Niche Construction, Biological Evolution and Cultural Change.

The conceptual model in Figure 2b extends Figure 2a, yet it still over simplifies the causal pathways connecting genes and culture, because it requires that cultural inheritance should affect the fate of some human genes directly, in the absence of any other mediating process. In most cases where gene-culture coevolutionary theory has been applied, this assumption is reasonable. Culture may bias human mating patterns non randomly, it may bias other human interactions, such as trade or warfare, or it may bias the choice of which infants are selected for infanticide (Boyd & Richerson, 1985; Laland, 1994; Kumm et al., 1994). The assumption that human cultural inheritance can directly bias human genetic inheritance may also be acceptable even when the source of the natural selection pressure that is modified by culture is no longer human, provided the relationship between whatever cultural trait is being expressed, and whatever natural selection pressure it is modifying, is sufficiently direct. For example, the trait that affected human genetic evolution in the lactose tolerance case was milk usage (Durham, 1991). Here, gene-culture theory is again applicable, because the link between milk usage and its genetic consequences are sufficiently simple to allow it to be modelled without bringing in any intermediate variables (Feldman & Cavalli-Sforza, 1989) .

1.2.3. Gene-Culture Coevolution plus Niche Construction: The gene-culture coevolutionary approach, however, fails in more complicated situations. Take, for example, the case of Kwa-speaking yam cultivators in West Africa, who increased the frequency of a gene for Sickle-cell anaemia in their own population as a result of the indirect effects of yam cultivation. These people traditionally cut clearings in the rainforest, creating more standing water and increasing the breeding grounds for malaria carrying mosquitoes. This, in turn, intensifies selection for the Sickle-cell allele, because of the protection offered by this allele against malaria in the heterozygotic condition (Durham 1991). Here the causal chain is so long that simply plotting the cultural trait of yam cultivation against the frequency of the Sickle-cell allele would be insufficient to yield a clear relationship between the cultural trait and allele frequencies (Durham 1991). The crucial variable is probably the amount of standing water in the environment caused by the yam cultivation, but standing water is an ecological variable, not a cultural variable, and it partly depends on factors (i.e. rainfall) that are beyond the control of the population. So here the simplifying assumption of a direct link between cultural and genetic inheritance distorts reality too much to allow their interaction to be modelled in the standard way. This time the two human inheritance systems can only interact via an intermediate, abiotic, ecological variable, which should be included to complete the model.

This shortcoming leads us to propose an extended gene-culture coevolutionary theory, a conceptual version of which is shown in Figure 2c. The novelty here is the replacement of the genetic-inheritance scheme, described by standard evolutionary theory (Figure 1a), as the proper basis of gene-culture coevolution, by the extended evolutionary scheme incorporating niche construction, summarised in Figure 1b. Thus in Figure 2c, niche construction from all ontogenetic and cultural processes modifies human selective environments. Culturally modified selection pressures are now regarded, not as unique, but as just a part of a more general legacy of modified natural selection pressures which are bequeathed by human ancestors to their descendants. Hence, instead of being exclusively responsible for allowing us to co-direct our own evolution in contrast to what happens in every other species, culture now becomes merely the principal way in which we humans do the same thing that most other species do.

We now take a closer look at the set of processes by which populations of complex organisms, such as humans, acquire adaptive information, and how this information is expressed in niche construction. It is now widely recognised that several of the major evolutionary transitions involved changes in the way information is acquired, stored and transmitted (Szathmary & Maynard Smith, 1995). This is reflected in Figure 3, where we acknowledge that populations of complex organisms can acquire relevant semantic "information", or more accurately "knowledge" (Holland, 1992), through a set of information-acquiring processes (Holland, 1992), operating at three different levels, and with the knowledge gained being influenced by niche-constructed environments at each level. In various combinations, these are the processes which supply all organisms with the knowledge that organises their adaptations. Every species is informed by naturally selected genes, many are also informed by complex, information-acquiring ontogenetic processes such as learning or the immune system, while hominids, and perhaps a few other species, are also informed by culture. It is generally recognised that any comprehensive treatment of the gene-culture relationship requires the inclusion of all three sets of processes because the links between genetic evolution and culture cannot be understood without some reference to the intermediate ontogenetic processes, such as individual learning, that connect them (Plotkin & Odling-Smee 1981; Boyd & Richerson, 1985; Durham, 1991; Feldman & Laland, 1996).

Figure 3: This figure zooms in on one of the boxes labelled "populations of phenotypes" in Figure 2. Human evolution results from information-acquiring processes at three levels. Adaptation in populations of complex organisms, such as humans, depends on population genetic processes, information-acquiring ontogenetic processes, and cultural processes, all of which can generate niche construction.

The most phylogenetically ancient process ultimately responsible for niche construction is genetic evolution. As a consequence of the differential survival and reproduction of individuals with distinct genotypes, genetic evolution results in the acquisition, inheritance and transmission of genetically encoded "knowledge" by individuals in populations. Each individual inherits this genetic information from its ancestors, which then translates into developmental processes, expressing different phenotypes in different environments, the so-called "norm of reaction". Each individual may also contribute to the modification of its population’s selective environment by genetically guided niche construction.

Many species have also evolved a set of more complicated ontogenetic processes, that allow individual organisms to acquire another kind of information. These processes are themselves products of genetic evolution, but are nevertheless distinct from it. They comprise "facultative" or "open" developmental processes, based on specialised information-acquiring subsystems in individual organisms, such as the immune system in vertebrates, or brain-based learning in animals, and they are capable of additional, individually-based, information acquisiton. Here the information acquiring entity is no longer an evolving population but is instead each individual organism in a population. As a result, the adaptive knowledge acquired through these ontogenetic processes cannot be inherited because all the knowledge gained by individuals during their lives is erased when they die. Nonetheless, processes such as learning can still be of considerable importance to subsequent generations because learned knowledge can guide niche construction.

A few species, including many vertebrates, have also evolved a capacity to learn from other individuals, and to transmit some of their own learned knowledge to others. In humans this ability is facilitated by a further set of processes (e.g. language), which collectively underlie culture. Culture adds a second knowledge inheritance system to the evolutionary process through which socially learned information is accrued, stored, and transmitted between individuals. Here, the information-acquiring entity is again a group of interacting organisms rather than an individual. Although all cultural knowledge is traceable to the innovation and learning by particular individuals (with ontogenetic processes the ultimate source), major cultural changes may also occur through learning from neighbouring groups or immigrants, and may come with a baggage of associated ideological or organisational requirements.

