these threats by experiencing persecutory or grandiose delusions. Why would enemies be targeting a particular child? Could a child reasonably pass him or herself off as being high status or having valuable social connections? Adults were unlikely to view children as important social partners. Although all humans require food and protection, during the SEEA children would have been unable to acquire these resources through reciprocal relations with others since it is unlikely that adults viewed children as plausible partners for cooperative defense or food sharing arrangements. Children acquired these resources through investments by parents and kin. Conversely, children did not require many of the benefits that go with defensive cooperation such as protection of mates and offspring. DD should therefore be extremely rare among children, a surprising conclusion given that children often fear imaginary dangers.

The demographic data on DD supports this hypothesis. Very few individuals younger than 20 are diagnosed with DD (See fig. 1). Cases of DD in individuals younger than 15 are almost unheard of, though mental illness in children is itself not uncommon.

A diagnosis of DD requires the presence of a delusional system, and such systems appear very rarely in pre-adolescents. Of the very few scattered case reports on delusional systems in pre-adolescents, almost all are examples of shared psychosis, with a parent as the primary partner. I have found one small study, however, that presents convincing evidence of delusional systems with both persecutory and grandiose delusions in 13 pre-adolescents (Arthur and Schumann 1970). The subjects ranged in age from 8 yrs, 11 months to 13 years, 9 months (mean = 12 yrs, s = 18 months). IQs ranged from 80 to 121 (mean = 101, s = 10.5). These children may be the exceptions that prove the rule. Subjects shared two characteristics of relevance to this hypothesis. First, "[a]ll of the boys had severe difficulties in getting along with other children and were avoided, scapegoated, ridiculed, or bullied by peers because of their 'odd' and unpredictable moods and behaviors. By the time we saw these children most of them were 'loners.'" The children are experiencing social isolation and rejection. Second:

most had an astonishing facility for engaging in rather appropriate though pseudo-adult and affectless conversations and interactions with adults. This capacity to simulate normality with adults was so well developed in the three boys who became inpatients that our highly trained and experienced psychiatric nursing staff occasionally seriously doubted the diagnoses of these youngsters. To some extent this same capacity was also noted during psychiatric interviews. It broke down completely in most peer relationships and in their highly disturbed response to the pressures of psychological testing.

It appears that the very rare child who presents persecutory delusions may also present themselves as an adult, a suggestive correlation-delusions only make sense for individuals participating in adult style relationships. These children did not have exceptional IQ's as a group, so their adult behavior cannot be interpreted as an artifact of intelligence.

In contrast to persecutory delusions, I argue that somatic delusions are a viable means for children to exploit the investment mechanisms of others. This argument has two parts. First, children require social and material investments, and, as discussed above, somatic delusions are designed to elicit such investments. Second, children are eligible for these investments from close kin. The demographics of somatization disorders supports this hypothesis. 40% of individuals with somatization disorder report onset at age 10 or earlier. 55% report an age of onset of 15 years or earlier (Swartz et al. 1991), a striking contrast to the demographics of DD. It is logical for children to attempt to obtain benefits by mimicking illness, but not by mimicking fears of enemy attack, high status, or sexual relations with high status individuals.

Although I have argued that DD cannot function properly in large Western societies, it should still function in contexts that resemble human social groups during the SEEA, i.e. small, kin-based communities. The mimetic hypothesis predicts that individuals experiencing DD in these societies will receive social benefits. Only a tiny fraction of the world's population currently lives in such societies, however, making this prediction extremely difficult to test. Although the data are sparse, there are a handful of case studies that appear to illustrate aspects of the proper functioning of delusions not seen in Western contexts: close relatives appear to take delusions seriously and often provide benefits as a result: "A 22 year old male Laotion was chronically afraid that certain spirits and deceased ancestors wanted to kill him. He heard the threatening voices of these spirits and at times conversed or argued with them....At times he would act on these paranoid delusions and had to be restrained." The family of this individual in fact believed that spirits with malevolent intentions were speaking to the young man, and perhaps did want to kill him, although they also believed he was baa (insane, Westermeyer 1988). This illustrates that individuals with DD can convince others of the veracity of their delusions.

Eastwell (1982) reports on the Five-Year Epidemiologic Survey of Northern Aboriginal Reserves. Using the diagnostic framework of the International Classification of Disease, 8th revision (ICD 8), 435 individuals out of a total population of 10,500 were identified as suffering from a psychiatric disorder. Of these, 57 were diagnosed as reactive psychosis, or fear-of-sorcery syndrome. Characterized as an anxiety state with paranoid features magnified to psychotic proportions, the patient fears imminent death from sorcery, and the severity of the concurrent autonomic signs is noteworthy. Paradoxically, the actual practice of sorcery is extremely difficult to document among these groups, and some doubt its existence altogether (Berndt and Berndt 1951; cited in Eastwell 1982). Eastwell observes that members of the clan "close ranks [with the patient] in indignation against the putative sorcerers," exactly the outcome predicted by the mimetic hypothesis.

