R1. What counts as evolutionary?

We are not surprised to find this question posed by a number of commentators, as we encounter it almost daily in our interactions with colleagues across disciplines. We think that the best articulation of the evolutionary paradigm is by Niko Tinbergen, who shared the Nobel Prize in medicine with Konrad Lorenz and Carl von Frisch for helping to found the science of ethology, the study of animal behavior. Tinbergen wrote his classic Reference Tinbergen1963 paper, “The Methods and Aims of Ethology,” to explain why the study of behavior should be regarded as a branch of biology. At the time, it was not obvious that it was possible to study a behavioral trait, such as aggression, in the same way that an anatomical trait, such as a deer's antlers, could be studied. Tinbergen pointed out that four questions need to be asked for any product of evolution: its function, its mechanisms, its development, and its history. These questions can be as profitable for behavioral traits as for physical traits.

We think that Tinbergen's four questions work as well for a science of intentional change as for the science of ethology, or indeed all the behavioral and life sciences. Every trait with a functional basis must be understood in functional terms. It will have a physical mechanism, which is typically the result of a developmental process, and it will have a history. This is true for culturally derived traits no less than genetically derived traits. When all four questions are linked to issues of variation and selective retention, they define an evolutionary approach.

There is no consensus among evolutionary biologists that the term “evolutionary” should be equated with Tinbergen's four questions. Even Tinbergen used the word evolution primarily to describe the history question. It is common for other evolutionists to associate evolution with function but not with mechanism or development. There is unlikely to be a consensus anytime soon, but at least we can be clear with our own usage: We associate the evolutionary paradigm with Tinbergen's fully rounded four-question approach, which can be applied to any product of evolution. We stated this briefly in our target article (sect. 2.2) and perhaps should have featured it more strongly. It does occupy center stage in the lead article of a special issue of the Journal of Economic Behavior & Organization titled “Evolution as a General Theoretical Framework for Economics and Public Policy” (Wilson & Gowdy Reference Wilson and Gowdy2013), which provides a useful complement to our BBS article.

A common complaint about broad definitions of evolution is that they lose meaning. By including everything, they explain nothing. But consider the standard definition of genetic evolution as any change in gene frequency, whether by selection, drift, linkage disequilibrium, or any other force. It is important to have a general accounting system that includes all forms of genetic change. The accounting system is meaningful when it includes meaningful categories. Thus, just noting that something evolves has little meaning, but documenting that it evolves by selection, by drift, or by linkage disequilibrium is meaningful.

The same considerations apply to a science of intentional change. It is important to have a general accounting system that includes all products of variation-and-selection processes. The accounting system is meaningful when it includes meaningful categories. There are more categories for a science of intentional change than for genetic evolution, to include Darwin machines with nongenetic inheritance mechanisms. We will return to this point below when we discuss various forms of forward-looking change discussed by some of the commentators.

Another common critique of an evolutionary perspective is that it does not add value to other perspectives, but merely reinvents what has already been discovered. This will certainly be true some of the time; other perspectives do not get everything wrong! Nevertheless, there are strong reasons for concluding that a fully rounded four-questions approach can add value to any sizeable human-related subject, as discussed in more detail by Wilson and Gowdy (Reference Wilson and Gowdy2013).

Against this background, we can consider some of the specific points raised by the commentators. Grinde thinks that the differences between genetic evolution and other evolutionary processes are greater than their similarities. Perhaps the term “evolution” should be reserved for genetic evolution. At the very least, policy efforts should focus on cultural, not genetic, evolution. We share Grinde's distaste for the “unsavory” eugenics policies inspired by evolutionary theory in the past, but we do not think that they should unduly influence our current conceptual framework or current policies that meet appropriate ethical standards (see next section). Evolutionary theory has been so gene-centric during the last century that a central message for the future must be “there is more to evolution than genetic evolution” (Jablonka & Lamb Reference Jablonka and Lamb2006). Furthermore, the thrust of concepts such as gene-culture coevolution, developmental systems, and niche construction is that genes play a role in larger systemic processes, but should not be conceptualized as privileged units within the larger system. We, therefore, prefer to maintain our broad use of the term “evolution” and think that Grinde's concerns can be addressed by defining meaningful categories of evolution inside a large systems-based approach.

Andersson, Törnberg, & Törnberg (Andersson et al.) stress,

Similarly, Grotuss seems to equate our treatment of the Darwin machine concept with narrow-school evolutionary psychology, when we think that our treatment is consistent with his own position. We hope that our reliance on Tinbergen's fully rounded four-questions approach makes this clear. That said, we acknowledge that our target article spent less time on questions of causal mechanisms, so we appreciate the emphasis and additional information provided by these and other commentators, as is addressed in more detail below.

Commenting on the successful change efforts that we review in section 3 of our target article, Burghardt, Stuart, & Shorey (Burghardt et al.) state, “it is unclear what is evolutionary about them. More specifically, it is unclear how evolutionary theory either guided their development or anticipated their success, other than their being a product of variation and selection.” This critique conflates two questions: (1) When does a process count as evolutionary? and (2) When does an explicitly evolutionary perspective add value to understanding an evolutionary process that has not yet been explicitly approached from an evolutionary perspective? We can imagine a similar critique being leveled against Darwin's effort to interpret past knowledge from an evolutionary perspective!

Wilson and Gowdy (Reference Wilson and Gowdy2013) provide a general argument for why explicit evolutionary theorizing is likely to add value to the study of any sizeable human-related topic. On the issue of prospective impact, it is worth noting the role that explicit evolutionary thinking has played in the history of the applied behavioral sciences, such as B. F. Skinner's concept of selection by consequences, which have been central historically to some of the very programs that we are examining.

Burghardt et al. provide an interesting and illustrative challenge as an expression of their concern that integration must go beyond post hoc stories to “the prospective analysis we need to move forward”: Would it have been possible to predict beforehand the failure of a program such as the Drug Abuse Resistance Education (D.A.R.E.) program over comprehensive cognitive-behavioral exposure treatments for anxiety, such as those developed by Barlow et al. (Reference Barlow, Craske, Cerny and Klosko1989)? We think that indeed it was possible, and in two ways.

Evolutionarily sensible therapy programs have found ways to foster healthy variation, selected by criteria of importance to those being worked with, in a way that is sensitive to contextual features of their life, occurring at the right level of selection (individuals; couples or groups), and in which retention of gains is fostered. The scientific development programs of such therapies often reflect the same process: They focus on data relevant to a variety of stakeholders and, as data come in, theories, methods, and techniques are altered in a continuous development cycle sensitive to those outcomes. In both of these areas (focus and development) it is easy to predict on evolutionary grounds that programs such as Barlow et al.'s cognitive behavior therapy (CBT) program will succeed, whereas programs such as D.A.R.E. may not.