Within a population, individuals share at least some of their learned knowledge with others, within and between generations. This information sharing can depend on several kinds of cultural inheritance, including vertical (from parents), horizontal (from peers), oblique (from unrelated older individuals), indirect (e.g. from key individuals) and frequency-dependent (e.g. from the majority) cultural transmission systems (Cavalli Sforza & Feldman, 1981; Boyd & Richerson, 1985). Cultural inheritance therefore requires at least one non-genetic channel of communication among organisms via which learned knowledge can be shared and spread. It probably also requires that organisms can decompose their store of cultural "knowledge" into discrete transmittable "chunks" (Feldman & Cavalli-Sforza, 1976), or "memes" (Dawkins, 1989), perhaps equivalent to psychologist’s "schemata", either in simple or compound form (Holland, et al., 1986; Plotkin, 1996). In practice, in every species that is capable of sharing learned information, cultural inheritance depends on some kind of social learning (Galef, 1988; Durham, 1991), while in humans, it may also depend on language (Pinker, 1994).

Much of human niche construction is dominated by socially learned knowledge and cultural inheritance, but the transmission and acquisition of this knowledge is itself dependent on pre-existing information acquired through genetic evolution or complex ontogenetic processes. Thus, while the variants that occur during genetic evolution, (i.e. mutations), are random (or at least, blind relative to natural selection), those acquired through ontogenetic processes are not. As well as being selected by ontogenetic processes, any knowledge gain by individuals is guided by genetic information. For example, animals are genetically predisposed to respond to both specific internal cues (e.g. hunger) and environmental contexts (e.g. sensory cues indicating food is nearby) by generating appropriate behaviour patterns from their repertoires. Hence, during learning, animals typically demonstrate "a priori" biases in their associations and patterns of behaviour that are most likely to be adaptive (Seligman 1970; Bolles, 1970). In addition, those associations and patterns of behaviour that animals do learn critically depend on which stimuli are perceived as reinforcing (pleasant or painful) under the influence of species-specific motivational systems, and these perceptual and motivational processes are constrained by genes (Hinde & Stevenson Hinde, 1973; Plotkin & Odling-Smee, 1981). This means that the behaviours of an animal, the associations it forms, the antibodies it generates, the developmental pathways it takes, are typically, although not universally, functional and adaptive. By the same reasoning, as well as being selected by cultural processes, cultural knowledge is guided and constrained by both genetic information and by ontogenetic processes (Odling-Smee, 1994). Social learning and transmission are affected by individual reinforcement histories and past associations partly because the cultural selective processes are themselves guided by what Durham (1991) calls "primary (or developmental) values", as well as by socially transmitted cultural values. Therefore, with some caveats that we discuss in the final section, we expect that the ideas, values, and acquired knowledge that comprise a culture would usually be adaptive.

We have now come some way from the simple, sociobiological descriptions of gene-culture interactions, captured by Figure 2a. We have brought together two different bodies of theory, gene-culture coevolution and niche construction. As a result, there is a proliferation of interactions between niche-constructing processes, biological evolution and culture, shown in Figure 2c. This proliferation is summarised in Table 1, which illustrates each interaction with an example, organised in terms of the sources and consequences of niche construction.

Source of Niche Construction	Feedback to Biological Evolution	Feedback to Cultural Change
Population Genetic Processes	eg. web spiders marking web or building dummy spiders (Edmunds, 1974)	eg. sex differences in human mating behaviour (Daly & Wilson, 1983, Barkow et al., 1992)
Information acquiring Ontogenetic Processes	eg. woodpecker finch, by learning to grub with a tool, alleviates selection for a woodpecker's bill (Alcock, 1972, Grant, 1986)	eg. learning and experience, influence the adoption of cultural traits (Durham, 1991)
Cultural Processes	eg. dairy farming selects for lactose tolerance (Feldman & Cavalli-Sforza, 1989)	eg. the invention of writing leads to other innovations such as printing, libraries, e-mail

Table 1: Niche construction resulting from population genetic processes, information-acquiring ontogenetic processes, and cultural processes, can influence both biological evolution and cultural change. The point is illustrated by a single example in each cell.

We believe that due recognition of the role of niche construction in the evolutionary dynamic should advance our understanding of the relationship between human culture and human genetic evolution. As such, our perspective may be regarded as part of a movement working towards a framework for integrating biology and the behavioural sciences (Hinde 1987; Durham, 1991).

What happens to those evolutionary and cultural issues that concern the human sciences when our new framework is substituted for the standard evolutionary perspective? In this section we will start to answer this question by describing some examples of how the feedback that occurs in our extended version of gene-culture coevolution might work. The examples we have chosen are hominid evolution, altruism and cooperation, and the processes of human adaptation.

Archeologists and anthropologists currently seek to reconstruct the evolutionary history of modern humans from the fossil and molecular data, in the context of standard evolutionary theory (Figure 1a). Since this theory does not incorporate niche construction, it encourages the idea that human evolution must have been directed solely by independent natural selection pressures in human selective environments, that is by selection pressures that have not been modified by niche construction. These selection pressures may sometimes include those arising from hominid social interactions (e.g. Byrne & Whiten, 1988; Durham 1991; Dunbar, 1993; Foley 1995), but they do not include other sources of selection in the external environment that have been modified by ancestral hominid niche constructors. When human adaptation is treated not only as dependent on natural selection, but also on niche construction (Figure 2c), the suite of hypotheses about the causes, rates and processes of evolutionary change is considerably enlarged.

2.1.1. Processes of Human Evolution: Consider the possible ways in which a new evolutionary episode might be initiated in hominid evolution. For illustrative purposes only we consider a suite of explanations for the divergence of the lineages leading to the Pongidae and Hominidae families in the late Miocene, and we assume two ancestral populations in allopatry. The subsequent divergence of these two populations could have been triggered by any of the following events. First, each population could have been exposed to different external environments with different selection pressures, leading to allopatric speciation in the manner proposed in standard evolutionary theory. Alternatively, both populations may have been exposed to the same novel selective pressures, say a changed habitat, but only one population, say the ancestors of the Pongidae, was able to respond with counteractive niche construction by retreating to a still unchanged habitat, while the Hominidae ancestors remained where they were and became adapted to the new environment. Third, both populations may have responded to the same novel selection pressures with counteractive niche construction, but in different ways, each subsequently generating a different array of novel modified selection pressures which fed back on themselves. Fourth, the divergence might have been initated by inceptive niche construction on the part of one population, by which we mean by a novel form of niche construction, initiated by a change in organisms in one population, possibly because of a mutation, or a new cultural discovery, which subsequently caused, rather than resulted from, a change in the environment. For instance, this might have entailed one population discovering a new habitat, or discovering a new form of niche construction, most likely because of the spread of a newly learned behaviour (West-Eberhard, 1987; Bateson, 1988; Plotkin, 1988). Culture, or proto-culture, say on the part of the Hominidae, may have initiated some novel biological evolutionary change (Feldman & Cavalli Sforza, 1976; Boyd & Richerson, 1985; Wilson, 1985). Fifth, the ancestors of both lineages may have initiated different kinds of inceptive niche construction, again with no key environmental event triggering their divergence.

