Are Plants Rational?

Elias L. Khalil Department of Economics Monash University Clayton, Victoria, Australia [email protected]

Abstract Organisms change their shape and behavior during ontogenesis in response to incentives—what biologists call “phenotypic plasticity” or what is called here more specifically “behavioral plasticity.” Such plasticity is usually in the direction of enhancing welfare or fitness. In light of basic concepts in economics, such behavioral plasticity is nothing but rationality. Such rationality is not limited to organisms with neural systems. It also characterizes brainless organisms such as plants, fungi, and unicellular organisms. The gist of the article is the distinction between rationality and intelligence. Whereas rationality is ubiquitous in all organisms, intelligence varies in degrees depending on division of labor that may involve the evolution of a neural system. This article aims to defend the universality of rationality—the Organismus oeconomicus hypothesis. It argues that neither the notion of bounded rationality of behavioral economics nor Herbert Simon’s “procedural rationality” can ultimately undermine the Organismus oeconomicus hypothesis. Keywords bounded optimization, bounded rationality, Organismus automaton, Organismus oeconomicus, procedural rationality, rationality optimization, selection optimization, unbounded optimization

April 15, 2009; accepted April 4, 2010 c 2010 Konrad Lorenz Institute for Evolution and Cognition Research Biological Theory 5(1) 2010, 53–66. 

53

Are Plants Rational?

People usually refer to humans in a coma as being in a “vegetative state.” Insofar as such an allegory conveys the belief that plants do not make decisions, this article disputes it. Plants are living organisms and have to make decisions. A plant has many parallels with a business firm: From the environment a plant must acquire resources (water, nutrients, carbon, etc.) that it can then save (storage) or spend in various ways to construct a product (leaves, stems, roots, etc.). This product can then assist in the acquisition of additional resources. In addition, both plants and businesses must operate under changing conditions. A highly successful strategy in one climate may prove disastrous in another: Optimal solutions are usually ephemeral. Thus, plants and businesses must engage in long-term as well as short-term planning. (Bloom et al. 1985: 363–364)

But these authors quickly note that their description only amounts to a useful analogy. The purpose of this article is to argue the opposite. Plants are rational decision makers in the substantive and not only in the analogical sense. Plants are rational in the same sense as are business firms, humans, and one’s favorite species of animals. According to standard economic theory, humans are rational insofar as they adjust behavior in the direction of welfare enhancement in response to a change in the environment (e.g., Becker 1976: ch. 1). Plants are rational in exactly the same sense. Plants adjust behavior in response to the change of scarce minerals, water, sun exposure, and so on, as evident in any elementary textbook on the life of plants (e.g., Tudge 2005). Such nutrients are mostly fungible, i.e., can be substituted in light of their scarcity. Such adjustment of behavior in light of incentives, i.e., rationality, is not limited to plants (Eichinger et al. 2005). Given that organisms adjust their shape and behavior in the direction of welfare enhancement, such organisms—from brainless plants to amoeba, fungi, and apes—are rational. This thesis is not an audacious conclusion for neoDarwinians who postulate that rationality is anyhow the product of natural selection. These neo-Darwinians argue that rationality is a trait and as such is favored by natural selection in the same way as are, say, beak length, color of fur, body power, mental power, or intelligence. All these traits have diverse lineages that are subjected to the editing forces of natural selection. But the thesis advanced here, viz., plants are rational, does not hinge on the postulate that rationality is the product of natural selection.1 One can even be a non-Darwinian and still uphold the hypothesis that plants are rational in the same substantive sense as are humans. In any case, while neo-Darwinian theory begins with given intra-population differences, the supposed engine of evolutionary change, this article focuses on a commonality, i.e., 54

rationality, among all organisms across all phyla and all kingdoms. It is a commonality that is characteristic of all living entities. If so, decision-making might be the proper point of commencement if we want to understand intra-population differences. That is, if we start with decision-making, i.e., what is the best behavior or shape an organism can adopt in light of environmental constraints, we can have an endogenous account of intra-population differences. So, this article is not in the tradition of neo-Darwinian theory, i.e., starting from differences. It is rather in the tradition of systematic biology, aimed at detecting and examining commonalities across living entities. The approach of systematic biology expresses the thrust of scientific inquiry, viz., the pursuit of parsimony of scientific principles—Ockham’s razor (Sober 1983). The pursuit of parsimony, though, is full of perils and wrong turns. Seen in this light, the expected professional skepticism toward calling plants rational is understandable. At first look, though, plants are engaged in decisions that cannot be called anything but rational. They have to make tradeoffs as they develop and branch out. Plants are very sensitive to their environments. As they reach out for more sun, they may have to give up easy access to water or other resources. There is no free lunch in nature. Plants adjust their behavior as the soil becomes depleted of one resource, i.e., they react properly to the rise of the price of the said resource. But one may object and assert that such an adjustment is merely a manifestation of behavioral plasticity. And botanists have observed for a long time that plants exhibit diverse phenotypic flexibility in light of changing climates, water, resources, predators, and so on (e.g., Grime 1997; Huber and Stuefer 1997; Dudley 2004; Wijesinghe et al. 2005; Griffith and Watson 2006; Richards et al. 2006). As the convention goes, behavioral plasticity is to a great extent allowed by non-rigid genetic coding (e.g., Price et al. 2003). As the argument stipulates, plants can improvise, but only within a range where the genetic code is silent. Such a range of plasticity, although it may not be coded in the genes, nonetheless expresses some programmed norms of behavior. Such a conception of behavior does not differ from how social scientists (e.g., Simon 1957, 1977; Nelson and Winter 1982; Nooteboom 2000) conceive of the behavior of humans as the outcome of routines or norms that express a reaction to environmental stimuli rather than the outcome of rational decision-making of costs and benefits. This article boldly hypothesizes that phenotypic plasticity, rather than being a programmed norm or the outcome of deep genetic code, is the outcome of rational decision-making. The central thesis of this article is that behavioral plasticity is nothing but rationality as understood by economists. In fact, unicellular organisms, such as the amoeba (Schaap 2007; Khalil 2009), are rational insofar as they exhibit behavioral plasticity. If so, plants and other simple organisms do not need brains to be rational. This amounts to the triumph Biological Theory 5(1) 2010