He also discusses the case of a 19 year old who lived in a remote part of the reserve isolated for months at a time by monsoon. He had been having an affair with a young woman forbidden to him by tribal law, and was convinced that he would soon die as a result of sorcery by the tribal elder responsible for bestowing the girl. Noises outside the hospital ward were interpreted as evidence of the sorcerer waiting to kill him, and he was vigilant through the night. A classificatory brother who accompanied him to the hospital felt the affair remained a well kept secret. Those who violate taboos are especially at risk for serious cooperative deficits, since they simultaneously threaten several relationships, perhaps explaining this individual's persecutory delusions.

Waxler (1977) presents a case study of a psychotic episode that may illustrate both the cause and function of delusions. A young Sinhalese woman was given in marriage to a man from an adjacent village, who had chosen her against the wishes of his own family. The women feared that her new sister-in-law was attempting to poison her, and she fled during the night to a nearby temple, where her husband found her the next morning crying and incoherent, and seeing demons and hearing voices. Her father organized small protective rituals, and then arranged an expensive all-night thovil ceremony to which her in-laws as well as local villagers were invited. After the ceremony, her symptoms disappeared, and she went to live with her husband's family with no recurrence in the succeeding five years. Although there is no direct evidence in the case study, it is possible that the women viewed her impending marriage into a family that didn't really want her as a very socially threatening situation-she may have believed she was being excluded from her own family. Her subsequent persecutory delusion that her sister-in-law was attempting to poison her appears to have increased both her and her in-law's connections with her family. Though there is no evidence that her delusion about her sister-in-law was believed, her beliefs about demons seemed to have been taken seriously by her father.

El-Islam (1980) studied the disappearance of delusions among a group of deluded psychotics from the Arab Gulf states. The disappearance of delusions is based on the existence of traditionally shared beliefs. The patient often attributes the disappearance of his delusions to relatives dealing with the object of delusion through prayer or through traditional healers, or the delusion may become "absorbed" into a cultural belief system and lose its force. These phenomena illustrate that close kin may take delusions seriously, and by doing so, alleviate the symptoms. The phenomena where relatives accept an individual's delusional framework is also noted by Murphy (1967; cited in Westermeyer 1988) and El Sendiony (1976).

Cases of Shared Psychosis (Folie à Deux) may provide additional evidence for the ability of individuals with DD to acquire cooperative benefits. According to DSM IV (APA 1994):

The essential feature of Shared Psychotic Disorder is a delusion that develops in an individual involved in a close relationship with another person (sometimes termed the "inducer" or "the primary case") who already has a Psychotic Disorder with prominent delusions.... The [secondary] individual comes to share the delusional beliefs of the primary case in whole or in part....Usually the primary case in Shared Psychotic Disorder is dominant in the relationship and gradually imposes the delusional system on the more passive and initially healthy second person. Individuals who come to share delusional beliefs are often related by blood or marriage and have lived together for a long time, sometimes in relative isolation. If the relationship with the primary case is interrupted, the delusional beliefs of the other individual usually diminish or disappear. Although most commonly seen in relationships of only two people, Shared Psychotic Disorder can occur among a larger number of individuals, especially in family situations in which the parent is the primary case and the children, sometimes to varying degrees, adopt the parent's delusional beliefs.

Thus, it appears that one individual is responsible for creating a delusional framework which forms the basis for cooperation with another. Newhill (1990) presents a case of shared psychosis that also illustrates the receipt of cooperative benefits from relatives. After being placed under a voodoo curse, a Jamaican man had immediately become very sick. His wife, with the help of her family, was able to move both her husband and herself to San Francisco to escape the magic. Although the husband initially experienced some relief, he and his wife soon became convinced that they would die from the magic. The couple became so terrified and agitated that both required immediate admission to the local psychiatric inpatient unit. Separated from her husband, the wife's psychotic symptoms soon cleared, while the husband remained acutely paranoid. No information is given on the husband's social circumstances, or the wife's family's motive for helping the couple move.

Arguing against this interpretation of Shared Psychotic Disorder as an example of the successful functioning of DD are the following facts: first, the disorder appears to be quite rare, although it is likely that many cases go unnoticed (Sacks 1988). Second, many of the secondaries have a preexisting psychiatric disorder, personality disorder, dementia, mental retardation, a physical disability or language barrier (Soni and Rockley 1974). Finally, since individuals with Shared Psychosis are often closely related, there may be genetic factors involved as well (Lazarus 1985). Nevertheless, the receipt of social benefits by deceiving others via a delusional system is recognized by the DSM, and is also supported the very few reports on the social response to delusions in kin-based societies.