This enlarged suite of processes operating in hominid evolution raises the possibility that some new traits might pay for their own fitness costs through niche construction. One possible example is the evolution of the (large) human brain. The mass-specific metabolic rate of the human brain is about nine times higher than the average metabolic rate of the human body as a whole, but there is no elevated basal metabolic rate in humans which would pay for it (Aiello & Wheeler, 1995). Aiello and Wheeler (1995) found that this was possible because the human gut, and in particular, the gastro-intestinal tract, requires fewer energetic resources. They hypothesized that our ancestors could afford a reduction in gut size because they used their brains to improve their diets in proportion to their loss of gut. Aiello and Wheeler suggest that this probably happened in two different episodes of brain evolution, the first coinciding with the appearance of the genus Homo, approximately two million years ago, and supported by increased meat eating, the second coinciding with the appearance of archaic Homo sapiens during the latter half of the Middle Pleistocene, and supported by the cultural invention of cooking, and therefore by the externalisation of part of the digestive processes. This is an example of how a trait, the human brain, might have evolved despite fitness costs by paying for itself by its "inventive" niche construction. Big brains would not be adaptive without niche construction.

If niche construction were an important evolutionary agent, then for any clade of organisms, it should also be possible to predict a priori which phenotypic traits (which we will call "recipient characters" because they are receptive of selection pressures that have been modified by niche construction) might have been selected as adaptive in environments that have been niche constructed. Pertinent characters, and environmental states, could be measured in populations of closely-related organisms that do and do not exhibit this niche construction. It would then be possible to use comparative methods to determine if a selected recipient character change correlates with a particular niche construction activity, if the niche constructing activity is ancestral to the recipient character, and whether the recipient character in question is derived. If we are correct, there should be a significant relationship between the pertinent environmental state and the recipient character only when the niche constructing activity is also present. Since the same logic applies at the cultural level, this method could be applied to hominids, or contemporary human populations, where it may shed light on the relationship between particular genes and memes. For instance, in the Kwa, there is only a strong correlation between amount of standing water and incidence of Sickle-cell anaemia in populations that grow yams. In this case, a cultural practice has left a measurable genetic signature, in the form of a different allele frequency. In theory, it is therefore possible that genetic signatures for other cultural practices, evident in archaeological or ethnographic records, could be identified and used as evidence for the presence or absence of the cultural trait in particular populations, or to trace the diffusion of the cultural practice across geographic regions. If so, advances in molecular techniques could eventually aid this line of inquiry.

2.1.2. Rates of evolution: Niche construction may also have influenced the rate of hominid evolution. Much attention has focused on how cultural transmission affects evolutionary rates. Allan Wilson and colleagues have argued that changes in niche, resulting from complex social behaviour and cultural (or proto-cultural) transmission, may generate a "behavioral drive", which accelerates morphological evolution by fixing a greater proportion of genetic mutations (Wilson 1985). Wilson notes that there is a monotonic relationship between relative brain size and rate of anatomical evolution among vertebrates, which he argues is consistent with his behavioral drive hypothesis. However, theoretical analyses suggest that cultural processes may act both to accelerate and decelerate evolution (Feldman & Cavalli-Sforza 1976). These apparently contradictory findings make better sense in the light of our new perspective, because culture is a powerful medium for human niche construction, and niche construction can both counteract and support evolutionary change. If cultural innovations modify natural selection pressures, then genetic change due to modified natural selection is likely to follow. If, as seems likely, the rate of change of cultural niche construction is rapid relative to independent changes in the environment, biological evolutionary rates may be accelerated. A number of gene-culture coevolutionary models have found that, as cultural transmission may homogenise a population’s behaviour, and because culturally transmitted traits can spread through populations rapidly compared with genetic variants, culture can generate atypically strong selection (Feldman & Laland, 1996).

It is widely recognised that culture can also shield genetic variants of low fitness from selection (Boyd & Richerson, 1985; Feldman & Laland, 1996). For instance, improved levels of health care and sanitation are examples of culturally mediated counteractive niche construction that damp out selection against individuals with some gene-related disorders, who may survive and reproduce in the modified environment. In fact, ontogenetic processes may also damp out selection, for example, if individuals develop antibodies that counter disease, or learn to avoid parasites or predators (Bateson, 1988; Plotkin, 1988). In addition, the recent culturally enhanced mobility of peoples, facilitates greater mixing of genes between populations, eradicating differences, and slowing down the divergence of populations. Moreover, a new culturally induced environmental change may be responded to exclusively by a new cultural adaptation. For example, even though smoking during pregnancy probably has a significant effect on the survival rate of offspring, the spread of tobacco smoking is unlikely to select for genes for resistance to smoking-related disease, because advances in medical technology allow smokers and their offspring to survive, and because campaigns to prohibit smoking are increasing awareness of its dangers. Under such circumstances, culture is unlikely to affect the rate of genetic evolution.

More generally, organisms have evolved many niche-constructing behaviours that allow them to regulate the environment in such a way as to buffer out particular natural selection pressures. Niche construction that mitigates a selection pressure may allow populations to maintain greater levels of genetic variation at those loci that would have been affected by selection had the population not expressed that particular niche-constructing traits, because it shields such variation from selection. For example, in mammals genetic variation in the ability to deal with heat, through body size, or shape of ears or tail, may be exposed to less intense selection in populations that escape extreme temperatures in burrows, than in those that do not. However, if the counteractive niche construction breaks down, say a new predator forces the mammals into the open, the presence of significant levels of variation in genes affecting heat exchange may facilitate rapid genetic evolution. In other words, for specific traits, counteractive niche construction may sometimes facilitate periods of evolutionary stasis, punctuated by rapid genetic change. Moreover, following such change, since the niche construction of many organisms, particularly "keystone" species, modifies the selective environments of other species, subsequent niche construction could trigger a cascade of evolutionary events that realign ecosystems (Jones et al., 1997). Although we do not anticipate all macroevolutionary patterns to be dominated by punctuated equilibria, niche construction does provide a novel, readily observable, and testable micro-evolutionary process to account for punctuated macro-evolutionary trends in particular traits, frequently observed in the fossil record (Eldredge & Gould, 1972).

The particular significance of this for human evolution is that, as unusually potent niche constructors, hominids should be particularly resistant to genetic evolution in response to changing environments, while at the same time capable of dramatic evolutionary change following major innovations. If we assume that hominid niche construction is more flexible than that of other mammals, and that culture enhances the capacity of humans to alter their niches, such that the more technically advanced a culture the greater its capacity for counteractive niche construction, then a number of hypotheses follow. First, consider Vrba’s (1992) hypothesis of "turnover pulses". We would expect hominids to show less response to fluctuating climates than other mammals. We would also expect more technologically advanced hominids to exhibit less of a response to fluctuating climates than less technologically advanced homininds. Second, consider Bergmann’s and Allen’s Rules. These rules suggest, respectively, that populations in warmer climates will be smaller bodied and have bigger extremities than those in cooler climates. Again we would expect hominids to show less correspondence to these rules than other mammals. We would also expect more technically advanced humans (e.g. moderns) to exhibit less correspondence to these rules than less technically advanced humans (e.g. Neanderthals), assuming that the latter must have been less well equipped than the former to invest in counteractive niche construction. Third, by the same logic, we would expect an inverse relationship between "robusticity" and the capacity for expressing counteractive niche construction. Fourth, it should be possible to reverse the inference, and to use the fossil record to infer something about the niche constructing capabilities of animals, including hominids. Here we suggest the greater the phenotypic (as opposed to extended phenotypic) response to environmental change by hominids, the more restricted must have been their capacity for niche construction. We are well aware that some related ideas have been proposed before (see Lewin, 1998), but we think that our niche construction perspective could provide a basis for new and much more detailed predictions along these lines, based on a more comprehensive understanding of the underlying processes, and therefore for further empirical work.