The Barlow et al. approach originally included a number of traditional CBT techniques such as cognitive modification, relaxation, and breathing training, among others, but went beyond traditional stimulus exposure to include exposure to emotions and sensations themselves. Research showed that these innovative methods increased tolerance of anxiety and fostered greater flexibility and greater ability to learn in the presence of anxiety cues (Craske et al. Reference Craske, Kircanski, Zelikowsky, Mystkowski, Chowdhury and Baker2008). Indeed, measures of psychological flexibility drawn directly from those used in Acceptance and Commitment Therapy (ACT) have been shown to mediate the outcomes of such programs (e.g., Arch et al. Reference Arch, Wolitzky-Taylor, Eifert and Craske2012b). The program itself has been continuously revised since its creation, dropping and adding elements. These data-driven changes increasingly aligned the protocol more with evolutionarily obvious ideas, as would make sense if the analysis in our target article is correct.

The latest and highly effective iteration to grow out of this development work, for example, now consists of only four modules: “increasing emotional awareness, facilitating flexibility in appraisals, identifying and preventing behavioral and emotional avoidance, and situational and interoceptive exposure to emotion cues” (Ellard et al. Reference Ellard, Fairholme, Boisseau, Farchione and Barlow2010, p. 88). All of these remaining modules are focused on increasing the contextual sensitivity of action and on augmenting emotional, cognitive, and behavioral flexibility. Although these changes were not driven by evolution science, they are easy to understand from that perspective, and any evolutionist examining the data and changes would appreciate the consilience between the focus of these remaining elements and evolution science findings.

In contrast, D.A.R.E. sought to use social persuasion methods to ingrain a rigid “zero tolerance” approach to drugs. Police became teachers to increase the authoritative impact of rule-based training; specific rules were modeled and repeated, such as “Recognize, Resist, and Report,” as if greater behavioral rigidity could solve the problem of drug use. Dissemination was aggressively based on fundraising and publicity, not on data collection and continuous program modification. When research suggested that increased awareness actually led to greater curiosity and increased drug experimentation (Rosenbaum & Hanson Reference Rosenbaum and Hanson1998), the response was to criticize the research findings and even to attempt to keep journals from reporting them, even as negative evidence kept pouring in for decades (Lilienfeld Reference Lilienfeld2007). Therefore, at both content and process levels, we think that evolutionary concepts would readily meet the challenge Burghardt et al. described.

The best test of the value of evolutionary thinking, of course, is empirical. Although we are unaware of any meta-scientific studies of this kind, some programs described in our target article were indeed explicitly driven by evolutionary thinking, whereas other programs developed in the same time frame or through the same funding mechanisms were not. This allows a naturalistic test of Burghardt et al.'s concerns. For example, the design of the PeaceBuilders violence prevention strategy was explicitly grounded in the evolutionary paradox that human groups are the greatest vertebrate predator of humans, and other human groups are the best source of safety from predatory humans (Embry Reference Embry1991; Embry & Flannery Reference Embry, Flannery, Flannery and Huff1999; Embry et al. Reference Embry, Flannery, Vazsonyi, Powell and Atha1996). This program showed larger effect sizes (Flannery et al. Reference Flannery, Vazsonyi, Liau, Guo, Powell, Atha, Vesterdal and Embry2003; Vazsonyi, et al. Reference Vazsonyi, Belliston and Flannery2004) than did more than a dozen other strategies not informed by evolutionary theory that were funded at the same time, most of which had iatrogenic or weak results (Grossman et al. Reference Grossman, Neckerman, Koepsell, Liu, Asher, Beland, Frey and Rivara1997; Guerra et al. Reference Guerra, Eron, Huesmann, Tolan, Van Acker and Björkqvist1997; Jaycox et al. Reference Jaycox, McCaffrey, Eiseman, Aronoff, Shelley, Collins and Marshall2006; The Multisite Violence Prevention Project 2009), and it was the only strategy randomized trials showed reduced medically coded violent injuries (Krug et al. Reference Krug, Brener, Dahlberg, Ryan and Powell1997).

Smaldino & Waring express concern that “an imperative to unite all social science under an evolutionary framework risks turning off researchers who have their own theoretical perspectives that can be informed by evolutionary theory without being exclusively defined by it.” We agree that very important framing issues must be kept in mind. The problem is even worse than Smaldino & Waring describe, because many people are already turned off by past associations with evolution within their disciplines and so are not starting from a neutral position. Nevertheless, it is important not to let past associations interfere with future goals – a first rule of therapy – and our goal is for evolutionary theory to have the same kind of generality in the human-related disciplines as it does in biology. Dobzhansky's (Reference Dobzhansky1973) dictum, “Nothing in biology makes sense except in the light of evolution,” needs to be expanded to include our own species, including our cultural and behavioral diversity; and this expansion needs the same positive connotation that the dictum already has for biologists.

Getting to there from here for any particular person or discipline might be difficult. Personal and cultural evolution is a path-dependent process across a rugged adaptive landscape. But we have found that with the appropriate framing, the transition can also be easy. The fear associated with evolution as a “totalizing” theory is that one's current perspective and expertise will be rendered obsolete. In contrast, it can be affirming to learn that one's perspective (e.g., social constructivism) plays a central role in the human evolutionary story, that it can be generalized beyond one's current disciplinary boundaries, and that one's empirical expertise (e.g., the history of witchcraft in Europe) can be just as important from an evolutionary perspective as from other perspectives, because all theories must draw on empirical information. It can be even more affirming to realize that one's area (e.g., human symbolic thought) can fill an important gap in core evolutionary theory, as we tried to stress in our target article. Integration as a two-way street is alluring, not threatening.

R2. Ethical considerations

Ethics is a foremost consideration for any change effort. The question we need to consider is whether the ethical considerations are different for a change effort centered on evolution. We think that for the most part, the answer is no.

Consider one reason change efforts often fail: because people have incomplete information, and what they think will work falls victim to unforeseen consequences. This hazard exists for anyone attempting to accomplish anything on the basis of any rationale. We emphatically do not think that evolutionists already have answers that can be implemented in a top-down fashion. In fact, there are grounds for thinking that such omniscience might never be possible for highly complex systems, as pointed out by some of the commentators (Andersson et al.; Bodor & Fokas; Grotuss; Ho, Torres-Garcia, & Swain (Ho et al.); Kostrubiec & Kelso; O'Brien; Read; and Smaldino & Waring). This humbling fact calls for a different approach to policy formulation and implementation that is more experimental; in other words, one that involves highly orchestrated variation-and-selection practices. We will say more about this in the section on complexity.