Elias L. Khalil

of the view that all organisms are rational, what is called here the Organismus oeconomicus hypothesis. If so, we can have a unified conceptual apparatus for the study of all living entities. The hidden agenda, not explicitly developed here (Khalil 2009), is that the Organismus oeconomicus hypothesis is as important—if not more—for the study of behavioral plasticity as the theory of natural selection. Sections 1 and 2 define the terms rationality and natural selection, respectively. Section 3 contrasts the advanced Organismus oeconomicus hypothesis with the conventional Organismus automaton hypothesis, viz., the view that organisms are basically machine-like entities. Section 4 defends the Organismus oeconomicus hypothesis in light of the fact that rationality is bounded and, in the process, contrasts rationality optimization with natural selection optimization. Section 5 defends the Organismus oeconomicus hypothesis in light of the fact that behavior is greatly routinized. Routinized behavior has given apparent credence to evolutionary psychology and Herbert Simon’s notion of “procedural rationality.” It is shown that the Organismus oeconomicus hypothesis remains warranted. What Is Rationality? If rationality exists, still how could brainless organisms, from Amoeba oeconomicus to plants, ever be rational? This question is based on the assumption that rationality depends on neural structures such as the brain. The brain is a specialized organ that executes actions also undertaken by brainless organisms with simpler forms of division of labor. In fact, the definition of rationality, as economists understand it, does not require brain or mental/neural activities. Although the neural system is not the focus here, it is sufficient to add that it is beneficial to understand the origin of the neural system. As Rodolfo Llin´as (2001) argues, plants never developed brains because it would be a wasteful capital specialization. Plants do not need brains because they do not need to move when they undertake economic decisions. Brains, as he argues, are needed by animals that move. Brains would help such organisms in making predictions about the environment, which is ever changing with their motion. Of course, plants do move when they climb rocks, seek light, and so on—a subject that fascinated Darwin ([1880] 1989). However, such motion is greatly limited and, hence, according to Llin´as, it would be inefficient for plants to evolve a brain. So, neural structures are something organisms, such as Amoeba oeconomicus, and plants can live without while still being able to make decisions. Still, almost all social scientists would find the above statement somewhat surprising. They usually conflate rationality with mental abilities such as intelligence (e.g., Plotkin 1994: 125–127; Ng 1996; Rumbaugh and Washburn 2003; Herman 2006; Pepperberg 2006). The conflation is not new; it is asserted without argument in key philosophical treatises (e.g., Dewey and Tufts 1908: 306). This article draws Biological Theory 5(1) 2010

a clear distinction between rationality with intelligence. Darwin ([1871] 1998: chs. 3–4) has called intelligence “mental power,” signifying that intelligence is a trait and, as such, does not differ from “body power” such as muscles. If rationality has zero correlation with body power, it has zero correlation with mental power for the same reason. As defined by the American Psychological Association, intelligence is an “ability to understand complex ideas, to adapt effectively to the environment, to learn from experience, to engage in various forms of reasoning, to overcome obstacles by taking thought.”2 Intelligence has been associated with animal innovation (Ramsey et al. 2007), long recognized by ethologists (e.g., Baldwin 1896). As Ramsey et al. (2007) define it, animal innovation is the novel change of repertoire at the level of the organism that cannot be explained by change of incentives (environmental inducement), changes brought about by innate maturation, or changes introduced by conspecifics. Economists treat intelligence, and likewise innovation, as part of the constraint function, which, hence, does not involve rational decision. Animal intelligence likewise does not involve rational decision. Intelligence, the probable source of innovation, is a mental faculty not different from memory, color recognition, and musical talent upon which there is no choice. Intelligence is not even different from quasi-mental faculties such as personal temperament. These faculties are traits, ultimately similar to sharpness of beaks, length of legs, hearing, eyesight, and so on. As for intelligence, it is an ability concerning the processing of information/time. Such ability is costly. Insofar as rationality is not confused with sharpness of beaks, which is also costly, it should not be confused with intelligence. If rationality is not intelligence, what is it? Rationality is simply the idea that the organism changes its behavior in response to the change of constraints, i.e., incentives. When the organism does so, it tries to maximize the product given the constraints. If the product is calories per unit of effort, the organism chooses an action, such as hunting in area A as opposed to area B, based on expectation or causality pattern. The causality pattern is usually called cognitive “belief” when held by human animals. The causality pattern specifies what to expect if one action is taken rather than another. So, it is about cause-and-effect relationships in the environment. The causality pattern, in turn, is also the product of rational decision, viz., it must be the best causality pattern given the information. Furthermore, the information is also the product of rational decision, viz., it must be the best information given the expected cost and benefit of search. We then have the following three layers of rationality: First, the agent decides on the best information given the cost and benefit of search. Second, the agent decides on the best causality pattern given the information. Third, the agent decides on the best action given the causality pattern and other 55

Are Plants Rational?

resources. The best action amounts to the most effective way to coordinate/allocate activity or resources. Rationality at all layers amounts to the same thing: the organization of the diverse traits, including intelligence, so that the organism can make the best decision possible in light of the internal and external constraints.3 Economists call such behavior optimization. It is called elsewhere (Khalil 2009) “rationality optimization” to distinguish it from “selection optimization,” which refers to the diffusion or spread of the most effective trait throughout the population. Rationality is underpinned by a few assumptions: the preferences must be coherent (transitivity axiom), and the agent must be decisive on how to rank them (completeness axiom). These two assumptions, along with minor technical ones (see Kreps 1990), are merely about the consistency of the agent’s goal and hence considered as necessary conditions of rationality. The first assumption, coherence, assures us that agents do not make logical errors: If agents prefer X to Y, and prefer Y to Z, then they would prefer X to Z. The second assumption, decisiveness, assures us that agents do not standby helplessly, i.e., in doubt of what they want. These assumptions are not sufficient to define rationality as understood here. Rationality is action undertaken to maximize an objective function given the constraints. Agents are rational insofar as they modify action in response to a change in constraints (i.e., incentives). Such behavioral flexibility, exhibited by brainless organisms as well as organisms with brains, is the core of rationality as defined here—irrespective of whether one conceives rationality as bounded or unbounded as discussed below. Of course, organisms, mostly human organisms, in many cases choose suboptimal behavior (Khalil 1997b). To uphold the rationality approach does not entail that decisions are always rational, as supposed by the revealed preference approach (Samuelson 1938).4 The actions of plants are rational insofar as plants react to incentives, i.e., they exhibit behavioral flexibility in response to a change in constraints. Plants react to incentives when they change their orientation if re-planted, react to changes of soil moisture, or adjust behavior when exposed to new predators. The repertoires of behavior flexibility can be minute as in the case of the behavior of amoeba and other simple organisms. Nonetheless, amoeba are rational insofar as they do not “walk away” from what they usually consider to be food, or “walk towards” what they usually consider as toxic matter— notwithstanding that some ingestion of toxic matter can be beneficial to make themselves unappealing to predators. If so, plants and amoeba do not need brains to make rational decisions. In simple organisms and in organisms that do not need to move extensively such as plants, there is no single specialized organ to take care of coordination. Instead, in brainless organisms, coordination of functions must be undertaken by 56