2.1.3. The Evolutionary Roots of Culture: Modern culture did not suddenly emerge from some pre-cultural Hominid ancestor (Plotkin, 1996). The psychological processes and abilities that underlie culture have evolved over millions of years, and can often be found in rudimentary form in animals. Cultural inheritance depends on the transmission of learned "knowledge" among individuals by one or more kinds of social learning (Cavalli-Sforza & Feldman, 1981; Durham, 1991). Hence a first step towards an understanding of the evolution of culture is to consider the evolution of social learning.

Over the past 15 years a variety of mathematical analyses have been conducted, exploring the adaptive advantages of social learning, relative to learning asocially, or expressing an unlearned pattern of behaviour that has been adapted over the course of genetic evolution (Boyd & Richerson, 1985; Rogers, 1988; Cavalli-Sforza & Feldman, 1983; Aoki & Feldman 1987; Bergman & Feldman, 1995; Laland et al., 1996b; Feldman, et al. 1996). Despite a plurality of methods, this body of theory has reached a surprising consensus as to when social learning is expected to be favoured. When environments change very slowly, adaptive knowledge should be gained at the level of population genetics because there are only modest demands for knowledge updating, which can easily be met by genetic systems responding to gradually changing selection pressures. In contrast, where environmental change is very rapid, or when there are sudden environmental shifts, tracking by individual learning should be favoured, as should horizontally (within-generation) transmitted information. In such environments, the genetic system will change too slowly to cope, while social learning from the parental generation is likely to be too error prone, as individuals would pick up outdated information. It is when individuals encounter intermediate rates of environmental change that social learning from parents should be favoured. Here, "intermediate" means when changes are not so fast that parents and offspring experience different environments, but not so slow that appropriate genetically transmitted behaviour could evolve instead.

The term "social learning", as currently applied to animals, describes a ragbag of heterogeneous processes, with a variety of functions, found in a broad array of vertebrate and invertebrate species. A more narrow use of the term would restrict it to those processes that might reasonably be regarded as homologous to processes operating in human social learning, and that mediate a general capacity to acquire information from others, regardless of the nature of the information, its function, or the sensory modality employed. Within the narrow category of social learning, humans probably transmit more information vertically from parent to offspring than all other species. Protocultural species typically depend primarily on horizontal transmissions based on social enhancement, rather than on imitation or teaching (Galef, 1988; Laland et al., 1993). A comparative perspective thus implies that the earliest forms of social transmission were probably horizontal. In contrast, humans appear to acquire large amounts of information from their parents, and from the parental generation (Hewlett & Cavalli-Sforza, 1986; Guglielmino et al., 1995), suggesting that the lineage leading to Homo sapiens has been selected for increasing reliance on vertical and oblique cultural transmission.

The theoretical analyses described above imply that a shift from transient horizontal traditions towards increased transgenerational cultural transmission reflects a greater constancy in the environment over time. Such a shift is difficult to reconcile with the traditional evolutionary perspective, since there is no evidence to suggest that environments have become more constant over the last few million years, but rather the opposite. Moreover, even if they had, other proto-cultural species would be expected to show more vertical transmission too. However, the increasing reliance of hominids on vertical transmission is consistent with our perspective, since here, a significant component of the selective environment is self-constructed by the species concerned, and this component could have favoured vertical transmission. We are suggesting that our ancestors constructed niches in which it "paid" them to transmit more information to their offspring. The more an organism controls and regulates its environment, and the environment of its offspring, the greater should be the advantage of transmitting cultural information from parent to offspring. For instance, by tracking or anticipating the movements of migrating or dispersing prey, populations of hominids may have increased the chances that a specific food source was available in their environments, that the same tools used for hunting would always be needed, and that the skin, bones and other materials from these animals would always be at hand to use in the manufacture of further tools. Such activities create the kind of stable social environment in which related technologies, such as food preparation or skin processing methods, would be advantageous from one generation to the next, and could be repeatedly socially transmitted from parent to offspring. It is possible that, once started, vertical cultural transmission may become an autocatalytic process: greater culturally generated environmental regulation leading to increasing homogeneity of environment as experienced by parent and offspring, favouring further vertical transmission. With new cultural traits responding to, or building on, earlier cultural traditions, niche construction sets the scene for an accumulatory culture. This might result in offspring learning higher-order "packages" of cultural traits from their parents, as appears to be the case in pre-industrial societies (Hewlett & Cavalli-Sforza, 1986; Guglielmino et al., 1995).

Clearly, the transition from animal proto-culture to human culture involved much more than a shift towards vertical transmission of information and the development of an accumulatory culture. Nonetheless, it is clear that the new evolutionary perspective portrayed in Figure 2c can generate novel hypotheses that may help to reconstruct some aspects of the evolution of human culture.

We suggest that any organism, O₁, should be prepared to cooperate in ways that benefit any other organism, O₂, provided the total niche-constructing outputs of O₂, or any of O₂’s decendents, modify resources in the environment of O₁, or any of O₁’s decendents, with resulting fitness benefits to O₁ that exceed the cost of O₁’s cooperation. This reasoning applies to relatives, and implies that cooperation should be more likely among kin that niche construct in mutually beneficial ways, and less likely among kin that niche construct in mutually detrimental ways, than a strict interpretation of Hamilton’s rule might suggest. The same logic also applies to non kin. In fact it is obvious that reciprocal altruism is a special case in which O₁ and O₂ are unrelated individuals of the same species that directly modify each other’s environment. Recent analyses here suggest that most cases currently described as reciprocal altruism, and many cases of kin selection, are actually forms of intra-specific mutualism that are the incidental outcomes of selfish behaviour (Mesterton-Gibbon & Dugatkin, 1992; Connor, 1995). In addition our statement justifies a variant of reciprocal altruism in which individuals do not trade altruistic acts with each other, but rather an individual aids a second individual if the latter acts in ways that benefit the first individual’s decendents.