According to the mimetic hypotheses, delusions and persecutory fears should remit in individuals who receive sufficient social benefits. Jørgensen and Aagaard (1988) studied the relationship of a number of social variables to impairment, remission, and relapse. They found that being married, living with others, having frequent social contacts, working full-time, and belonging to high status social groups were important predictors of good outcome. On the other hand, the variables of living alone, having few social contacts, and not working prior to admission were by far the best predictors of poor outcome for this group of patients.

Jørgensen and Aagaard conclude that social variables like having social contacts and useful work are more valuable than any of the clinical variables in predicting outcome. Unfortunately, the data presented by these researchers do not favor social causation over social selection theories, but do demonstrate the strong and necessary association of positive social variables with the remission of DD.

There are at least two major problems with the mimetic hypothesis, one theoretical, and one empirical. Theoretically, we should expect individuals who are being exploited by mimicked signals to evolve mechanisms to detect mimesis. Detecting signal mimickry is fairly easy in this case: if many or most individuals reliably start mimicking cooperative signals after they are ostracized, then other group members should just evolve to ignore all cooperative signals of ostracized individuals. There are several solutions to this problem, none completely satisfying. One, it could have been the case that most ostracized individuals did not mimick signals, perhaps because they could successfully switch to another group. Ostracism did not then reliably indicate that an individual was likely to mimic signals. Perhaps only individuals who were ostracized and who couldn't switch to another group would employ mimesis, for example.

Two, because ostracism was a deadly threat, while being exploited by signal mimicry was likely a less-than-deadly threat, the selection pressure on mimetic adaptations was stronger than it was on mimesis detection mechanisms. Mimetic adaptations can then be expected to outperform mimesis detection mechanisms.

Three, individuals who evolve to successfully discriminate against mimics risk inadvertantly discriminating against real signals. Because the benefits received through cooperative signaling are so valuable, individuals may evolve to tolerate some mimetic parasitism rather than risk losing the benefits obtained from geniune signals.

Four, mimetic parasites may attempt to target their signals at individuals who have little or no information concerning the ostracized status of of the mimic. These could include individuals from other groups-delusions would then have to be reconceptualized as a group switching adaptation-or they could be individuals from a competing faction within the group. All groups have internal politics and competing factions. Some factions may be significantly less aware of the ostracized status of an individual than are other factions. An ostracized individual could then exploit the "naive" faction by manipulating their fears of the other factions, for example.

Five, individuals who have committed a serious transgression that has not yet been discovered by other group members could have preemptively employed mimesis in order to divert attention from themselves. For example, if an individual had an affair with the headman's wife, a transgression that, if discovered would have led to ostracism (or death), they could have preemptively diverted the group's attention towards an imaginary threat. Since their transgression has not yet been discovered, there would be no reason to question their signals. In this case, delusions would have to be reconceptualized as an adaptation to prevent, rather than mitigate, ostracism.

Finally, natural selection can lead to stalemates. Perhaps delusions were initially successful at mitigating ostracism, but then individuals evolved to detect signal mimickers. Selection for mimickry would then decrease because it was no longer effective. Selection would then also decrease on mimickry detection, renewing the opportunity for the evolution of mimickry. Mimickry would be maintained in the population even though it was not particulary effective.

Another problem with the mimetic hypothesis involves the content of paranoid delusions. The EMA hypothesis leads one to predict that the content of paranoid delusions should involve threats to the group, but the empirical data on these delusions reveals that the threats are almost always directed at the delusional individual. Group members should not care about threats to an ostracized individual, however, and should be even less inclined to cooperate with them, in fact. Again, there are several somewhat unsatisfactory solutions to this problem. First, and most plausibly, in ancestral environments threats directed at individuals may have been very reliably associated with threats to the group. Intuitively, individuals that have serious criticisms of close family members get very defensive when others criticize those same family members. Paranoid delusions of a personal threat from an outside source may have been able to exploit this ingroup/outgroup psychology.

Second, perhaps delusions about threats to the group were far less plausible than personal threats. This may have especially been the case if mimicks were attempting to exploit individuals who were naive about their ostracized status. The personal nature of delusional threats may be necessary for them to effectively manipulate others.

Finally, I may have misidentified the adaptive function of paranoid delusions. It could be that they functioned like somatic delusions. Instead of attempting to instill fears of an external threat, paranoid delusions could be designed to exploit reciprocal investment mechanisms: help me with this big problem now and I will repay in spades later. The deal may have been too good to pass up, especially if the costs of helping seemed low (not too plausible for a powerful external threat, however).

Resolving these problems (if, indeed, that is possible) requires a far better understanding of ancestral social environments than currently exists.