Another reason why change efforts fail is that they benefit some people at the expense of others and the group as a whole (the fundamental evolutionary paradox of humans and common-pool resources), even when the decision makers think that they have the welfare of everyone in mind. This is a hazard for change efforts from any perspective and speaks to the need for checks and balances, including inclusive decision-making processes, which might be consensus decision making in small groups or representative forms of decision making in large groups. An evolutionary perspective reinforces this message, if reinforcement is needed, because it highlights the potential conflict between levels of functional organization. What counts as good for any given social entity is likely to undermine functional organization at higher levels. This is in contrast to the economic metaphor of the invisible hand, which makes it seem as if lower-level self-interest robustly contributes to higher-level functional organization. The evolutionary perspective makes it crystal clear that adaptation at any given level requires a process of selection at that level, which includes the deliberative selection of policies in addition to other variation-and-selection processes.

The standard formulation of the naturalistic fallacy is that “is” does not imply “ought.” Just because something evolves does not make it good or right. We hope it is obvious from our target article that we are mindful of the naturalistic fallacy: One of our major themes is that left unmanaged, evolution often takes us where we do not want to go. For the most part, we think that the evolutionary perspective helps to clarify the “is” without changing the “ought.” That is why most of the practical applications that we review in the target article are ethically unproblematic. Who does not want to empower neighborhoods, improve learning outcomes, and so on?

At a deeper level, however, the concept of “ought” needs to be understood from an evolutionary perspective. One of us (DSW) interviewed the distinguished moral philosopher Simon Blackburn (available at http://www.thisviewoflife.com/index.php/magazine/media/evolution-and-morality-i-simon-blackburn). The interview begins with a discussion of morality in general terms before exploring the relevance of evolution. Here is a transcript of Blackburn answering the question “What is morality?”

That definition of morality, which was stated without reference to evolution, cries out for an evolutionary explanation as a set of mechanisms that enables groups to function as cooperative units. That is why moral philosophers such as Blackburn are taking evolution seriously to clarify the nature of morality. In this fashion, the evolutionary perspective is relevant to our understanding of “ought” in addition to our understanding of “is” without committing the naturalistic fallacy in any naïve sense.

We do not mean to underestimate the complexity and divisiveness of many ethical issues, but this is a problem for any and all perspectives regarding “what is moral.” Moreover, an evolutionary perspective can shed light on the cultural and situational diversity of moral values that contribute to differences that are hard to reconcile (e.g., Haidt Reference Haidt2012). Tinbergen's four questions are useful in this regard. If the function of a moral system is to orchestrate the functional organization of groups, and if moral systems evolve by gene-culture coevolution, then different mechanisms are likely to evolve in different cultures or for different situations within the same culture. These historically based differences are likely to become problematic with the admixture of cultures during modern history. Detailed knowledge of function, mechanism, and history (along with development) in conjunction with one another will be more useful for resolving difficult ethical issues than will the absence of such knowledge.

Against this background, we can address the major ethical points raised by the commentators. We agree with Rottschaefer on the need to establish a body of facts as value free as possible, which can be consulted by a set of values to derive courses of action. Therefore, it is important to be explicit about a fact/value distinction. The body of facts can be greatly clarified from an evolutionary perspective, which does not challenge the naturalistic fallacy. However, the concept of a value system itself needs to be understood from an evolutionary perspective, a project that is already in progress among moral philosophers, without committing the naturalistic fallacy in any naïve sense. From a practical standpoint, we think it is important to be as explicit as possible about one's facts, one's values, and how they are combined to lead to a practical course of action, so that every component of the process can be examined and challenged if necessary.

It is true that any science invented by humans has been used, and most certainly will be used, for good or ill – especially toward other human groups in the latter case, if past is prologue for the future. Our ancestors' invention of stone tools bears witness that we used such tools not just to secure a meal or build shelter: We also used those tools to kill other humans.

The inherent question in our paper is not whether humans can have scientifically proven tools for intentional cultural change: We do, by any sensible measure, which we resolved to illustrate. Most of our reviewers seem to agree that we have done so. The question that Rottschaefer asks and we have tried to answer is how will we use that explicit technology of intentional change in ways that humans agree is “good” or valued by all. The technology of intentional change is not the question: It is how we use it.

Nearly all the commentators who focused on ethics mentioned the sorry history of social Darwinism in the past. This is understandable, but it is also important to draw the right conclusions from this history. Take the slaughter of one group by another, as an example. It is first and foremost an outcome of the between-group competition that takes place throughout the animal world (e.g., ant colonies raiding other colonies) and has taken place throughout human history. It has been justified by religion, patriotism, and intellectual ideologies of all stripes. It should surprise no one that evolutionary theory was added to the arsenal of justifications for exterminating other groups, once it became available. When used, it became a causative factor in a mechanistic sense, but does this mean that removing it as a causative factor will reduce the problem of between-group conflict? Perhaps not, given so many other functionally equivalent justifications to choose from (back to Tinbergen's distinction between function and mechanism).

We certainly agree with the need to be vigilant about ethics in general terms, but we should not be unduly influenced by the justificatory role that evolution has played in the past. Efforts to alter the course of genetic evolution provide an example. The idea of selectively breeding people predates Darwin, as it is suggested by the artificial selection practices of plant and animal breeders for centuries. It was implemented as a policy in nations such as England, America, and Germany under the belief that because traits such as criminality and joblessness were under strict genetic control, it was ethically acceptable for the state to deny some of its citizens the right to reproduce. We should never forget this history and the role played by evolutionary theory per se, but neither should we allow the history to place current considerations of genetic evolution for “evolving the future” off-limits (contra Grinde). A key discovery in evolutionary biology during the last half century is that evolution takes place on ecological time scales, in our species no less than in others. We think that current knowledge, which emphasizes the interplay of cultural, developmental, situational, genetic, and epigenetic factors in the expression of undesirable traits, will provide the best protection against naïvely eugenic prescriptions and lead to policies that are ethically humane.