Mutualisms and commensalisms also slot into this framework. Like altruism, mutualism also depends on the modification of the selective environments of recipient organisms by the niche-constructing activities of donor organisms. However, in the case of mutualisms, O₁ and O₂ are individuals in different species. The same is true of commensalisms, which for our purpose can be regarded as asymmetric versions of mutualism. In commensalisms O₁ cooperates with O₂ because O₂ benefits O₁ by ameliorating O₁’s environment, but O₂ is unnaffected by O₁, so O₂ does not cooperate with O₁.

Our general statement sketching the conditions under which organisms should cooperate, may, under restricted circumstances, generalise to cases in which O₁, and O₂ are groups of organisms, in one or more species. In many cases, the niche-constructed by-products of several organisms are exploited by a population. For example, shoals of fish, flocks of birds, and herds of animals enjoy reduced predation risks, relative to solitary animals, merely because their combined presence changes the selective environment of each individual (Hamilton, 1971). Coordinated fish driving by cormorants, or seal hunting by killer whales, provide examples of individuals coordinating their niche construction so that it results in more effective food acquisition for everyone (Connor, 1995).

Mutualistic interactions that result from the exploitation of incidental by-products are more evolutionarily robust than altruistic interactions as, because each individual is already acting selfishly, they are less likely to be broken down by selfish cheaters. Organisms may also invest in others so as to enhance the benefits which eventually return to themselves (Connor, 1995). A large amount of human cooperation can probably be explained in terms of the exploitation and investment in the by-products of others, in other words in terms of mutualisms resulting from human niche construction. For example, barter and exchange are mutualistic interactions in which individuals or organizations trade products for more desirable alternatives. Moreover, human individuals and institutions "invest", metaphorically, or even literally, in other individuals or institutions in order to enhance their own returns.

The original, or "naive", group selection hypothesis failed, primarily because the processes that maintain group differences and select between groups are typically weak compared with the processes that break down group differences and select within groups (Williams, 1966). If group-level adaptations are based on the cooperation of altruists, then any individual who refuses to cooperate can reap the benefits without paying the costs, and selfish strategies should be favoured by natural selection. It is not clear how Wilson and Sober can surmount this obstacle to explain the cooperation of large groups of individuals, since they have not proposed any feasible process that could have reduced individual differences within groups, yet promoted differences between groups, during hominid evolution. Here it may be worth reflecting on the fact that group-level cooperation typically depends on niche construction, because group-level adaptations are generally expressed outside of human bodies. Groups may remain cohesive because it pays each individual to invest in mutually beneficial niche construction.

Although we recognize some utility in the replicator-vehicle distinction, and in Wilson and Sober’s hierarchical approach, we regard it as a distortion. The entities that are selected, and between which there are fitness differences, are not well described as "vehicles", or even "interactors" (Hull, 1988), but rather are "organism-environment systems". Similarly, what is replicated, from one generation to the next, is a complex of information (both genetic and cultural) and some environmental (including developmental) resources (Gray, 1992; Oyama, 1985). Once these distortions are removed, it becomes easier to see how a form of group selection could help account for some human cooperation.

The hypothesis that comes closest to a solution comes from gene-culture coevolutionary theory, and places explicit emphasis on culturally inherited niche construction. In this hypothesis, proposed by Boyd and Richerson (1985), group selection works at the cultural level, with group-level cultural traits being selected. They place emphasis on a "when in Rome do as the Romans do" conformity, where individuals adopt the behaviour of the majority. The significance of this "conformist transmission" is that it minimises behavioural differences within groups, while maintaining differences between groups. Thus group selection of cultural variation "for the good of the group" is possible because if an altruistic cultural trait becomes frequent in a cultural group, the transmission process should subsequently discriminate against selfish individuals. Group selection of cultural rather than genetic variation, requires a "frame shift" of replicator, because it is not genes that are selected for, but rather groups of individuals expressing a particular culturally transmitted idea. Since cultural selection between groups may favour beliefs that benefit the group at the expense of the individual, Boyd and Richerson provide a new explanation for human cooperation.

Several properties of cultural inheritance, as opposed to genetic inheritance, make Boyd and Richerson’s idea attractive. One is that cultural inheritance, unlike genetic inheritance, may depend on more than two parents. It is therefore possible for individuals to be sensitive to the most frequent cultural traits in their society, and to conform to them. Second, group selection of cultural variants can be faster than group selection of genetic variants, because cultural death does not imply the physical death of all the people in a culture. A threatened or defeated people may switch to the traits of a new conquering culture, either voluntarily, or under duress. Thus, unlike Wilson and Sober’s group selection, here migration will not weaken the process. Third, symbolic group marker systems, such as totem animals, human languages, and flags, make it considerably easier for cultures to maintain their identities, and to resist imported cultural traits from immigrants, than it is for local gene pools, or demes, to maintain their identity by resisting gene flow (Boyd & Richerson, 1985). Fourth, cultural transmission of information about cheaters (e.g. gossip) reduces the efficacy of non-cooperative strategies (Dugatkin, 1992).

More negatively, conformist transmission may potentially be exploited by powerful individuals, groups, or institutions, that dominate the dissemination of information through societies, to promote their own interests. Powerful "cultural parents" may stand to gain from persuading other less powerful humans to conform, perhaps by recruiting extra assistance in modifying environments in ways in which they, rather than the helpers benefit. These processes may be amplified by tool use, for instance by the technology of the modern media, by weapons, by art, or by deceit. Religious, commercial and political propaganda, for example, may all be used to persuade, trick or coerce conformity from others against their own individual interests, yet in favour of the interests of a dominant class of cultural transmitters.

We can also reverse our earlier logic to suggest that any organism O₁, should act in a hostile manner, to the disadvantage of any other organism O₂, provided the total niche-constructing outputs of O₂, or of any of O₂’s descendents, modify resources in the environment of O₁, or any of O₁’s descendents, to the detriment of O₁, if the resulting reduction in the fitness costs to O₁ of O₂’s outputs exceeds the cost of O₁’s agonistic behaviour. It is easy to see how this reasoning might account for a great deal of aggressive behaviour, including a form of reciprocal hostility, in which individuals and their descendents trade antagonistic acts. In other words, we predict that organisms should actively harm other organisms by investing in niche construction that destroys other organism’s selective environments, provided the fitness benefits that accrue to the investing organisms from doing so, are greater than their fitness costs. Since this is a general idea, it should extend to the human cultural level, with the qualification that at this level other processes may be operating.