less specialized tissues. If this is the case, we should not conflate rationality with the brain, as much as we should not conflate mobility with limbs or digestion with stomachs. Rationality, mobility, and digestion can be the prerogative of generalpurpose parts of the body such as simple cells communicating chemically with their environment or among a group of cells. What Is Natural Selection? As a result of refinements in light of genetics, the modern synthesis of Darwin’s ([1859] 2006) theory and genetics, known as “neo-Darwinism,” can be stripped down to three basic principles: (1) blind mutation; (2) blind random assorting of genes during meiosis and chromosomal crossovers; and (3) differential replication. The third principle means that natural selection allows one type to replicate more than the other types, which eventually leads to its dominance. For the third principle to work, we need variety of types, which the first two principles assure us (Khalil 1993). In fact, for neo-Darwinism to work, we must start with heterogeneity of lineages for each trait. Heterogeneity is the motor of change here. Given the heterogeneity of genotypes, we have more or less corresponding heterogeneity of phenotypes. The phenotype, the traits, can undergo change and development as a result of learning and environmental factors. But such phenotypic change, even when it involves somatic genotype change, cannot feed back and manipulate the germline/germ genotype (e.g., the DNA of semen or egg). This one-way causality is known as the Weissman barrier or central dogma of molecular biology. Contrary to Lamarck’s thesis on the inheritance of acquired characteristics, the Weissman barrier insists that no acquired characteristic can influence the germline.5 Thus, the germline and its heterogeneity in a population are the starting point of the causal process, and protein cannot manipulate it. Of course, the germline undergoes change—but only as a result of blind mutation. The mutations are neither the product of learning nor do they express any ex post benefit. To be clear, while germline mutations are blind, they are not totally random, such as in the case where a lioness can give birth to a cub with bird-like wings. After all, the previous scheme or plan of the organism, or what Gould and Lewontin (1979) call “structure,” imposes limits on the range of viable mutational changes. Such phylogenetic inertia is widely recognized by biologists (Orzack and Sober 2001a, 2001b), and it is not a challenge of natural selection as discussed below. Stated briefly, while genetic mutation is restricted by historical inertia, it is still “blind” in the sense that it does not take place in light of ex post payoff—which is the main pillar of the neoDarwinist program of adaptationism. Blind genetic mutation may not be favored after all. If the new lineage happened to be fitter than others, it would replicate itself more successfully. That is, it would copy itself in a higher number of viable Biological Theory 5(1) 2010

Elias L. Khalil

offspring than other less fit traits. Through a number of generational rounds, the fittest trait becomes more frequent than other trait lineages in the population. For neo-Darwinism, evolution amounts to the inadvertent change of the population—neither as the result of any preordained plan nor as the result of the actors’ intention. There is empirical evidence that may or may not support the seemingly convincing natural selection theory. While empirical and experimental studies are, of course, highly relevant (Orzack and Sober 1994, 2001b), this article is not about whether the data support the selection theory. Also, this article ignores many issues in evolutionary theory such as the debate around the unit of evolution: is it the gene, the organism, or a colony of organisms? (The unit of selection, along with the status of the “species” concept, has bedeviled recent philosophy of biology; Brandon and Burian (1984); Sober (1984); Wilson and Sober (1989).) Whatever is the highest unit of selection, or the relation between population and species, the question of optimization will still persist. That is, the question of optimization is orthogonal to the definitions of unit of selection, species, and population. Organismus oeconomicus vs. Organismus automaton Many behavioral and biological scientists explicitly subscribe to the Organismus automaton hypothesis (Pal 1998; Price et al. 2003; De Jong 2005). This hypothesis portrays behavior as the product of inherited programs or norms, where the organism is supposed to be machine-like and does not make decisions. Interestingly, these scientists would concede that, generally speaking, Homo sapiens are an exception to the Organismus automaton hypothesis. That is, human animals are more-orless good shoppers, effective planners, and careful strategists in office politics—i.e., they are not oblivious to incentives. But does this mean that the upper primates are nonrational as supposed by the Organismus automaton hypothesis? If they are also rational, does this mean that the other primates are nonrational? If all primates are rational, why stop there? Why not extend the blanket of rationality to other mammals, vertebrates, animals, and eventually all organisms? Let us suppose that a biologist takes an anti-naturalist philosophical stance, viz., restricts rationality to humans or to the species with which he or she is most familiar with only (Khalil 1996). If so, the biologist would group all other species under a peculiar category, “non-familiar species,” and assume that they are nonrational, i.e., that the members of this category behave according to genetic programs. After all, the biologist would think: “An ant just cannot be as rational as the organisms I am familiar with” (which would typically mean humans). However, if one adopts such an anti-naturalist stance, it is a daunting task, as this article shows, to explain the origin of rationality for the organisms in this peculiar category. Biological Theory 5(1) 2010