Sarkar raises a number of important ethical issues, including (1) the distinction between group and individual interests, (2) acknowledging cultural diversity, (3) focusing on normative issues before rushing to intervention, (4) insights from a variety of disciplines such as philosophy and cultural anthropology, and (5) the difficulty of achieving normative consensus on the largest and most recalcitrant problems. We agree, and wish only to stress the added value of considering these issues from an evolutionary perspective. People distrust being told what to do “for the good of the group” for the best of reasons: because it makes them vulnerable to exploitation. Moral systems that include respect for individual rights are protected against exploitation from within. Taking cultural evolution seriously means attaching the same importance to cultural diversity that evolutionary biologists accord to biological diversity. Human moral systems require shared norms of good conduct enforced by rewards and punishment. These norms must be established before action is taken. Disciplines such as philosophy and cultural anthropology are not rendered obsolete but are essential to the integration. Elements of moral systems that take place relatively easily at small scales, such as reaching a consensus on normative issues, take place with more difficulty at large scales, requiring proceeding with caution. To Sarkar's point, this is why we emphasized the Nobel Prize work of Elinor Ostrom on common-pool resources, which provides an empirical way to resolve some of the ethical issues associated with human exploitation and fairness.

Khalidi and Aitken stress the poor track record of social engineering in the past and the inherent unpredictability of complex human social systems that makes prediction difficult or impossible. They seem to think that we have Watson's (Reference Watson1925) boast about molding individuals or Skinner's Walden Two (Reference Skinner1948) in mind. We turn to the issue of complexity in the next section.

R3. Complexity

A major theme of the commentaries concerns the need to adopt a systemic approach that does justice to the mechanistic complexity of human social systems (Andersson et al.; Bodor & Fokas; Grotuss; Ho et al.; Kostrubiec & Kelso; O'Brien; Read; and Smaldino & Waring). Particular approaches to complexity that were discussed include developmental systems theory, niche construction theory, generative entrenchment, gene-culture coevolution, and dynamical systems theory. Bodor & Fokas make the important point that the most profound and important social changes, such as the attainment of Western-type social organization, evolve over a period of centuries and not by anyone's intentions. The same goes for many technological innovations. The implication seems to be that intentional change should not and perhaps cannot entirely substitute for unintended cultural evolution.

Our reliance on Tinbergen's four questions indicates our willingness to acknowledge the importance of mechanisms and development in all their complexity – but that is a far cry from doing them justice. There is a reason why the approaches listed above are relatively new trends in evolutionary biology. Their very complexity makes them more difficult to study than functional approaches that treat heritable variation as a black box. Only very recently has our understanding of biological mechanisms reached the point where all four empirical questions can be asked in conjunction with one another. Comparable understanding of the behavioral and symbolic Darwin machines that underlie cultural evolution lags far behind. As the theoretical biologist Robert Boyd (personal communication) puts it, our understanding of the mechanisms of cultural evolution is comparable to our understanding of the mechanisms of genetics before Mendel.

Learning how to accomplish positive change in systems that are too complex to make accurate predictions about requires us to combine the best of our knowledge with experimentation. This observation might seem mundane, but explicit experimentation is by no means the norm in many policy circles. Even when policies are evidence based, the evidence is typically gathered in certain settings and may not apply to other settings. One of the messages of Elinor Ostrom's work is that even though the core design principles provide a recipe of sorts for creating efficacious groups, cookie-cutter implementations do not work because the best implementations depend so much on local context. There can be no single fisheries policy for the coast of Maine, for example, because every bay will require special considerations. Creating effective social cooperation requires an approach to experimentation that is sensitive to local context, which is new even for policy circles that strive to be evidence based in other respects.

We think that an explicit evolutionary perspective will add considerable value to this enterprise, beginning with the observation that the human ability to learn and transmit useful information evolved by genetic evolution in the context of small groups and can be expected to break down in larger groups. Unless additional mechanisms evolve by behavioral and symbolic evolution that interface with our genetically evolved mechanisms, then cultural evolution will work poorly. Such mechanisms have evolved over the course of human history, which means that we should study traditional cultures (including religions) very respectfully to learn how they function as well as they do. Read's description of the Netsilik Inuit provides an example of this kind of functional contextualist approach.

If the mechanisms of cultural evolution are a product of gene-behavior and gene-symbolic coevolution, then what worked in the historical past does not necessarily work in the present, given the enormous changes in social organization and communication technology that have taken place. We should focus on examples of failures of cultural evolution, such as the absence of sufficient variation, practices that work but do not spread (such as that discussed by Gray), and practices that are harmful but nevertheless spread because they are falsely perceived as good. We think that many examples of dysfunction in modern life can be traced to a breakdown of the mechanisms underlying cultural evolution, similar to the breakdown of the immune system that occurs when modern environments depart from ancestral environments (Hanski et al. Reference Hanski, Von Hertzen, Fyhrquist, Koskinen, Torppa, Laatikainen, Karisola, Auvinen, Paulin, Mäkelä, Vartiainen, Kosunen, Alenius and Haahtela2012).

Elinor Ostrom provides a simple example for common-pool resource groups in one of the last articles that she completed before her death (Ostrom Reference Ostrom2013). In a small-scale society, useful practices that are learned by some individuals are quickly noticed and copied by other individuals. Not so for a modern common-pool resource group, which has a limited ability to communicate with other groups – despite the massive growth of communication technologies that are driven by aberrations in selection by consequences such as advertising revenues but have not yet been behaviorally integrated into group functioning. By the simple expedient of facilitating interactions among common-pool resource groups, they were able to compare notes with one another, enabling best practices to spread when they had not before. This humble example illustrates a critical point: The parameters of an efficient cultural evolutionary process in modern life must be constructed. Neither our genetically evolved abilities nor past behavioral and symbolic adaptations are sufficient. This is what we mean by becoming wise managers of evolutionary processes. It means not knowing the exact solutions, but orchestrating a process whereby the solutions can be derived and spread.

Another important point illustrated by Ostrom's example is that the variation part of a variation-and-selection process is often unplanned. Members of common-pool resource groups tinker with their arrangements. Candidate solutions are often based on serendipity and happenstance. The best solutions quickly spread within a given group, but spreading to other groups will not take place unless a comparison-and-selection process is orchestrated. In short, a well-managed cultural evolutionary process can leave room for unplanned variation, as stressed by Bodor & Fokas.

One important property of complex social systems is that variation can be expected at all spatial scales, because small chance differences result in larger divergences rather than remaining small (sensitive dependence on initial conditions; Gleick Reference Gleick1987). Hence, large-scale social units such as nations vary in ways that no one planned but that are highly consequential for military and economic competition over the course of centuries, as stressed by Bodor & Fokas. Variation among modern nations remains highly consequential for their capacity to function as corporate units (Acemoglu & Robinson Reference Acemoglu and Robinson2012; Pickett & Wilkinson Reference Pickett and Wilkinson2009), but the comparison and selection of best practices is highly unlikely to take place at this scale without an orchestrated process informed by evolutionary theory.