Social systems that rely on signals for the exchange of substantial benefits are very susceptible to exploitation by parasites that can mimic these signals. For humans, these "parasites" are most likely to be individuals excluded from social exchange since the costs of cheating are small to non-existent for these individuals, and the potential benefits are large. The mimetic hypothesis is consistent with many of the clinical, epidemiological and demographic aspects of Delusional Disorder: the presence of delusions, specific categories of delusions, emotional involvement with delusions, lack of impaired functioning, distribution of age-of-onset, association with immigrant and low socioeconomic status populations, and apparent etiological role of isolation, exclusion, and other social deficits. During our evolutionary history, individuals facing the deadly threat of social exclusion would have had to monitor their social and physical environment very carefully, and may have had no choice but to convincingly mimic, albeit unconsciously, important social cues in order to maintain their status as members of the group. Delusional themes appear designed to generate precisely those cues that would have elicited cooperation from the group: fears of external threat, possession of important information, illness, and intimate relations with high status individuals. Each of these situations would have been difficult for other group members to verify, at least in the short term, making them ideal candidates for exploitative mimesis. Many lines of evidence indicate that social deficits cause delusions, and the sparse data that are available indicate that delusional individuals in small, kin-based societies receive social benefits, at least in the short term. Proof of the mimetic hypothesis will require controlled studies of the etiological role that social deficits play in delusional symptomology, convincing evidence that delusional individuals are frequently able to obtain social benefits in small societies, and the resolution of the two major problems discussed in the previous section.

From a clinical perspective, the mimetic hypothesis has important implications for treatment. First, an individual's assessment of their social environment will need to be obtained. This is non-trivial. Humans may or may not have conscious access to this information, and delusional individuals may have particularly limited access in order to conceal their precarious situation from themselves and others. In addition, current models of human sociality are primitive and untested. The kinds of information that humans use to assess their social environment, and the emotional and other forms in which this information is encoded are largely unknown. Once the social environment is determined, strategies will need to be devised that allow the individual to compete successfully for important social resources. Again, this is non-trivial. It is difficult to regain status after it has been lost. Nevertheless, none of these obstacles is insurmountable, and successful treatment of delusional individuals should be possible in most cases.

Alexander, R.D. (1979) Darwinism and Human Affairs. University of Washington Press.

Alexander, R.D. (1987) The Biology of Moral Systems. Aldine.

APA (1994) Diagnostic and Statistical Manual of Mental Disorders. American Psychiatric Association.

Arthur, B., and Schumann, S. (1970) Family and peer relationships in children with paranoid delusions. Child Psychiatry & Human Development 1:83-101.

Axelrod, R., and Dion, D. (1988) The Further Evolution of Cooperation. Science 242:1385-1390.

Axelrod, R., and Hamilton, W.D. (1984) The Evolution of Cooperation in Biological Systems. In: The Evolution of Cooperation, ed. R. Axelrod. Basic Books, Inc.

Badcock, C. (1990) Oedipus in Evolution. Blackwell.

Berndt, R.M., and Berndt, C.H. (1951) The concept of abnormality in an Australian society. In: Psychoanalysis and Culture, eds. G.B. Wilbur and W. Muensterberger. International Universities Press.

Boulding, K.E. (1962) Conflict and defense: a general theory. Harper.

Bowlby, J. (1969) Attachment and Loss. Basic Books.

Brown, D.E. (1991) Human Universals. McGraw-Hill, Inc.

Bull, J.J., and Rice, W.R. (1991) Distinguishing mechanisms for the evolution of cooperation. Journal of Theoretical Biology 149:63-74.

Cameron, N. (1943) The Development of Paranoic Thinking. Psychological Review 50:219-234.

Carpenter, L., and Brockington, I.F. (1980) A study of mental illness in Asians, West Indians and Africans living in Manchester. British Journal of Psychiatry 137:201-205.

Carpenter, W.T., Strauss, J.S., and Bartko, J.J. (1973) A flexible system for the diagnosis of schizophrenia. Report from the WHO International Pilot Study of Schizophrenia. Science 182:1275.

Chance, M.R.A., ed. (1988) Social Fabrics of the Mind. LEA.

Chiu, L.P., and Rimon, R. (1987) Relationship of migration to paranoid and somatoform symptoms in Chinese patients. Psychopathology 20:203-212.

Cohen, M.N. (1977) The food crisis in prehistory : overpopulation and the origins of agriculture. Yale University Press.

Cooper, A.F., and Curry, A.R. (1976) The pathology of deafness in the paranoid and affective psychoses of later life. Journal of Psychosomatic Research 20:97-105.

Cooper, A.F., Garside, R.F., and Kay, D.W. (1976) A comparison of deaf and non-deaf patients with paranoid and affective psychoses. British Journal of Psychiatry 129:532-538.

Cooper, A.F., and Porter, R. (1976) Visual acuity and ocular pathology in the paranoid and affective psychoses of later life. Journal of Psychosomatic Research 20:107-114.

Coser, L.A. (1957) The Functions of Social Conflict. In: Sociological theory: a book of readings, eds. L.A. Coser and B. Rosenberg. Macmillan.