Let us suppose that a biologist takes the opposite stand, i.e., adopts the naturalist philosophical position that either all organisms are rational or none are. Let us focus on the “none are rational” position. Our biologist would postulate that all organisms act according to genetic programs: signals (what economists call “incentives”) stimulate a gene to turn on or off, and correspondingly entice a particular phenotypic behavior. Some of these supposed programmed phenotypic behaviors, our biologist would highlight, are imperfect. For instance, humans would eat non-healthy food and engage in risky behavior. Thus, humans are not rational. However, it must be noted that, contra the “none are rational” position, rationality cannot—as natural selection cannot—be refuted by pointing to imperfections. Every action is costly, and agents must make compromises. Some agents, such as students, might rationally choose a non-healthy diet because they want to spend their scarce resources on something else. The idea of tradeoff is not unfamiliar to biologists: They understand that the design of organisms cannot be perfect, as every design involves a tradeoff between competing functions. Our “none are rational” biologist would counter that organisms cannot be rational because their decisions often involve systematic errors. He or she would argue that organisms act according to patterns of behavior that are too routinized, i.e., cannot change in response to changes in signals (incentives). Many social scientists use the same observation (viz., that humans are creatures of habit or routines) to support different versions of the hypothesis that humans cannot be rational (Khalil 2007). According to them, humans also make systematic errors—such as when they are deceived by a new advertisement gimmick on how to save money. Rationality, though, is not equivalent to error-free decision-making. As shown below, when we discuss bounded rationality and biases as uncovered by behavioral economists and psychologists, error-making can be explained on the basis of rationality: Given that search is costly, humans use rules-of-thumb (routines) that have, on some occasions, negative by-products, such as falling victim to gimmicks and failure to react to change of signals. This does not mean that humans do not commit irrational acts. These include “recklessness” and “addiction,” such as self-destructive behavior that may lead to suicide. But it is often the case that human organisms recognize such acts of irrationality and seek to rectify them. Such desire to rectify irrational acts cannot be explained if one adopts the position that the irrational acts are the products of genetic programming. If they are the products of genetic programming, the irrational agents should be as happy as the carriers of the rational lineage. The fact that irrational humans seek help implies that they are, at a deeper level, rational. There is another challenge that faces our biologist who denies the existence of rationality. Such a position faces an 57

Are Plants Rational?

empirical anomaly, namely, organisms have a variety of responses that cannot be explained by “turned on” and “turned off” genetic programs. There is simply too little time available for genetic instructions when an organism is faced with a new environment. The organism has to improvise. It is simply impossible for all possible environmental scenarios to be coded in the DNA (DeWitt and Scheiner 2004). The organism has to learn about predators and toxic substances and, hence, life of experiences shapes development and explains behavioral plasticity as emphasized by advocates of the EvoDevo research program (Matsuda 1987; West-Eberhard 1989; Raff 1996; M¨uller and Newman 2003; Hall et al. 2004). However, even when we can explain a phenotypic response to an incentive by tracing it to the “turning on” of genes, it is not straightforward how to model this genetic “turning on.” Namely, it does not necessarily mean that the product (the phenotypic behavior), is the outcome of an exogenous genotypic change, as is argued in the Organismus automaton hypothesis. Rather, the genotypic “turning on” can be modeled as an endogenous response—i.e., the decision of the organism would be based mainly on a cost-benefit calculation and the genetic “turning on” simply assists the organism in carrying out the already decided, optimum action. If the organism decided to take a different action in response to a change of incentives, it would “turn on” alternative genes for assistance. The genetic assistance, in this endogenous genetic “turning on” view, merely saves energy and time: It helps the organism to use ready-made repertoires, corresponding to the changing environment, in order to avoid having to reinvent the wheel each time. Such an endogenous view of genetic “turning on” certainly needs to be explored, but this cannot be done here. I only mention it to show that genetic “turning on” does not necessarily entail the Organismus automaton hypothesis. It suffices to add that the plausible endogenous view of genetic “turning on” is perfectly analogous to a particular rational view of action vis-`a-vis cultural repertoires. Namely, human agents first decide on what is the rational action, i.e., undertake a forward-looking calculation of cost and benefit. They then search into their past looking for a suitable cultural trait (equivalent to genetic “turning on”) that helps them to avoid reinventing the wheel in dealing with the given problem. So the claim that genetic “turning on” is endogenous is analogous to the claim that cultural norms are endogenous, contrary to the claim of some institutional economists (e.g., North 2005; Khalil 2008b, 2008c). Cultural repertoires or motifs are rather revived from the past only insofar as they advance forwardlooking interest. On the other hand, the thesis that cultural repertoires, routines, or genetic “turning on” are exogenous—the Organismus automaton hypothesis—begs the question: Why did a genetic “turning on” take place in response to one incentive but not in 58

response to others? For instance, if a predator spots some prey, the gene for “pursuit” would not “turn on” if the prey is too far. Why is there no response in this case? The neo-Darwinian answer is that the signal (i.e., what economists call an incentive) would be too weak to stimulate a response. On the other hand, if the prey is too close, the predator may again not pursue the prey if it senses that the prey is bigger than normal and, hence, cannot be subdued easily. The neo-Darwinian answer, again, is that the “incentive” is not strong enough. So, ultimately, the predator is making a calculation of costs and benefits. This is precisely what economists call rationality when they study humans. Why should we then call it “as-if” rational decision-making when we study nonhuman organisms? Why not use the same framework for all organisms—and hence avoid the arbitrary human/nonhuman divide (Khalil 1996)? As Sober (1996, 1998) shows, but with some qualification (Khalil 2008a), both hypotheses—the Organismus oeconomicus and Organismus automaton—predict the same optimum allocation of resources (see also Maynard Smith 1978). Why would we want, then, to adopt the roundabout Organismus automaton hypothesis? Why would we want to violate Ockham’s razor with a redundant theoretical route, i.e., claim that minute rational calculations are only “as-if” rational and attribute them to a genetic program selected by natural selection? In addition, we would have to undertake a somewhat unwarranted theoretical route to avoid further violations of Ockham’s razor and postulate that humans, given that they are not special, also do not behave rationally. In short, the Organismus automaton hypothesis makes the theory unnecessarily roundabout. The only advantage of such roundaboutness is metaphysical: The theory allows one to avoid accounting for rationality in nature, which was the main reason behind the triumph of Darwinism (Dennett 1995; Hodgson 2002). On the other hand, some biologists have already moved somewhat toward the Organismus oeconomicus hypothesis (Schoener 1971; Ghiselin 1974, 1992; McFarland 1977; Griffin 1984, 1991, 1992; Smith and Winterhalder 1992; Detrain et al. 1999; Cassill 2003; Franks et al. 2003; Vermeij 2004; Hurley and Nudds 2006; Raby et al. 2007). Many of them have moved on their own, without being prompted by economists who have been advancing the bioeconomics agenda (Tullock 1971, 1994; Landa 1986; Landa and Wallis 1988; see Khalil and Marciano 2010). These biologists were not responding to Ockham’s razor argument explicitly. Rather, they were responding to the burgeoning evidence of behavioral plasticity, documented even in plants and brainless organisms, which undermines the old-fashioned Organismus automaton hypothesis. The Organismus automaton hypothesis basically advances the “all-in-the-genes” view. On this view, there is a sufficiently high one-to-one correspondence between the genotype and the phenotype. Although natural selection theory works as long as there is some non-zero correspondence between genotype Biological Theory 5(1) 2010