2.3.1. Units of Selection in Human Evolution: A prima facie problem with our multiple-processes-in-evolution approach is that it raises questions about the currency of human evolution. Should cultural traits be measured in terms of reproductive success only, as the sociobiologists advocate, or should they be measured by a cultural transmission rate parameter as well as, or instead of genes? This problem was one of the earliest hurdles faced by contemporary gene-culture theory. Initially genetic and cultural processes were treated as independent, with separate fitness scores and transmission coefficients allocated for natural selection and cultural transmission. However, it rapidly became clear that genes and cultural traits can interact in the same way that two genetic loci can interact, to generate associations of genotype and cultural phenotype in non-random frequencies (Feldman & Cavalli-Sforza, 1984). This means that treating genetic and cultural processes as independent is a distortion. A straightforward, pragmatic solution is to allocate fitnesses and transmission rate parameters directly to combinations of genotypes and cultural traits, a package known as a 'phenogenotype'. For instance, in Feldman and Cavalli-Sforza’s (1989) theoretical exploration of the coevolution of lactose tolerance and milk usage, one phenogenotype was a milk using individual expressing two copies of the allele conferring lactose tolerance, while another was a non-milk using individual which no such alleles. For human sociobiologists, the most appropriate way to think about evolution is from the perspective of the gene: those characteristics that have been favoured by selection are the expression of the "selfish genes"(Dawkins, 1989) that were best able to increase their representation in the next generation. For gene-culture coevolutionary theory, the logic is the same, but the replicator is different: Instead of the selfish gene there is the "phenogenotype". Those human characteristics that have been favoured in the face of both natural selection and cultural transmission are the expression of the phenogenotypes that were best able to increase their representation in the next generation, by whatever process. As an intuitive shorthand, a phenogenotype can be thought of as a human with a package of genes and experience. In this sense, the phenogenotype approach re-establishes the organism (or rather, classes of organism) as the central unit of human evolution, not as vehicle but as replicator. In fact, what is really replicated is a bio-cultural complex, with a composite array of information (acquired through multiple processes, and stored at different levels) and inherited resources. However, we recognize the need for simple conceptual and formal models that have the utility to explore and shed light on the dynamics of such systems, and we regard the phenogenotype approach as the best method currently on the market⁵.

2.3.2. Multiple-Process Adaptation: Controversy has surrounded the sociobiological postulate that human beings typically behave in ways that increase their inclusive genetic fitness (Sahlins, 1976; Montagu, 1980). It is trite to point out that the processes that underlie culture are adaptations, and that socially learned information, and cultural inheritance, may increase reproductive success. Mathematical models that have explored the evolution of social learning reveal that it is a truism of the modelling exercise that the capacity for social learning cannot be favoured unless it generally increases some measure of fitness. Obviously, the same is true of knowledge-gaining ontogenetic processes. So what characteristics of human culture could allow humans to behave in a maladaptive way, or to transmit maladaptive information?

Our perspective suggests that, in each generation, populations of organisms persistently construct or reconstruct significant components of their environments. This means that, as they evolve, organisms may, in effect, drag part of their own environments along with them, thereby transforming their own "adaptive landscapes". If ontogenetic processes, culture, and counteractive niche construction in general, have consistently damped out the need for a genetic response to changes in the population’s environment, hominid populations may have become increasingly divorced from their ecological environments. At the same time, our hominid ancestors may increasingly have responded to novel selection pressures initially generated by inceptive niche construction, and subsequently dominated by cultural traditions. In this case, the common conception that modern human populations are adapted to an ancestral Pleistocene environment (Barkow et al., 1992) can only be partly correct. In particular, components of the social environment, for example, traits related to family, kinship and social stratification, may have been increasingly vertically transmitted by culture to the extent that contemporary human populations may have become largely divorced from local ecological pressures. Support for this argument comes from Guglielmino’s et al.’s (1995) study of variation in cultural traits among 277 contemporary African societies, in which most traits examined correlated with cultural (linguistic) history, rather than ecology.

In the short term organisms typically niche construct in ways that enhance their immediate fitness, but in the long term organisms can also "niche destruct" relative to their own genes. For example, they can build up polluting detritus, or strip their environments of non-renewable, or too slowly renewing, resources, until they have made their own environments hostile to themselves and to their offspring (Diamond 1993). Among plants, this process typically leads to auto-ecological succession, while animals typically respond by dispersing to other environments. Failure to respond to the feedback from negative niche construction is a possible recipe for extinction.

Could humans drive themselves to extinction? There are two reasons for supposing that this is a possibility. First, culture greatly enhances the human capacity for niche construction. For example, science based technology is currently making an enormous impact on the human environment. It has made many new resources available via both agriculture and industry, it has influenced human population size and structure via hygiene, medicine and birth control, it has drastically changed human warfare, it is drastically reducing biodiversity, and it may already have resulted in the degradation of large areas of our global environment. These are all potential sources of modified natural selection pressures. Second, human cultural processes can work much faster than human genetic processes, generating new adaptive problems at a faster rate than the human genetic processes can respond to. In these circumstances, human culture might drive either local, or general, self-induced extinctions.

In many respects, this is a reoccurrence of an old evolutionary problem. Many relatively long-lived species encounter rates and types of environmental change, whether self-induced or independent, that exceed the capacity of their genes to handle, and they frequently go extinct. Clearly, one way in which human beings could adapt to culturally induced environmental changes is through quicker acting responses at some non-genetic level, especially through further cultural change.

Unfortunately there are well known snags with this kind of solution. First, the population may not recognise the source of the novel, culturally induced selection pressure. This was the case with the Fore of Papua New Guinea who maintained a cannibalistic tradition despite the fact that it perpetuated a deadly disease (Durham 1991). Second, the required corrective technology may not always be available, or may be too costly to introduce. For example, in theory the Kwa could have responded to the increased selection by malaria by the cultural control of this disease, but in fact they lacked the technology to do so. Third, the feedback from cultural niche construction may be indirect, which may make it difficult to recognise any longer term negative consequences of the niche construction. Rogers (1995) documents how the adoption of wet rice cultivation in Madagascar had a range of diffuse indirect effects only manifest several generations later, including changes in tribal government, patterns of warfare and the role of the father. Fourth, responding to cultural change with further cultural change always risks introducing a "runaway" situation, in which each new solution generates the next problem, at an ever accelerating rate. The phenomenon of antibiotic resistance is a recent example (Ewald, 1994).

In the preceding sections we have begun to develop a new type of evolutionary framework for the human sciences by emphasing niche construction, and ontogenetic and cultural processes. We have also illustrated the conceptual model with a number of ideas, related to sample topics. Our hope is that these suggestions will encourage others to use the evolutionary framework we are proposing in this paper, either to take some of the ideas we have already discussed further, to the point where they can be empirically tested, or to generate other hypotheses of their own.

Acknowledgements

We are grateful to Robert Hinde, Adam Kuper, Richard Lewontin, Henry Plotkin, Peter Richerson, and Kate Robson-Brown for helpful comments on an earlier draft of this manuscript. Kevin Laland was supported by a Royal Society University Research Fellowship, and Marcus Feldman by NIH grant GM28016.

Aiello, L.C. & Wheeler, P. (1995) The expensive-tissue hypothesis. Current Anthropology 36: 199-221.

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

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

Aoki, K. & Feldman, M.W. (1987) Toward a theory for the evolution of cultural communication: Coevolution of signal transmission and reception Proceedings of the National Academy Science, USA 84: 7164-8.

Aoki, K. & Feldman, M.W. (1997) A gene-culture coevolutionary model for brother-sister mating. Proceedings of the National Academy Science, USA 94: 13046-13050.