Dahrendorf, R. (1964) The new Germanies. Encounter 22:50-58.

Dawkins, R., and Krebs, J.R. (1978) Animal Signals: Information or manipulation? In: Behavioral Ecology, eds. J.R. Krebs and N.B. Davies. Blackwell.

Dawkins, R., and Krebs, J.R. (1979) Arms races between and within species. Proceedings of the Royal Society of London B 205:489-511.

Eastwell, H.D. (1982) Psychological Disorders Among the Australian Aboriginals. In: Extraordinary Disorders of Human Behavior, eds. C.T.H. Friedmann and R.A. Faguet. Plenum Press.

El Sendiony, H.F.M. (1976) Cultural aspects of delusions: A psychiatry study of Egypt. Australia New Zealand Journal of Psychiatry 10:201-207.

El-Islam, M.F. (1980) Symptom onset and involution of delusions. Social Psychiatry 15:157-160.

Ettinger, L. (1959) The incidence of mental disease among refugees in Norway. Journal of Mental Science 105:326-338.

Ettinger, L. (1960) The symptomology of mental disease among refugees in Norway. Journal of Mental Science 106:947-966.

Gardner, R.J. (1982) Mechanisms in major depressive disorder: an evolutionary model. Archives of General Psychiatry 39:1436-1441.

Gilbert, P. (1989) Human Nature and Suffering. LEA.

Gillin, J.P. (1934) Crime and Punishment Among the Barama River Carib. American Anthropologist 36:331-344.

Glantz, K., and Pearce, J. (1989) Exiles From Eden: Psychotherapy From An Evolutionary Perspective. Norton.

Goodall, J. (1986) Social rejection, exclusion, and shunning among the Gombe chimpanzees. Ethology & Sociobiology 7:227-236.

Gouldner, A. (1960) The norm of reciprocity: a preliminary statement. American Sociology Review 47:73-80.

Grassian, S. (1983) Psychopathological Effects of Solitary Confinement. American Journal of Psychiatry 140:1450-1454.

Green, S., and Marler, P. (1979) The analysis of animal communication. In: Handbook of Behavioral Neurobiology 3. Social Behavior and Communication, eds. P. Marler and J.G. Vandenbergh. Plenum.

Hamilton, W.D. (1964) The Genetical Evolution of Social Behavior. Journal of Theoretical Biology 7:1-16.

Henderson, S. (1974) Care-eliciting behavior in man. Journal of Nervous Mental Disorders 159:172-181.

Henderson, S. (1981) Social Relationships, Adversity and Neurosis: An Analysis of Prospective Observations. British Journal of Psychiatry 138:391-398.

Henderson, T.E. (1995) Madrid v. Gomez. Federal Supplement 889:1146-1284.

Heron, W., Bexton, W.H., and Hebb, D.O. (1953) Cognitive effects of a decreased variation to the sensory environment. The American Psychologist 8:366.

Hoebel, E.A. (1954) The Law of Primitive Man: A study in Comparative Legal Dynamics. Harvard University Press.

Hölldobler, B. (1977) Communication in social hymenoptera. In: How animals communicate, ed. A. Sebeok. Indiana University Press.

Huxley, J.S., Mayr, E., Osmond, H., and Hoffer, A. (1964) Schizophrenia as a genetic morphism. Nature 204:220-225.

Jorgensen, P., and Aagaard, J. (1988) A multivariate predictor analysis of course and outcome in delusional psychosis. Acta Psychiatrica Scandinavica 77:543-550.

Jorgensen, P., and Jensen, J. (1988) An attempt to operationalize reactive delusional psychosis. Acta Psychiatrica Scandinavica 78:627-631.

Kaffman, M. (1981) Paranoid disorders: The core of truth behind the delusional system. International Journal of Family Therapy 3:29-41.

Kay, D.W., Cooper, A.F., Garside, R.F., and Roth, M. (1976) The differentiation of paranoid from affective psychoses by patients' premorbid characteristics. British Journal of Psychiatry 129:207-215.

Keeley, L.H. (1996) War Before Civilization. Oxford University Press.

Kendler, K.S. (1980) The nosologic validity of paranoia (simple delusional disorder): A review. Archives of General Psychiatry 37:699-706.

Kendler, K.S. (1982) Demography of paranoid psychosis (delusional disorder): A review and comparison with schizophrenia and affective illness. Archives of General Psychiatry 39:890-902.

Kendler, K.S. (1984) Paranoia (delusional disorder): A valid psychiatric entity? Trends in Neurosciences 7:14-17.

Kendler, K.S. (1987) Paranoid disorders in DSM-III: A critical review. In: Diagnosis and classification in psychiatry: A critical appraisal of DSM-III., ed. E. Gary L. Tischler. Cambridge University Press, New York, NY, US.