Elias L. Khalil

and phenotype, the assumption of high one-to-one correspondence is necessary by definition for our biologist who insists on tracing every minute difference in phenotype to genotypic instructions, assuming that such instructions are exogenous. The Organismus automaton hypothesis, which supposes a one-to-one correspondence of genotype and phenotype, has been debated extensively under the rubric of reductionism and holism throughout the history of biology (Mayr 1982; Khalil 1993). The consensus is that it is simply impossible to trace every minute variation in the phenotype to heritable genetic material. After all, behavioral plasticity is too ubiquitous to dismiss—it is found effectively in all taxa (Raff 1996; Schlichting and Pigliucci 1998; Gould 2002; West-Eberhard 2003; DeWitt and Scheiner 2004; Ananthakrishnan and Whitman 2005). Furthermore, such phenotypic ubiquity is not limited to humans or their favored “smart” animals. The claim that humans, or their favored organisms, are the only ones to be free from the shackles of the genes is therefore simply the product of species-centric social science (Khalil 1996, 1997a). Behavioral plasticity amounts to the responsiveness of the organism (in terms of organ development or behavior) to incentives, which is the focus of developmental biology. This position can be pushed to underemphasize the role of genes (Levins and Lewontin 1985). But one does not have to go to that extreme. The basic finding is that the organism is not an automaton in the face of the environment (Lewontin 1984). Rather, the organism varies its decisions in light of the change of incentives. So, this article assumes the Organismus oeconomicus hypothesis—viz., it is best to model actions as the result of calculation of costs and benefits. This means that inherited patterns of behavior and bodily capital (organs) can be taken as given constraints, and are thus somewhat similar to environmental constraints in rational decision-making. So, within the inherited and environmental constraints actions can vary. The same genotype could therefore give rise to a variety of actions. Organisms make decisions in light of incentives, which is the core meaning of rationality as explained below. Does Bounded Optimization Challenge Optimization? For rationality theory, optimization is the maximization of the objective function, viz., the choice of technology (lineage) and behavior that attains the highest profits or utility. Evolutionary biologists call such optimization the maximization of fitness. For neo-Darwinian biologists such maximization is not attained via rational deliberation but via selection. Selection optimization amounts to editing out the less fit lineages and behavior, which means retaining the organisms that can attain the highest fit offspring. As such, fitness means the organism best adapted to its niche becomes dominant in the population. Economists call such optimization market selection, Biological Theory 5(1) 2010

Figure 1. Optimization: unbounded and bounded.

where low-profit firms are deleted out from the market in favor of high-profit firms. Either in nature or the market, lowperforming units cannot persist in the long run along with the higher performers. We then have two concepts of optimization (Khalil 2008a): rationality optimization and selection optimization. As shown in Figure 1, both concepts can be expressed in two different flavors: as bounded and unbounded optimization. There are thus four possible combinations. In unbounded selection optimization (e.g., Dawkins 1976, 1982), a population becomes fully adapted to its environment when the fittest lineage, in comparison to other lineages in the population, becomes dominant. The optimization is “unbounded” in the sense that there are no enduring obstructions, transaction costs, or information imperfections that prevent the population from reaching full adaptation, i.e., becoming fully dominated by the fittest lineage. Likewise, with the absence of obstacles, market competition ensures the dominance of the best technology, so that there is a uniform rate of profit across the economy. In unbounded rationality optimization, on the other hand, agents make the best decisions given the information available and the cost of further search. The optimization is “unbounded” in the sense that there are no obstacles that prevent the agent from calculating the optimum allocation. Both positions are disputed, across disciplines, by the bounded view. Concerning bounded selection optimization, “soft” Darwinians argue that evolution is not only the outcome of natural selection but can also be the outcome of genetic drift, environmental accidents, mutational shocks, intrusion of migrants, and so on. But these objections only strengthen the unbounded view, because they amount to changes in data, which can be included in the calculus of optimization. A more challenging critique of unbounded selection optimization is that many traits are the products of “structure” or “phylogenetic intertia.” This is similar to the critique voiced by some institutional economists who argue that decisions or behavior is greatly constrained by cultural norms, historical inertia, or path-dependent institutions (see Khalil 2008b). There is no question that pigs, in the face of drastic change of environmental conditions, cannot adjust and develop wings that would enable them to fly or to invent other drastic technologies to save 59

Are Plants Rational?