R4. Symbotypes, culture, and the future

To varying degrees, some commentaries (Baumard; Costanza & Atkins; Grinde; O'Brien; Read; and Wang, Li, & Rao (Wang et al.)) drew an equivalence between our comments about symbotypes and evolution of cultural practices per se. Cultural evolution includes behavioral processes that are not symbolic, and the implications at times differ. In some areas our comments were misunderstood or overextended as a result of failing to track this distinction. For example, Read is concerned that we have the idea that “what is expressed in the symbolic/cultural domain is an epiphenomenon of prior patterning.” The basis of the criticism was our suggestion that learning to relate events symbolically is in part a matter of abstracting from physical relationships but becoming somewhat independent from them. For example, small children initially learn to compare based on physical properties such as relative size, but as comparison per se is abstracted and brought under the control of social cues (e.g., “this is bigger than that”), it can be arbitrarily applied (e.g., a dime is bigger than a nickel). This view of one of the key relational processes that allow symbotypes to be constructed is similar to the processes of combining and reading nucleotides in the genotype, but it is quite different from assuming that culture is a “codification of already existing patterns of behavior,” as Read understands us to be saying. To the contrary, the combinatorial complexity of these basic relational or symbolic units in their capacity to produce actions is as flexible in various physical and social environments as the combinatorial complexity of the genotype is in its ability to produce a variety of phenotypes in different environments.

Grinde likewise reduces our multidimensional and multilevel argument about evolving the future to distinctions between genetic and cultural evolution. We agree there are important differences, but the plasticity of evolutionary processes is impressive everywhere we look. Baumard is skeptical that cultural practices can change human psychology – but symbolic functions are part of that psychology, and we cannot know how far we can take intentional change until we bring together all of the sciences needed to make such an attempt. In this area Fox's central point seems apt: A more detailed awareness of how symbolic behavior evolved and functions is a key part of the task ahead for an evolution science focused on intentional change.

We agree that knowledge is very much needed, and it will be central to answering MacDonald's call for a program of research that carefully teases out “domain-specific, modular mechanisms from domain-general mechanisms, and in the case of the latter, must be clear on exactly how they are domain-general and how this promotes change.” To the extent that MacDonald is correct that domain-general processes in the behavioral area evolved via affective cues signaling the attainment of evolutionary goals, symbolic behaviors challenge this general process, as human beings can visit pain on any situation through symbolically driven memories, comparisons, and fears. So far it is applied behavioral scientists and psychotherapists who have lived at the edge of this conflict, and many of their insights seem to provide guidance for moving forward, as we emphasized in the target article. Indeed, we know that acceptance and mindfulness-based therapies work in part by reducing the avoidance of such responses, thus increasing emotional, cognitive, and behavioral flexibility (Hayes et al. Reference Hayes, Luoma, Bond, Masuda and Lillis2006). But without a deeper and evolutionarily sensible understanding of symbolic functions, the precision of predictions in a variety of important areas such as these will be weak.

Fox points to a key reason why a deeper understanding of symbotypes is needed: the possibility that we could learn to recast our cultural and personal dialogues about war and peace, much as we are learning to do in approaches that have psychotherapeutic uses. He is right: “If we desire to survive and flourish as a species, then we will need to select for ways of thinking and behaving that favor peaceful outcomes.” As psychotherapeutic ideas have been applied to social issues such as stigma and prejudice, it has become clear that some of the ideas about symbolic evolution can indeed scale from the clinic to the society (e.g., Masuda et al. Reference Masuda, Hayes, Fletcher, Seignourel, Bunting, Herbst, Twohig and Lillis2007).

Nowhere is knowledge of symbolic functions more critical than in our understanding of the future. In the absence of symbolic behavior, the actual experience of the future boils down to the experience of change from one present moment to another. That experience is built in the past, resides in the present, and is “about” the future in the original etymological sense of the word about: that is, it is on it or near it, while at the same time being outside of it.

The learning process of the extension of the experience of change into the future is put on steroids once symbolic functions arrive. Human symbolic behavior is bidirectional and combinatorial. We can now construct futures that have never been. As LaFreniere points out, our “human cognitive ability [is critical in creating] novel solutions to various human goals.” Costanza & Atkins likewise emphasize the human ability to bring foresight to the evolutionary table by deliberately envisioning the future we want through the use of symbols. They are right to suggest that having different symbotypes available is advantageous, but not if diversity overwhelms the ability to transmit and share common visions for the future. In the same way, the construction of the future needs to be guided, as they note in their explication of scenario planning.

The processes of symbolic construction of change (if…then; before…after) are not bound by the changes that have actually been experienced in the past. The combinatorial properties of the symbotype allow us to do that in a way that is both creative and based in experience, as the human construction of the future can be guided by how such novel constructions have previously helped orient us to the future. That is precisely what scientific knowledge does.

Dowrick rightly focuses on how we learn from envisioned futures. Humans cannot be in the future, but they can project and entertain possible futures. Fundamental to the idea of intentional cultural change is a playful and purposeful envisioning of futures. Dowrick's work is cited in Embry and Biglan's (Reference Embry and Biglan2008) article on evidence-based kernels, and is used in some of the example studies to change human behavior to decrease unwanted outcomes (e.g., less violence, less addiction) and increase wanted or desired outcomes (e.g., more peaceful, prosocial environments; better health). For example, the underlying implementations of the real-world versions of the Good Behavior Game (Embry et al. Reference Embry, Richardson, Schaffer, Rosen, Darney, Kelly, Rouiller, Muempfer and Pitchford2010), as well as earlier work on peaceful, positive schools (Embry et al. Reference Embry, Flannery, Vazsonyi, Powell and Atha1996), and the national injury control strategies in New Zealand and the United States (Embry Reference Embry, Paine and Bellamy1984; Embry & Peters Reference Embry and Peters1985; Embry et al. Reference Embry, Rawls and Hemingway1985) directly involved Bandura's intentional imaginary modeling and Dowrick's self-modeling work in future contexts.

There are multiple ways of using the envisioned future. One way is to use data trajectories, which we do not believe Dowrick opposes. That is what programs such as those just described have done. Another is to extend what is meaningful and purposeful: that is, to consider human values. As Peschl & Fundneider note, our paper implicitly includes the change strategy of “learning from the future as it emerges.” We concur, and thank them for their observation.