Axelrod, R. (1984) The Evolution of Cooperation. Basic Books.

Barkow, J., Cosmides, L., & Tooby, J. (1992) The Adapted Mind: Evolutionary Psychology and the Generation of Culture Oxford University Press.

Bateson, P.P.G. (1988) The active role of behaviour in evolution. In: Evolutionary Processes and Metaphors ed.M.W. Ho & S. Fox Wiley.

Bergman, A. & Feldman, M.W. (1995) On the evolution of learning: Representation of a stochastic environment. Theoretical Population Biology 48(3): 251-276.

Bodmer, W.F. & Cavalli-Sforza, L.L. (1976) Genetics, Evolution and Man Freeman.

Bolles, R.C. (1970) Species-specific defense reactions and avoidance learning. Psychological Review 77: 32-48.

Borgerhoff Mulder M (1991) Human behavioural ecology. In: Behavioural Ecology: An Evolutionary Approach. ed. J.R. Krebs & N.B. Davies. Blackwell.

Boyd, R. & Richerson, P.J. (1985) Culture and the Evolutionary Process. University of Chicago Press.

Boyd, R. & Richerson, P.J. (1988) The evolution of reciprocity in sizable groups. Journal of Theoretical Biology 132: 337-356.

Boyd, R. & Richerson, P.J. (1989) The evolution of indirect reciprocity Social Networks. 11: 213-236.

Boyd, R. & Richerson, P.J. (1992) Punishment allows the evolution of cooperation (or anything else) in sizable groups. Ethology & Sociobiology 13: 171-195.

Byrne,R. & Whiten, A. (1988) Machiavellian Intelligence: Social Expertise and the evolution of intellect in monkeys, apes and humans. Clarendon Press.

Cavalli-Sforza, L.L. & Feldman, M.W. (1973) Models for cultural inheritance I. Group mean and within-group variation. Theoretical Population Biology 4(1): 42-55.

Cavalli-Sforza, L.L. & Feldman, M.W. (1981) Cultural Transmission and Evolution: A Quantitative Approach. Princeton University Press.

Cavalli-Sforza, L.L. & Feldman, M.W. (1983) Cultural versus genetic adaptation Proceedings of the National Academy Science, USA 79: 1331-5.

Connor, R.C. (1995) The benefits of mutualism: A conceptual framework. Biological Reviews70: 427- 457.

Cronk, L. (1995) Review of Laland, Kumm, and Feldman’s "Gene-Culture Coevolutionary Theory". Current Anthropology 36: p147-8.

Daly, M. & Wilson, M. (1983) Sex, Evolution, and Behavior. 2nd Edition. Wadsworth.

Darwin, C. (1881) The Formation of Vegetable Mold through the Action of Worms, with Observations on their Habits. Murray.

Dawkins, R. (1982) The Extended Phenotype. Freeman: Oxford.

Dawkins, R. (1989) The Selfish Gene, 2nd Edition. Oxford University Press.

Diamond, J. (1993) The Third Chimpanzee. The Evolution and Future of the Human Animal. Harper Perennial.

Dugatkin, L. (1992) The evolution of the con artist. Ethology and Sociobiology 13: 3-18.

Dunbar, R.I.M. (1993) Coevolution of neocortical size, group size and language in humans. Behavioral and Brain Sciences 16: 681-735.

Durham, W.H. (1991) Coevolution: Genes, Culture and Human Diversity. Stanford University Press.

Edmunds, M. (1974) Defense in Animals. Longman.

Eldredge, N. & Gould, S.J. (1972) Punctuated Equilibria: an alternative to phyletic gradualism. In Models in Palaeobiology (ed. TJM Schopf), pp82-115. Freeman & Cooper: San Francisco.

Ellis, S. & Mellor, A. (1995) Soils and Environment. Routledge: London.

Ewald, P.W. (1994) Evolution of Infectious Diseases. Oxford University Press.

Feldman, M.W., Aoki, K. & Kumm, J. (1996) Individual versus social learning: Evolutionary analysis in a fluctuating environment. Anthropological Science 104(3) 209-232.

Feldman, M.W. & Cavalli-Sforza, L.L. (1976) Cultural and biological evolutionary processes, selection for a trait under complex transmission. Theoretical Population Biology 9(2): 238-259.

Feldman, M.W. & Cavalli-Sforza, L.L. (1984) Cultural and biological evolutionary processes: Gene-culture disequilibrium. Proceedings of the National Academy Science, USA 81: 1604-1607.

Feldman, M.W. & Cavalli-Sforza, L.L. (1989). On the theory of evolution under genetic and cultural transmission with application to the lactose absorption problem, In: Mathematical Evolutionary Theory. ed. M.W. Feldman. Princeton University Press.

Feldman, M.W. & Laland, K.N. (1996) Gene-Culture Coevolutionary Theory. Trends in Ecology & Evolution 11: 453-7.

Foley, R. (1995) The adaptive legacy of human evolution: a search for the environment of evolutionary adaptedness. Evolutionary Anthropology 4 (2) 194-203.

Frisch, K. von (1975) Animal Architecture. Hutchinson: London.

Galef, B.G. Jr. (1988) Imitation in animals: History, definition, and interpretation of data from the psychological laboratory. In: Social Learning: Psychological and Biological Perspectives. ed. T.R. Zentall TR & B.G. Galef Jr. (eds.) Erlbaum.

Grant, P.R. (1986) Ecology and Evolution of Darwin’s finches. Princeton Press.

Gray, R.D. (1992) Death of the gene: Developmental systems fight back.In Trees of Life: Essays in the philosophy of biology. (ed.) P. Griffiths Kluver.

Guglielmino, C.R., Viganotti, C., Hewlett, B., & Cavall-Sforza, L.L. (1995) Cultural variation in Africa: Role of mechanisms of transmission and adaptation. Proc. Natl. Acad. Sci. USA 92: 7585-7589.

Hamilton, W.D. (1964) The gentical evolution of social behaviour I. Journal of Theoretical Biology 7: 1-16.

Hamilton, W.D. (1971) Geometry for the selfish herd. Journal of Theoretical Biology 31: 295-311.

Hansell, M.H. (1984) Animal Architecture and Building Behaviour. Longman: NY.

Hewlett, B.S. & Cavalli-Sforza, L.L. (1986) Cultural transmission among Aka pygmies. Am. Anthrop. 88: 922-934.

Hinde, R.A. (1987) Individuals, Relationships and Culture. Cambridge University Press.

Hinde, R.A.& Stevenson-Hinde, J. (1973) Constraints on Learning. Academic Press.

Holland, J.H. (1992) Adaptation in Natural and Artificial Systems. MIT Press.

Holland, J.H., Holyoak, K.J., Nisbett, R.E. & Thagard, P.R. (1986) Induction. Processes of Inference Learning and Discovery. MIT Press.