Kendler, K.S., and Tsuang, M.T. (1981) Nosology of paranoid schizophrenia and other paranoid psychoses. Schizophrenia Bulletin 7:594-610.

Kennedy, H.G., Kemp, L.I., and Dyer, D.E. (1992) Fear and anger in delusional (paranoid) disorder: The association with violence. British Journal of Psychiatry 160:488-492.

Koehler, K., and Hornstein, C. (1986) 100 years of DSM-III paranoia: How stable a diagnosis over time? European Archives of Psychiatry & Neurological Sciences 235:255-258.

Kraepelin, E. (1915) Der Verfolgungswahn der Schwerhörigen. In: Psychiatrie, ed. . Barth.

Kraepelin, E. (1971) Praecox and Paraphrenia. Robert E Krieger Publishing Co Inc.

Lazarus, A. (1985) Folie à Duex: Psychosis by Association or Genetic Determinism? Comprehensive Psychiatry 26:129-135.

Lee, R.B. (1993) The Dobe Ju/'hoansi. Harcourt Brace.

Lee, R.B., and DeVore, I., eds. (1968) Man the hunter. Aldine Publishing Company.

Leff, J.P. (1974) Transcultural influence on pyschiatrists' rating of verbally expressed emotion. British Journal of Psychiatry 125:336.

Lemert, E.M. (1962) Paranoia and the Dynamics of Exclusion. Sociometry 25:2-20.

LeVine, R.A., and Campbell, D.T. (1972) Ethnocentrism: Theories of Conflict, Ethnic Attitudes, and Group Behavior. John Wiley & Sons, Inc.

Lewis, W.H. (1961) Feuding and social change in Morocco. Journal of Conflict Resolution 5:43-54.

Lilly, J.C. (1956) Mental Effects of Reduction of Ordinary Levels of Physical Stimuli on Intact, Healthy Persons. Psychiatric Research Reports 5:1-9.

Mack, R.W., and Synder, R.C. (1957) The analysis of social conflict-toward an overview and synthesis. Journal of Conflict Resolution 1:212-248.

Manschreck, T.C. (1989) The paranoid syndrome and delusional (paranoid) disorders. In: Outpatient psychiatry: Diagnosis and treatment, ed. A. Lazare. Williams & Wilkins Co.

Markl, H. (1983) Vibrational Communication. In: Neuroethology and Behavioral Physiology, eds. F. Huber and H. Markl. Springer.

Markl, H. (1985) Manipulation, modulation, information, cognition: some of the riddles of communication. In: Experimental Behavioral Ecology and Sociobiology, eds. B. Hölldobler and M. Lindauer. Gustav Fischer Verlag.

Meissner, W.W. (1987) The diagnosis of paranoid disorders. In: Diagnostics and psychopathology. Directions in psychiatry monograph series, No. 1., ed. F. Flach. W. W. Norton & Co, Inc.

Mirowsky, J., and Ross, C.E. (1983) Paranoia and the structure of powerlessness. American Sociological Review 48:228-239.

Moore, N.C. (1981) Is paranoid illness associated with sensory defects in the elderly? Journal of Psychosomatic Research 25:69-74.

Murphy, H.B.M. (1967) Cultural aspects of delusion. Stadium Generale 20:684-692.

Murphy, R.F., and Kasdan, L. (1959) The structure of parallel cousin marriage. American Anthropologist 61:17-29.

Ndetei, D.M., and Vadher, A. (1984) Frequency and clinical significance of delusions across cultures. Acta Psychiatrica Scandinavica 70:73-76.

Nesse, R. (1990) The evolutionary functions of repression and the ego defenses. Journal of the American Academy of Psychoanalysis 18:260-285.

Nesse, R. (1991) What Good Is Feeling Bad - The Evolutionary Benefits Of Psychic Pain. Sciences 31:30-37.

Newhill, C.E. (1990) The role of culture in the development of paranoid symptomatology. American Journal of Orthopsychiatry 60:176-185.

Nitsche, P., and Wilmanns, K. (1912) The History of the Prison Psychoses. The Journal of Nervous and Mental Disease Publishing Company.

Nöe, R. (1992) Alliance formation among male baboons: shopping for profitable partners. In: Coalitions and Alliances in Humans and other Animals, eds. A.H. Harcourt and F.B.M. de Waal. Oxford University Press.

Nöe, R., and Hammerstein, P. (1994) Biological markets: supply and demand determine the effect of partner choice in cooperation, mutualism and mating. Behavioral Ecology and Sociobiology 35:1-11.

Nöe, R., van Schaik, C.P., and van Hoof, J.A.R.A.M. (1991) The Market Effect: an Explanation for Pay-off Asymmetries among Collaborating Animals. Ethology 87:97-118.

Ødegaard, Ø. (1932) Immigration and insanity. Acta psychiatrica neurologica scandinavia supplement 4.