Dowrick suggested that we made no effort to include the representation of valued future behavior as an important part of human learning; but although we did not do a deep dive into the issue, the importance of applying values to our construction of the future is mentioned half a dozen times in our article, primarily drawing on the psychotherapy methods we were discussing. It is a mischaracterization of the work of translating relational frame theory into ACT or the work cited by Embry and Biglan on prevention as “limited almost entirely to clinical, developmental, and educational problems” (italics original). The application of these ideas is not limited to problems – they apply to human development more generally. For example, when 30 minutes of online exposure of ACT-based values training is provided to college students and they consider what they want out of their education, their grade point averages increase significantly over the next semester, whereas mere academic goal setting has no such effects (Chase et al. Reference Chase, Houmanfar, Hayes, Ward, Vilardaga and Follette2014). Acceptance, mindfulness, and values methods help athletes compete (Bernier et al. Reference Bernier, Thienot, Codron and Fournier2009), chess players play (Ruiz & Luciano Reference Ruiz and Luciano2012), health professionals learn (Varra et al. Reference Varra, Hayes, Roget and Fisher2008), and gym members exercise (Butryn et al. Reference Butryn, Forman, Hoffman, Shaw and Juarascio2011).

Our article highlighted the potential for reducing major social problems because many believe those problems to be intractable or not avertable; but as Dowrick suggests and we affirm, these methods are means of improving development, not just remediating deficits. By showing that there are powerful, simple mechanisms for reducing costly problems through intentional change, a stage is set for a broader discussion that includes moving toward valued futures in a positive sense.

We agree with Manzotti & Moderato that we need to understand intrinsic motivation – how an intentional cognitive agent may produce a new goal. That is precisely in line with the need to understand the ontogeny of symbotypes and its resulting teleological effects, which is another way of restating their idea that we require an understanding of domain-general cognitive architectures.

It is worth noting that values constructions are still only “about” the future in that original etymological sense; but science itself provides more tools every day for these symbolic extensions, and in a deep sense, that task is a central aspect of the very history of science itself. That being said, we agree with commentators such as Sarkar, Khalidi, and others who suggest that we need to be humble and cautious. We are not claiming we currently have all or even most of the tools needed to know all aspects of the future or to control it with precision, nor that we can currently extend our knowledge with precision across very large time spans or contexts in most areas. Bodor & Fokas take caution too far, however, in claiming that it is actually undesirable to seek long-term change. Areas such as global climate change demand that we do a better job of acting with regard to the future. It is our argument that only by drawing together scientific knowledge into a greater degree of consilience across the full range of scientific disciplines are we likely to be able to acquire such knowledge. The umbrella provided by evolution science is the means we see to do that.

Kostrubiec & Kelso make a useful point in noting that some forms of selection (e.g., “selection via instability”) can be more transformational than other forms. The symbolic domain is one in which that kind of transformational change is especially likely. When the entire purpose or central organizing assumption of a cognitive network is changed, everything changes. That possibility undergirds many of the examples provided in our target article. A deeper understanding of symbotypes may therefore be an especially fruitful area in which to explore Kostrubiec & Kelso's ideas because it will reflect so directly on what we even mean by such issues as “intention.”

This concerned some commentators. We are using the word intentionality to mean that intervenors or those affected are guided by a verbal purpose or envisioned future. Aunger & Curtis argue that most of the behaviors people seek to change are habitual and therefore “not necessarily responsive to intentions.” This idea turns intentionality from a feature of the purpose of change to a narrow method of change (Aunger & Curtis appear to use “intentionality” as a synonym for “consciously known instructions”), which fundamentally narrows our arguments. Their randomized trial of a hand-washing intervention is a policy-driven public health effort using evidence-based kernels for cueing (the “eye-spots”) and the alteration of symbolic relational networks by pledges. It is a wonderful example of evidence-driven intentional cultural change. The intentionality is in the actions of the program organizers and in the values of the individuals who do not wish to contract deadly or harmful diseases while in a public restroom. The method itself certainly need not be instructional or even conscious.

Wang et al. draw the same distinction in their point that “education via symbotype is necessary but not sufficient” and that change depends on establishing an effective link between intentional behavior and an intrinsic and emotional consequence. We agree, especially as many of the behavior change methods we describe seem to work in that way; we simply wish to add that by “intentionality” we did not mean that the methods of change should be instructional or that all those involved in a cultural change must have a conscious intention.

R5. What intentional cultural change might look like

The human sciences are already guiding cultural change to a much greater extent than ever before. We presented examples of evolving science-based practices that are decreasing the prevalence of numerous common and costly problems of human behavior. Much of the work thus far has focused on altering these problems at a relatively small scale; even the largest-scale interventions reviewed in this article are at the level of counties and states. However, there are a growing number of efforts under way to bring evidence-based preventive efforts to scale.

Perhaps the most successful example of this effort is the tobacco control movement, which has reduced the prevalence of smoking in many countries by altering a wide range of policies and practices that at the outset constituted a culture of smoking. The principles that underlie the smoking control movement are generic ones that apply to the intentional cultural change we have in mind. Indeed, the movement evolved out of the more general evolution of public health efforts to improve human health that began with sometimes desperate efforts to combat and prevent epidemics.

At the same time, we believe that a substantial convergence is emerging from research in epidemiology, public health, prevention science, neuroscience, and epigenetics that points to specific patterns of behavior that have well-established relationships to human well-being and specific types of environments that select these behavioral patterns and are, therefore, appropriate targets for efforts to influence the evolution of cultural practices. In the interest of space, we summarize this evidence within the framework of the principles that have guided tobacco control.

Identify factors that harm human health, giving priority to the most prevalent and deadly. This principle was initially established with epidemics, but it has been steadily extended to the causes of epidemics once those causes were recognized (e.g., sanitation, the presence of pathogens). Smoking became a target of public health efforts because it was linked to death from cancer, heart disease, and over the years, a growing number of other threats to life (U.S. Department of Health and Human Services 2000; 2004). This same epidemiological principle has been followed in order to identify numerous other unhealthful behaviors, including excessive alcohol use, drug addiction, academic failure, depression, and anxiety (Ainsworth Reference Ainsworth2002; McEvoy & Welker Reference McEvoy and Welker2000; Munoz et al. Reference Munoz, Beardslee and Leykin2012; National Research Council and Institute of Medicine 2004; 2009a; 2009b; Shonkoff Reference Shonkoff2003; U.S. Department of Health and Human Services 1999; Walker et al. Reference Walker, Cheney, Stage and Blum2005).