Holloway, R.L. (1981) Culture, symbols and human brain evolution: A synthesis. Dialectical Anthropologist 5: 287-303.

Holmgren, M., M. Scheffer, & M.A. Huston. (1997) The interplay of facilitation and competition in plant communities. Ecology. 78. 1966 -1975.

Jaenike, J. (1982) Environmental modification of oviposition behaviour in Drosophila. Am. Nat. 119: 784-802.

Jones, C.G., J.H. Lawton & M. Shachak. (1997) Positive and negative effects of organisms as physical ecosystem engineers. Ecology. 78. 1946 -1957.

Kirkpatrick, M. & Lande, R. (1989) The evolution of maternal characters. Evolution 43(3): 485-503.

Krebs, J.R. & Davies, N.B. (1993) An Introduction to Behavioural Ecology. 3rd Edition. Blackwell Press.

Kumm, J., Laland, K.N. & Feldman, M.W. (1994). Gene-culture coevolution and sex ratios: The effects of infanticide, sex-selective abortion, sex selection, and sex-biased parental investment on the evolution of sex ratios, Theoretical Population Biology 46(3): 249-278.

Kuper, A. (1988) The invention of primitive society. Routledge. London.

Laland, K.N. (1994) Sexual selection with a culturally transmitted mating preference, Theoretical Population Biology. 45(1): 1-15.

Laland, K.N., Odling-Smee, F.J. & Feldman, M.W. (1996a). On the Evolutionary Consequences of Niche Construction. Journal of Evolutionary Biology. 9: 293-316.

Laland, K.N., Richerson, P.J. & Boyd, R. (1993) Animal social learning: towards a new theoretical approach. Perspectives in Ethology 10. Plenum Press.

Laland, K.N., Richerson, P.J. & Boyd, R. (1996b). Developing a theory of animal social learning, In: Social Learning in Animals: the Roots of Culture. ed. C.M. Heyes & B.G. Galef Jr. Academic Press.

Lee, K.E. 1985. Earthworms: their ecology and relation with soil and land use. Academic Press.

Lewin, R. (1998) Principles of human evolution: a core textbook. Blackwell Science. Oxford.

Lewontin, R.C. 1983. Gene, organism, and environment. In: Evolution from Molecules to Men ed. D.S. Bendall. Cambridge University Press.

Lovejoy, C.O. (1981) The origin of man. Science 211(4480): 341-350.

Lumsden, C.J. & Wilson, E.O. (1981) Genes, Mind and Culture Harvard University Press.

Maynard-Smith, J. (1989) Evolutionary Genetics. Oxford UP.

Mayr, E. (1982) The growth of biological thought: Diversity evolution and inheritance. Belnap, Harvard University Press: Cambridge, Mass.

Mesterton-Gibbons, M. & Dugatkin, L.A. (1992) Cooperation among unrelated individuals: evolutionary factors. Quarterly Review of Biology 67: 267-281.

Montagu, A. (1980) Sociobiology Examined. Oxford University Press.

Mount, A.B. (1964) The interdependence of the eucalypts and forest fires in southern Australia. Austr. For. 28; 166-172.

Nowak, R.M. (1991) Walker’s Mammals of the World, 5th Edition. John Hopkins UP.

Odling-Smee, F.J. (1988) Niche constructing phenotypes. In: The Role of Behavior in Evolution. ed. H.C. Plotkin MIT Press.

Odling-Smee, F.J. (1994) Niche construction, evolution and culture In: Companion Encyclopedia of Anthropology. ed. T. Ingold. Routledge.

Odling-Smee, F.J., Laland, K.N. & Feldman, M.W. (1996) Niche Construction. The American Naturalist. 147(4): 641-648.

Oyama , S. (1985) The Ontogeny of Information. Cambridge University Press.

Pinker, S. (1994) The Language Instinct: The New Science of Language and Mind. Penguin.

Plotkin, H.C. (1988) Learning and evolution. In: The Role of Behavior in Evolution. ed. H.C. Plotkin. MIT Press.

Plotkin, H.C. (1996) Non-genetic transmission of information: Candidate cognitive processes and the evolution of culture. Behavioral Processes 35: 207-213.

Plotkin, H.C. & Odling-Smee, F.J. (1981) A multiple-level model of evolution and its implications for sociobiology. Behavioral and Brain Sciences 4: 225-268.

Preston-Mafham, K. & Preston-Mafham, R. (1996) The Natural History of Spiders. Crowood Press: Ramsbury, Wilts.

Richerson, P.J. & Boyd, R. (In Press). The evolution of human ultra-sociality. In: Indoctrinability, Warfare and Ideology: Evolutionary Perspectives ed. I. Eibl-Eibesfeldt, & F. Salter. Berghahn Press

Robertson, D.S. (1991) Feedback theory and Darwinian evolution. Journal of Theoretical Biology 152, 469-84.

Rogers, A. (1988) Does biology constrain culture? American Anthropologist 90(4): 819-831.

Rogers, E.M. (1995) Diffusion of Innovations, 4th Edition. Free Press: New York.

Sahlins, M.D. (1976) The Use and Abuse of Biology: An Anthropological Critique of Sociobiology. University of Michigan Press.

Seligman, M.E.P. (1970) On the generality of the laws of learning. Psychological Review 77: 406-418.

Szathmary, E. & Maynard-Smith, J. (1995) The major evolutionary transitions. Nature 374: 227-231.

Trivers, R.L. (1985) Social Evolution. Benjamin/Cummings.

Tylor, E.B. (1871) Primitive Culture. London: Murray

Vrba, E.S. (1992) Mammals as a key to evolutionary theory. J. Mammol, 73: 1-28.

West, M.J., King, A.P. & Arberg A.A. (1988) The inheritance of niches. The role of ecological legacies in ontogeny. Handbook of Behavioral Neurobiology, vol. 9. Ed. E. Blass. Plenum: NY.

West-Eberhard, M.J. (1987) Flexible strategy and social evolution. In: Animal Societies: Theories and Facts. ed. Y. Brown & J. Kikkawa. Japanese Social Science Press.

Whelan, R.J. (1995) The Ecology of Fire. Cambridge UP.

Williams, G.C. (1966) Adaptation and Natural Selection. Princeton University Press.

Wilson AC (1985) The molecular basis of evolution. Scientific American 253(4): 148-157.

Wilson, E.O. (1975) Sociobiology: The New Synthesis. Harvard University Press.

Wilson, E.O. (1978) On Human Nature Harvard University Press.

Wilson, D.S. & E. Sober, (1994) Reintroducing group selection to the human behavioral sciences. Behavioral and Brain Sciences 17: 585-654.

Wolf, J.B., E.D. Brodie, J.M. Cheverud, A.J. Moore & M.J.Wade. (1998) Evolutionary consequences of indirect genetic effects. Trends in Ecology & Evolution. 13.64-69.

Wynne-Edwards, V.C. (1962) Animal Dispersion in Relation to Social Behaviour. Oliver & Boyd.