Oltmanns, T.F. (1988) Approaches to the definition and study of delusions. In: Delusional beliefs. Wiley series on personality processes., eds. T.F. Oltmanns and B.A. Maher. John Wiley & Sons, New York, NY, US.

Opjordsmoen, S. (1987) Toward an operationalization of reactive paranoid psychoses (reactive delusional disorder). World Psychiatric Association Symposium (1986, Copenhagen, Denmark). Psychopathology 20:72-78.

Price, J.S. (1972) Genetic and phylogenetic aspects of mood variation. International Journal of Mental Health 1:124-144.

Retterstöl, N. (1966) Paranoid and Paranoiac Psychoses. Charles C. Thomas.

Rich, K.C., Siegel, J.N., Jennings, C., Rydman, R.J., and Landay, A.L. (1995) CD4+ lymphocytes in perinatal human immunodeficiency virus (HIV) infection: evidence for pregnancy-induced immune depression in uninfected and HIV-infected women. Journal of Infectious Diseases 172.

Robins, L.N., Locke, B.Z., and Regier, D.A. (1991) An Overview of Psychiatric Disorders in America. In: Psychiatric Disorders in America: The Epidemiologic Catchment Area Study, eds. L.N. Robins and D.A. Regier. Macmillan.

Robins, L.N., and Regier, D.A., eds. (1991) Psychiatric Disorders in America: The Epidemiologic Catchment Area Study. Macmillan.

Sacks, M.H. (1988) Folie à Deux. Comprehensive Psychiatry 29:270-277.

Sahlins, M.D. (1972) Stone Age Economics. Tavistock.

Schrire, C. (1980) An Inquiry into the Evolutionary Status and Apparent Identity of San Hunter-Gatherers. Human Ecology 8:9-33.

Sherif, M., Harvey, O.J., White, B.J., Hood, W.R., and Sherif, C.W. (1961) Intergroup Conflict and Cooperation: The Robbers' Cave Experiment. University of Oklahoma Press.

Sherif, M., and Sherif, C.W. (1953) Groups in harmony and tension. Harper.

Simmel, G. (1955) Conflict and the web of group affiliations. The Free Press.

Slavin, M., and Kriegman, D. (1992) The Adaptive Design Of The Human Psyche: Psychoanalysis, Evolutionary Biology, And The Therapeutic Process. Guilford Press.

Slavin, M.O. (1985) The origins of psychic conflict and the adaptive function of repression: An evolutionary biological view. Psychoanalysis and Contemporary Thought 8:407-440.

Solway, J., and Lee, R.B. (1990) Foragers, Genuine or Spurious? Current Anthropology 31:109-146.

Soni, S.D., and Rockley, G.J. (1974) Socio-Clinical Substrates of Folie à Deux. British Journal of Psychiatry 125:230-235.

Starrett, A. (1993) Adaptive resemblance: a unifying concept for mimicry and crypsis. Biological Journal of the Linnean Society 48:299-317.

Sullivan, H.S. (1953) The interpersonal theory of psychiatry. Norton.

Sumner, W.G. (1906) Folkways. Ginn.

Swartz, M., Landerman, R., George, L.K., Blazer, D.G., and Escobar, J. (1991) Somatization Disorder. In: Psychiatric Disorders in America: The Epidemiologic Catchment Area Study, eds. L.N. Robins and D.A. Regier. The Free Press.

Symons, D. (1979) The Evolution of Human Sexuality. Oxford University Press.

Szasz, T.S. (1961) The Myth of Mental Illness. Dell Publishing Co.

Thornhill, N., and Thornhill, R. (1990) An Evolutionary Analysis Of Psychological Pain Following Rape. 1. The Effects Of Victims Age And Marital Status. Ethology And Sociobiology 11:155-176.

Tooby, J., and Cosmides, L. (1990) The past explains the present: Emotional adaptations and the structure of ancestral environments. Ethology & Sociobiology 11:375-424.

Tooby, J., and Cosmides, L. (1996) Friendship and the Banker's Paradox: Other pathways to the evolution of adaptations for altruism. In: Proceedings of the British Academy: Evolution of Social Behavior Patterns in Primates and Man, ed. J. Maynard Smith.

Trivers, R.L. (1971) The Evolution of Reciprocal Altruism. The Quarterly Review of Biology 46:35-57.

Trivers, R.L. (1985) Social Evolution. Addison-Wesley.

van den Berghe, P.L. (1990) Human Family Systems: An Evolutionary View. Waveland Press, Inc.

Wallace, A.F.C. (1960) The Biocultural Theory of Schizphrenia. International Record of Medicine 173:700-714.

Watt, J.A. (1985) Hearing and premorbid personality in paranoid states. American Journal of Psychiatry 142:1453-1455.

Waxler, N.E. (1977) Is Mental Illness Cured in Traditional Societies? A Theoretical Analysis. Culture, Medicine, and Psychiatry 1:233-253.