themselves from extinction. Organisms are almost locked in historic inertia, i.e., path- or phylogenetic-dependent paths of change. But such path-dependency cannot be used, pace Fodor (2007), to argue against unbounded selection optimization. Fodor’s argument, which repeats the argument of Gould and Lewontin (1979) about “spandrels,” assumes incorrectly, as Orzack and Sober (2001a, 2001b) note, that the phenomenon of phylogenetic inertia is the null hypothesis of unbounded selection optimization theory. That is, it is not the case that if path-dependency gains credence, it must be at the expense of unbounded optimization. While historical constraints—called “structures,” “vested interests,” or “institutions”—constrain the range of choices, they cannot be dismissed as outside optimization. Such structures were, at some historical time, also the products of optimization. That is, these structures were not always “immovable mountains.” They were malleable options or paths that were not chosen for a good reason. To see the point, optimization entails choices, which, if repeated, become hardened in structures that almost cannot—but theoretically can—be revised or questioned in light of changing environment. Even if structures or past constraints matter, natural selection or optimization is not weakened because the core of optimization is that any change or mutation occurs blindly— i.e., without regard to ex post benefit. The range of blind mutations is restricted, for sure. But they are not the products of innovative imagination in light of learning and so on. As such, the recognition of the structures does not weaken unbounded selection optimization. It can be argued along the economist’s practice that we can treat these structures as part of the constraint set, not different from environmental constraints. And then one can proceed to show how a new trait or a new behavior, although limited by such structures, is still the subject of selection optimization or, in case one is operating along rationality theory, rationality optimization. The fact that there is historical stickiness does not reduce the power of the theory of natural selection. A further critique of the unbounded selection optimization view is that one can hardly find a population with a homogeneous lineage. It is a fact that any trait in a population is never fully dominated by a single allele—as if a population can never reach equilibrium or attain full adaptation. But one can show, at least theoretically, that some variance of alleles is actually predicted by the unbounded selection optimization view. In this light, the diversity of alleles acts as a “hedge fund,” which, although maybe an “ill-investment” given the ex post environment, is in fact ex ante optimal given the uncertainties of environmental conditions. It is always good for a population to preserve such a hedge fund (diversity) so that in case of, say, a dropping of temperature or invasion by a disease, some individuals will not be wiped out. Such a statement is framed in a structural-functionalist (holistic) thesis, but it can easily be 60

operationalized into a non-holistic or individualist thesis: The environment is actually mired with uncertainty, where each “state of the world” has a different probability of coming into being. So it is selection-optimal in the unbounded sense to sustain a diversity of alleles. In this manner, each lineage is “fitted” to its temporally spread ecosystem state of the world or, in other words, to its geographically spread eco-niche. So, the diversity of alleles is actually an equalizing equilibrium, i.e., expressing efficient or unbounded selection optimization (Roughgarden 2004).6 The notion of “equalizing equilibrium” is identical to the notion of “equalizing differences” in economics (Rosen 1981). Examples of equalizing differences include the diverse prices of homes indicating their distance from the center of town, the diverse wages of employees indicating their differences in skill, and the diverse rents of land indicating location or fertility. However, it is also possible under bounded selection to interpret the persistence of diversity as something suboptimal, i.e., indicative of “flabby” (bounded) selection optimization forces.7 Likewise, sagging market selection optimization (competition) can allow firms within the same industry to coexist for a long time with various types of technologies (and hence productivities), as Luria (1996) documents. (On the other hand, as argued above, such a diversity of technologies can be conceived, instead, as optimal in the face of fluctuating market conditions.) But this should not concern us here. The point is that bounded market selection optimization is possible, at least theoretically. Such boundedness could feasibly arise as a result of increasing returns or high transaction costs (friction). It could also arise as a result of path dependency, in which accidental events, such as the technology that is first introduced, matter because first-comers can carve a path of change that makes them withstand competition from newcomers that turn out to have fitter technologies (Dosi et al. 1988; Arthur 1989). Considering bounded rationality optimization, neoclassical economists hijacked Simon’s (1957) notion of “bounded rationality” or “satisficing” to argue the following: The mental capacity of the agent is bounded by its intelligence, memory, and other cognitive abilities. Such a bounded mental capacity entails particular transaction costs to search for information, to digest the optimum information to come up with optimum beliefs, and to use the optimum beliefs effectively to come up with the optimum action. It is very costly for the agent to come up with the optimum decision. So, agents adopt institutions, heuristics, or rules that work most of the time, but occasionally cause some errors. Such heuristics, on average, allow agents to do better, given their limited capacity, than if they were to optimize unboundedly at each instant. If agents are going to optimize at each instant as supposed by unbounded rationality, they might avoid some pitfalls, but it would be enormously expensive. So, they adopt heuristics that, Biological Theory 5(1) 2010

Elias L. Khalil

as a by-product, make behavior deviate from the predictions of unbounded rationality. These deviations have been uncovered by behavioral economics and behavioral decision research (Camerer et al. 2004; Fudenberg 2006; Pesendorfer 2006). Such deviations have also been uncovered by ethologists and behavioral ecologists (e.g., Marsh and Kacelnik 2002). Even research in nonhuman animal behavior has shown that such agents also exhibit bounded rationality (e.g., Schuck-Paim et al. 2004). The point is rather how to interpret such deviations. As already mentioned, and following the argument of Gigerenzer (2005, 2006; see also Goldstein et al. 2001), such shortcomings or errors in logical reasoning are occasional failures of rules of thumb or “heuristics” that are, on average, effective. It pays on average to have heuristics, and the price of such shortcuts is that one occasionally makes mistakes. If so, we should not get rid of the “quirks” of cognition if it also means sacrificing the heuristics that are effective most of the time. Even if we “improve” computation and minimize errors in logical reasoning, we are not improving rationality. We are rather employing a different technology that allows us to make different decisions. Given that such technology is costly, one adopts it only when the payoff is greater than the cost. Put differently, if we take a broader account of all relevant information, given the expected costs and benefits of such information, to achieve the best solution, sometimes it is better to settle for the second-best. For instance, if a person calculates each time whether to take the stairs or the elevator, he or she would, on average, lose more resources than if he or she would settle on the second-best and adopt the habit of taking one means.8 So, even when the agent acts according to the second-best, i.e., acts with bounded rationality, he or she is using unbounded rationality. So, debates, often voiced, of whether a behavior expresses bounded or unbounded rationality is ultimately an empirical question. The disagreement often misses the implied agreement, viz., bounded optimization does not actually undermine the notion of optimization per se. When one advances the thesis that rationality optimization is bounded, one has to assume unbounded rationality optimization in order to account for bounded rationality optimization. These issues are replicated with debates of whether evolution can generate false beliefs. For instance, Stephens (2001) calls such inaccurate beliefs “better-safe-than-sorry arguments.” Such beliefs prompt agents to adopt risk-averse behavior when it appears that accurate beliefs would lead agents, at least in some occasions, to amass greater wealth or progeny. However, accurate beliefs that prompt agents to be risk-neutral may actually, on average, decrease wealth or progeny if the greater accuracy entails costs unjustified by marginal benefits. Here, again, when theorists want to show the limits of optimization, they have to use optimization. Likewise, Stich (1985, 1990) argues that suboptimal beliefs can be adaptive: In Biological Theory 5(1) 2010