Converging evidence points to two contrasting clusters of behavior that have different implications for the well-being of individuals and those around them. One cluster might be called “prosociality.” It consists of a set of behaviors, attitudes, and values that have to do with helping others, contributing to the community, and growing as a person (Biglan & Embry Reference Biglan and Embry2013; Kasser Reference Kasser2002; Reference Kasser, Linley and Joseph2004; Kasser & Ryan Reference Kasser and Ryan1993; Wilson et al. Reference Wilson, O'Brien and Sesma2009). This cluster is associated with greater personal well-being and is beneficial to the group. Indeed, there is evidence that nations that have higher proportions of prosocial individuals also have stronger public policies supporting families and lower levels of carbon emission (Kasser Reference Kasser2011).

In contrast, a set of psychological and behavioral problems that include antisocial behavior, drug abuse, cigarette smoking, excessive alcohol use, academic failure, and depression are highly inter-related (Biglan et al. Reference Biglan, Brennan, Foster and Holder2004).This cluster is harmful to individuals and to those around them. Young people with these problems are also more likely to endorse and pursue values having to do with fame and materialism, values that are associated with more problematical personal outcomes (Sheldon & Kasser Reference Sheldon and Kasser1998).

This epidemiological principle also extends to the environments. We are interested in the environments that select these two types of behavior. Considerable research shows that among children and adolescents, the complex of psychological and behavioral problems is selected in family and school environments that fail to nurture prosociality. Specifically, such environments (1) have high levels of socially and biologically toxic conditions, (2) fail to richly reinforce a wide variety of skilled prosocial behaviors, (3) fail to monitor and set limits on problem behaviors, and (4) fail to promote the kind of psychological flexibility that we described in our target article (Biglan et al. Reference Biglan, Flay, Embry and Sandler2012).

Establish a goal of reducing the incidence and prevalence of the problem. One reason for the progress of the tobacco control movement has been its laser-like focus on addressing all of empirically established factors that influence the prevalence of smoking. The application of this principle in the present case implies that we need to increase the prevalence of family and school environments that select prosociality and minimize psychological and behavioral problems. That, in turn, implies that from a cultural evolution perspective, we are interested in how we can select environments that select prosocial behavior (Biglan et al. Reference Biglan, Flay, Embry and Sandler2012).

Pragmatically implement whatever programs, policies, and practices can be shown to alleviate the targeted problem or reduce the risk factors that contribute to the problem. The tobacco control movement implemented whatever worked to affect smoking (Biglan & Taylor Reference Biglan and Taylor2000). In our target article, we described one example of a family intervention that has been successful in decreasing the prevalence of abusive and neglectful environments. Other studies of this intervention and numerous other behaviorally oriented family interventions (e.g., Forgatch et al. Reference Forgatch, Patterson, DeGarmo and Beldavs2009; Shaw et al. Reference Shaw, Dishion, Supplee, Gardner and Arnds2006; Webster-Stratton Reference Webster-Stratton2000) show that they can prevent the development of the range of psychological and behavioral problems described above. We also described interventions that make schools more nurturing of prosociality. And we described kernels that are beneficial in a wide variety of environments. In short, although there is much to be learned, we agree with the conclusion of the National Research Council and Institute of Medicine (2009b) report on prevention: In principle, we have the scientific knowledge to ensure that virtually every young person develops the skills, interests, values, and habits needed to become a productive and caring adult.

We are unclear about the distinction that Aitken makes between treatment and prevention interventions and “moulding the process of evolving more successful behaviour.” We believe that the further dissemination and refinement of the strategies we describe are the very processes needed to evolve more successful (prosocial) behavior.

Monitor the incidence and prevalence of the problem and of risk factors that would affect the problem. Monitoring is an essential component of the tobacco control movement and all other public health efforts so that increasingly more effective methods of reducing the problem can be selected (Biglan & Embry Reference Biglan and Embry2013). A system for monitoring the psychological and behavioral problems described here has been evolving for at least 40 years in the United States. Monitoring the prevalence of nurturing families and schools is in its infancy. We think it likely, however, that as the central importance of these environments becomes clear, appropriate monitoring practices will be widely implemented.

R6. An evolving science of cultural change

The development of a science of cultural change is itself an evolutionary process. Contra Aitken, we do not believe that scientific understanding of cultural change is fully developed, and doubt it ever will be, as each major scientific “answer” raises new questions. Rather, we see the processes we have just described as ones that will be continually refined in light of their consequences. Those consequences will include not only the effects of specific change strategies on their targets, but also the acceptance or opposition of citizens and policy makers. Indeed, for the kind of change that is needed, we will have to further develop an evolutionary understanding of the major social forces affecting policy development, including the evolution of capitalism (Biglan Reference Biglan2009; Reference Biglan2011; Biglan & Cody Reference Biglan and Cody2013; Biglan & Embry Reference Biglan and Embry2013).

We are more optimistic than Khalidi is about the prospects for predicting the impact of societal change efforts, but only within fairly circumscribed boundaries. Specifically, we believe that the accumulated evidence on the impact of family, school, and clinical interventions shows that we can construct environments that reliably (though far from infallibly) result in prosocial behavior and the prevention of diverse psychological and behavioral problems.

At the same time, we think that caution about any intentional effort to change cultural practices is warranted (Biglan Reference Biglan1995). The numerous examples of human excesses in the effort to influence others' behaviors (e.g., Pinker Reference Pinker2011) make clear that any effort to change the evolution of individual or societal behavior must be guided by a system of safeguards. We suggest that the core design principles derived by Ostrom (Reference Ostrom1990; Reference Ostrom2010) for common-pool resource groups and generalized by Wilson et al. (Reference Wilson, Ostrom and Cox2013) provide a very workable system of such safeguards, which cuts well across different cultures and issues.

As several commentators suggest, any effort to use evolutionary science to guide cultural evolution will require ongoing discussion and the development of safeguards against misuse of that science (as well as every other practice that could harm well-being). Such safeguards are emerging. Prominent examples include the 1975 Helsinki Accords, the 1948 Universal Declaration of Human Rights, and the policies of the National Institutes of Health regarding research involving human subjects.

Some of the most important practices that have evolved in the economically developed world involve corporations and the market system that shapes and maintains these practices (Biglan & Cody Reference Biglan and Cody2013). We raise this issue for two reasons. First, we believe that the system of corporate capitalism that has evolved is one of the most important influences on the further development of society. Any effort to improve the prospect that our societies will evolve in a direction that increases the prevalence of prosociality and well-being will require that we develop an effective analysis of how to guide the further evolution of corporate capitalism to support prosociality (Biglan Reference Biglan2009; Reference Biglan2011; Biglan & Embry Reference Biglan and Embry2013).