Wessely, S., Buchanan, A., Reed, A., Cutting, J., and others, a. (1993) Acting on delusions: I. Prevalence. British Journal of Psychiatry 163:69-76.

Westermeyer, J. (1988) Some cross-cultural aspects of delusions. In: Delusional beliefs, eds. T.F. Oltmanns and B.A. Maher. John Wiley & Sons.

Westermeyer, J. (1989) Paranoid symptoms and disorders among 100 Hmong refugees: A longitudinal study. Acta Psychiatrica Scandinavica 80:47-59.

Westermeyer, J., and Sines, L. (1979) Reliability of cross-cultural psychiatric diagnosis with an assessment of two rating contexts. Journal of Psychiatric Research 15:199-213.

White, L.A. (1949) The science of culture: a study of man and civilization. Farrar, Straus.

Williams, G. (1966) Adaptation and Natural Selection: A Critique of Some Evolutionary Thought. Princeton University Press.

Wilmsen, E.N., and Denbow, J.R. (1990) Paradigmatic History of San-speaking Peoples and Current Attempts at Revision. Current Anthropology 31:489-524.

Wilson, D.A. (1993) Evolutionary Epidemiology - Darwinian Theory In The Service Of Medicine And Psychiatry. Acta Biotheoretica 41:205-219.

Winokur, G. (1977) Delusional disorder (paranoia). Comprehensive Psychiatry 18:511-521.

Winters, K.C., and Neale, J.M. (1983) Delusions and delusional thinking in psychotics: A review of the literature. Clinical Psychology Review 3:227-253.

Ziskind, E. (1958) Isolation Stress in Medical and Mental Illness. The Journal of the American Medical Association 168:1427-1431.

Notes

1. Certain general elements of this hypothesis have been argued by others. Szasz (1961) has argued that "mental illness" is a form of deception, for example. The only previous (brief) suggestion that I have encountered that psychoses function to prevent social exclusion is Wallace (1960). He presents no rationale for this function however. Henderson (1974; 1981) has carefully investigated the hypothesis that "neuroses" (though not psychoses) function to elicit care. Sullivan (1953) is well known for his interpersonal approach to psychiatry in general.

2. The problem of "cheating" is dealt with extensively in the literature on reciprocal altruism and cooperation. Exclusion from future cooperative opportunities is generally viewed as a successful counter-strategy to cheating.

3. Certain species of beetles have evolved to mimic ant feeding signals in order to receive food.

4. Exploitative mimicry generally has a negative frequency dependent relation with fitness, which allows for the existence of polymorphisms--individuals wishing to successfully exploit others should pick strategies unused by other similarly situated individuals in their local environment.

5. These are the DSM IV criteria. Older DSM III criteria are still commonly encountered.

6. The difficulty in defining delusions centers on what is meant by "false." A person who holds a belief commonly accepted by his or her community is not considered delusional, for example, even if Western psychiatrists would consider this belief to be false. The number of categories of delusions in Delusional Disorder is actually quite limited. This largely eliminates concerns about the definition of what constitutes a delusion.

7. I have omitted his first two criteria specifying that the belief is not shared by others.

8. I consider the vigilance hypothesis very important for understanding the full spectrum of "paranoid" symptoms. Paranoid individuals often fear members of their in-group--family members for example--in apparent contradiction to the mimetic hypothesis. In-group members can be very dangerous to individuals in weak social positions, however, and must be carefully monitoredÑvigilant paranoia rather than deceptive (delusional) paranoia. I would predict that delusions regarding dangerous in-group members should attempt to link them to external threats, perhaps "tainting" them in the eyes of other in-group members.

9. These are the DSM IV categories. Delusions of reference are commonly mentioned also. These refer to delusions in which the individual feels that events have a special personal significance. DSM IV appears to have subsumed these under the grandiose, persecutory or other categories. Same for religious.

10. Sample included individuals diagnosed with schizophrenia and affective psychosis.

11. Cults may be the exception. Cult leaders may be, in some cases, paranoid and/or grandiose individuals exploiting lonely and vulnerable people.

12. Jorgensen and Aagaard claim that according to both ICD 8 and DSM III the majority of patients were not schizophrenic or suffering from affective psychosis, although some did experience hallucinations.

13. Median age for first symptom of any psychiatric disorder in the Epidemiologic Catchment Area Study is 16, although this study included alcoholism and drug abuse as psychiatric disorders.

14. I did find a single case study that claimed the 10 year old was the primary and the parent the secondary.

15. Arthur and Schumann note that the parents of these children showed obvious psychopathology.

16. A DSM III category, simplified and renamed Somatoform Disorder in DSM IV.

17. The World Health Organization (WHO) analog of the DSM.

18. A likely case of Brief Psychotic Disorder under DSM IV. Relationship to DD unclear.