evolution, what matters is not finding the truth, but finding beliefs that afford greater chances of survival and, hence, leaving greater offspring on average than other beliefs. If agents act very cautiously and develop inaccurate beliefs that engender risk-aversion that maximizes chances of survival, they can be favored by natural selection over agents who attend to the facts and develop more accurate beliefs that put them at greater risk. Again, such inaccurate beliefs do not indicate sub-optimality. They rather indicate the bounded nature of optimality. Does Procedural/Modular Rationality Challenge Organismus oeconomicus? Simon spent the latter part of his career insisting that his notion of bounded rationality had been hijacked by the neoclassical approach (Callebaut 1998, 2007; Mousavi and Garrison 2003). To salvage his notion, Simon (1977) devised an alternative term, “procedural rationality,” to stress that he is challenging bounded rationality or, specifically, how bounded rationality has been incorporated and diluted by mainstream, neoclassical economics. Simon is correct that bounded rationality, as incorporated in neoclassical economics, is not a challenge to Organismus oeconomicus. Many psychologists extend Simon’s insight that rationality is a procedure. They argue that the issue is not whether preferences are bounded or not in the neoclassical sense, but rather whether preferences are vague and, hence, partially determined by experience or procedure (e.g., Gigerenzer and Selten 2001; Rieskamp et al. 2006). Still, is Simon’s procedural rationality, as distinct from bounded rationality a` la neoclassical theory, a challenge to Organismus oeconomicus? Procedural rationality portrays humans as guided by routines. Humans are not decision makers who are trying to optimize. They are rather, at first approximation, creatures of habit. Simon’s theory can be generalized to all organisms and, hence, called here “Organismus habitus” hypothesis. This hypothesis differs from the Organismus automaton hypothesis in one important regard—it allows for cognitive processes to take place, which prompt the organism to change and adopt new routines; whereas in the Organismus automaton hypothesis change only takes place at the level of the population. Despite this difference, both hypotheses share one common element: Organisms do not respond to incentives, i.e., they do not act rationally. To elaborate, for Simon, agents are simply so enmeshed within their particular environment or particular experience that their decision-making is, rather, experiential or procedural. The procedural rationality notion ultimately means that agents would continue a particular routine, procedure, or habit of production as long as there is no sudden shock that upsets the status quo (Khalil 2007). Only with a shock, the agent would search outside the box for an alternative routine or procedure. 61

Are Plants Rational?

The default position for agents is to accept whatever is the status quo. Agents simply cannot stand outside environment and be involved in the kind of rational calculation about rules and heuristics, as assumed in neoclassical economics. The Organismus habitus hypothesis coheres with the modularity approach generally (e.g., Callebaut and RasskinGutman 2005), and with a particular strand of evolutionary psychology (Cosmides and Tooby 1994; Sperber 1994; Samuels 1998, 2000). For instance, Cosmides and Tooby argue that rationality is not a “content-independent” faculty. That is, rationality cannot be defined independently of the particular experience and, hence, it consists of whatever habit or behavior the agent finds feasible, which is Simon’s notion of “satisficing.” Paul Glimcher et al. (2005) likewise argue that we should not start with axiomatic rationality, but rather with evolutionary history that takes particular brain circuits, physiological structure, and the relevant environment seriously. The problem with Cosmides, Glimcher, and other advocates of the evolutionary approach is that they confuse rationality with what rationality uses, viz., eyesight, brain circuits, and other abilities. It is similar to the earlier discussed confusion of rationality with the intelligence ability. This recent upsurge of evolutionary explanation, which stresses historical experience, is not new. It was advanced in 1922 by Dewey (1957) and even earlier by the founders of American Pragmatism such as James (1893; see Khalil 2003, 2007; Mousavi and Garrison 2003). The pragmatic approach greatly influenced Veblen’s (1898) critique of the rationality approach of economics. The pragmatic/procedural approach in economics, or what is sometimes called “institutional economics,” emphasizes that what appears as rational action is actually the result of experience that has engendered useful routines on how to interact with the environment. There is no such thing as global rationality; there is only habituation, where the organism and the environment are united in what Dewey and Bentley (1949) called “transaction” or “union” of the known (environment) and knowing (the actively involved organism). So agents behave in particular ways because such ways are how they behaved in the past and, hence, cannot express “content-independent” rationality. The Organismus habitus hypothesis that there is no content-independent rationality faces one major anomaly. It cannot distinguish scientific theories from superstitions, on one hand, and superstitions from religious/spiritual convictions, on the other. All three beliefs—scientific, superstitious, and spiritual—are matters of habits of the mind or routines. But insofar as, at least in common sense, we can distinguish these beliefs, we must be applying rational deliberation (Khalil 2010). Agents must be selecting via Bayesian updating proper beliefs about the environment, as they try to avoid unsupported beliefs. So, the modular, evolutionary psychology, or pragmatic view solves the problem of the origin of behavioral 62