Second, if those with an understanding of behavioral and evolution science hesitate to use that understanding to influence cultural evolution in a prosocial direction, the further evolution of society is simply left to other groups and organizations, many of which are not explicitly focused on improving human well-being. Many of these organizations are themselves making deliberate use of behavioral sciences. For example, the tobacco industry has used very sophisticated research to develop effective ways to market cigarettes to young people and to influence smokers not to stop smoking (National Cancer Institute 2008).

Proposals to influence cultural practices have often been opposed because it was assumed that they must necessarily involve the coercive power of the state (Chomsky Reference Chomsky1971). However, the examples we provide make use of positive consequences. Indeed, they typically replace coercive approaches to change. Consider, in particular, the practices of parents. There is growing evidence about the kinds of parenting practices that harm children and make it more likely they will develop psychological and behavioral problems that are costly to them and to society (and that contribute to a lower life expectancy). Yet efforts to target coercive, neglectful, or abusive parenting practices must take place in the context of a democratic society with established policies that protect families from state intrusion. At present, the state can take children away from abusive and neglectful parents, but there are limits on when it can do that and, as a practical matter, there are limited resources for detecting such abuse.

But one can engage in intentional cultural change that prevents, averts, or reduces accepted indicators of child maltreatment by creating easy access for any family to learn scientifically proven noncoercive parenting practices in communities. That is precisely what happened when Prinz et al. (Reference Prinz, Sanders, Shapiro, Whitaker and Lutzker2009) offered evidence-based parenting supports to every family in an 18-county study. Simply making such practical tools easily accessible, with no coercion or shaming, had large effect sizes (es) on reducing multiple indicators of maltreatment (the smallest es = –1.09). The strategy cost only $15 per child. Given the effects of adverse childhood experiences on health, relationships, and life attainments (Shonkoff Reference Shonkoff2003; Shonkoff & Phillips Reference Shonkoff and Phillips2000; Shonkoff et al. Reference Shonkoff, Lippitt, Cavanaugh and Zeanah2000), this would seem to qualify as a social good. Prosociality does not imply an absence of social harm. It is possible that some were harmed by a major reduction in child maltreatment in a local sense. Fewer prison cells will be needed, and the businesses that supply those cells will be harmed; fewer people will need expensive health care over their lives, and those providing that care may make less money, and so on. Democratic choices need to be made to balance these social costs.

If one examines the details of the family interventions we described in our paper and those of the many other evidence-based family interventions, one finds that they do not employ coercive means. Indeed, a fundamental principle of all of these interventions (e.g., Chamberlain Reference Chamberlain and Chamberlain2003; Forgatch et al. Reference Forgatch, Patterson, DeGarmo and Beldavs2009; Shaw et al. Reference Shaw, Dishion, Supplee, Gardner and Arnds2006; Webster-Stratton Reference Webster-Stratton2000) is that interventionists empathically join parents around the goals that they have for their families and their children. Just as the coercive behavior of parents toward their children is problematic, it is counter-productive to try to coerce parents to change their parenting behavior. Far from abrogating individual rights and coercing change, the interventions that have been developed involve empathic involvement with children and adults that minimizes coercive efforts to change behavior.

So, while acknowledging that there are risks to any effort to change behavior, our understanding of the key ingredients of the most effective treatment and prevention interventions makes us less concerned than is Khalidi that there will be “ethical costs of interventions involving social control that may have unforeseen consequences.”

R7. Who decides?

Scientifically guided efforts to influence the direction of cultural evolution should have no special status in relation to other efforts. Each must compete in the marketplace of ideas, at least within democratic and capitalist societies. If 100 years from now, people look back at the history of cultural change and conclude that cultural change was increasingly guided by an evolutionary science of intentional change, it will be because a growing proportion of the population found that the guidance provided by such a science was preferable to other systems for enhancing their well-being.

We should not be naïve, however, about the social forces that are influencing cultural evolution currently. Social interests collide and groups often pursue their interests selfishly. For example, corporate capitalism has influenced numerous policies, some of which have harmed the well-being of families and schools (Biglan & Cody Reference Biglan and Cody2013). An effective analysis of how such practices are selected and can be modified will be key to learning how to guide cultural evolution. In the modern world, the “we” who decides has grown. In the modern world, it includes all of us.

Wang et al. cite cultural differences as well as organizational memberships affecting the important issue of “us versus them” for thinking and acting for intentional change on such issues as helping others, reducing conflict, and caring for the environment. “Us” and “them” can be modified, however, even when an individual or group within the larger group would benefit by being “selfish.”

Consider a couple of examples from the paper. The Good Behavior Game creates group interdependent rewards, which means that the child has to “give up” high levels of social reinforcement as an individual for engaging in deviant behavior. This simple strategy immediately has effects on “selfish” behavior in more than 70 studies (Tingstrom et al. Reference Tingstrom, Sterling-Turner and Wilczynski2006). What is extraordinary, however, is that prospective, randomized control trials show that a year of exposure to this simple recipe in first grade has a more than 20-year impact on reducing the pursuit of antisocial behaviors (Kellam et al. Reference Kellam, Mackenzie, Brown, Poduska, Wang, Petras and Wilcox2011). This is not the only example of using interdependent group rewards to reap long-term positive impacts on children (Greenwood Reference Greenwood1991a; Reference Greenwood1991b). These changes in classroom environments apparently “reset” the early developmental and evolutionary markers that predict human life histories.

Perhaps the best large-scale example from our paper of swapping out short-term individual or small group gains for longer-term social gains for unknown persons is the Reward & Reminder protocol that publicly recognized clerks and stores for not selling highly profitable tobacco products. Any rewards received by clerks were small, and no store received a financial incentive. About 30% or more of the profits of many outlets come from tobacco products. Yet, a status/reputation symbol and small rewards caused tobacco outlets to forgo substantial revenue in a short period of time and to maintain that behavior for many years.

As many commentators note, intentional change has its perils. The possibility of harm is not avoided by maintaining ignorance of, or denying the faculty for, the capacity for intentional change, nor by the pretense of innocence so as to avoid a decision to act one way or another. The lack of perfection for the prediction and influence of human behavior does not eliminate the need for us to continue to learn how to do so at the scale needed to confront the challenges we face. Humans have the capacity to alter our physical and social world intentionally, with effects good and ill, and we do so every day at both large and small scales. Explicitly and implicitly, we have already developed a number of technologies of intentional change. We have eaten from the tree of knowledge, and we left the garden of innocence long ago. Now we must learn how to systematize that knowledge and use it intentionally to leverage our best ideas and practices from the full range of relevant disciplines. Evolution science provides a way to do that. It is time to begin.