plasticity by denying, at first approximation, that agents update their beliefs. The modularity/procedural approach ultimately portrays organisms as the slaves of programs or modules of behavior. However, even if we go along with this procedural view, organisms still would have to make a decision in light of an environmental shock; organisms still have to search for a new routine in such situations. It might be as well that the procedural view can describe the persistence of procedures once they exist. It cannot ultimately explain why organisms are capable of changing them in light of drastic changes in their incentives. We cannot, at first approximation, entertain the Organismus habitus hypothesis simply because it denies that organisms initiate action and react to incentives. Further, the Organismus habitus hypothesis leads to the following problem: If action is simply a habit, it cannot draw a distinction between technological learning and scientific learning (Khalil 2007). In technological learning, the agent learns how to manipulate the environment in order to attain a goal. In scientific learning, there is no goal or benefit at hand: The agent only wants to know how the environment works. Dewey argues that the two types of learning are “habit.” If so, agents adopt a technology just because of an accident, not because of figuring out a better way to attain the goal. As such, the Simon/Dewey line cannot afford an endogenous theory of evolution of technology. Conclusion This article advances the Organismus oeconomicus hypothesis against the Organismus automaton. The advanced hypothesis provides a unified organizing view of the life sciences that identifies decision-making, rather than natural selection, as central. It argues that behavioral plasticity, which is not disputed to be characteristic of all organisms including plants, is nothing but an indication that organisms are rational. Biologists have been documenting how organisms change their shape, including behavior, in response to incentives. And such behavioral plasticity does not require brains. So, rational decision-making is simply characteristic of being a living entity. This article argues that the Organismus oeconomicus hypothesis is not undermined by the rising evidence that rationality is bounded: Bounded rationality presumes the operation of unbounded rationality. It is also not undermined by the fact that much of behavior is procedural in the sense of being routinized: Organisms still improvise and do exhibit behavioral plasticity in the face of new environmental constraints. The Organismus oeconomicus hypothesis highlights that rationality does not require complex mental processes—not to mention a neural system to start with. Rational deliberation does not involve conscious deliberation or high intelligence. For instance, when the government started to impose seat belt Biological Theory 5(1) 2010

Elias L. Khalil

laws, economists predicted that agents would drive less cautiously, resulting in more accidents but less (thanks to the seat belt) fatalities per accident (Peltzman 1975). The tradeoff between safety and time gained by greater speed was never conscious. Also, agents who have insurance tend to adjust to the new incentives and adopt less careful habits with regard to locking doors, making sure that the stove is off, and so on. Likewise, Brian Goff et al. (1997) show that baseball players in the American League adjusted their behavior in light of the 1973 change of the designated hitter rule. The aim of this article is to advocate the centrality of rationality in understanding behavior. More biologists are recognizing the importance of the tools of rationality in shedding light on the widely admitted phenomena called behavioral plasticity. The plastic behavior of organisms, including plants, cannot be “hard-wired.” It is simply impossible to replicate all possible environmental constraints and encode them into the DNA. The organism constantly learns and improvises. Even simple organisms adjust their behavior in light of new environmental constraints. The organism does not need a specialized neural system, such as the brain, to process information. It can process information at the level of the non-specialized cells and cells can communicate with each other. Cellular, organismic, and behavioral biologists can teach us about the details of such communications and its underpinning rationality. Therefore, it would be useful to openly acknowledge Organismus oeconomicus.

are concerned with another kind of rationality because they focus on the rationality of action. As advocated here, we only have one kind of rationality, viz., getting the best product given the constraints.

Acknowledgments

American Psychological Association (1995) Intelligence: Knowns and Unknowns. Report of a task force of the board of scientific affairs. http://www .lrainc.com/swtaboo/taboos/apa 01.html Ananthakrishnan TN, Whitman D, eds (2005) Insect Phenotypic Plasticity: Diversity of Responses. Enfield, NH: Science Publishers. Arthur WB (1989) Competing technologies, increasing returns, and lock-in by historical events. Economic Journal 99: 116–131. Baldwin JM (1896) A new factor in evolution. American Naturalist 30: 441– 451. Becker GS (1976) The Economic Approach to Human Behavior. Chicago: University of Chicago Press. Bloom AJ, Chapin FS III, Mooney HA (1985) Resource limitation in plants: An economic analogy. Annual Review of Ecology and Systematics 16: 363–392. Brandon RN, Burian RM, eds (1984) Genes, Organisms, Populations: Controversies Over the Units of Selection. Cambridge, MA: MIT Press. Callebaut W (1998) Self-organization and optimization: Conflicting or complementary approaches? In: Evolutionary Systems: Biological and Epistemological Perspectives on Selection and Self-Organization (Van de Vijver G, Salthe SN, Delpos M, eds), 79–100. Dordrecht: Kluwer. Callebaut W (2007) Simon’s silent revolution. Biological Theory 2: 76–86. Callebaut W, Rasskin-Gutman D, eds (2005) Modularity: Understanding the Development and Evolution of Natural Complex Systems. Cambridge, MA: MIT Press. Camerer C, Loewenstein G, Rabin M, eds (2004) Advances in Behavioral Economics. Princeton, NJ: Princeton University Press. Cassill D (2003) Skew selection: Nature favors a trickle-down distribution of resources in ants. Journal of Bioeconomics 5: 83–96.

This article was supported by the Konrad Lorenz Institute for Evolution and Cognition Research (Altenberg, Austria), the Max Planck Institute of Economics (Jena, Germany), and Monash University’s Faculty Research Grant Scheme, 2006. An older version received comments from Richard Posner, Ulrich Witt, Gerd M¨uller, Werner Callebaut, Steven Orzack, Steve Abedon, Jack Vromen, Brian Charlesworth, Gordon Tullock, Timothy Crippen, Michael Ghiselin, Howard Margolis, Robert Axelrod, Richard Levins, Richard Nelson, Joseph Lopreato, R. Preston McAfee, J.S. Metcalfe, Peter Taylor, Elliott Sober, Stanley Salthe, Casey Mulligan, Franz Weissing, and seminar participants at the Konrad Lorenz Institute. The current version received comments from Brian Skyrms, Jack Vromen, Michael Ghiselin, Yew-Kwang Ng, Richard Posner, Paul Griffiths, Ulrich Witt, Avi Waksberg, Martin Burd, Ellen Larsen, Deby Cassill, anonymous referees, and seminar participants at Monash University. The article benefited greatly from the assistance of Michael Dunstan. The usual caveat applies.

Notes 1. As argued elsewhere (Khalil 2008a), rationality in fact cannot be a trait and hence cannot be explained as the product of natural selection. 2. “Intelligence: Knowns and Unknowns,” report of a task force of the board of scientific affairs, American Psychological Association, released August 7, 1995. 3. The distinction among the three layers of rationality should not entail that there are different kinds of rationality, as Kacelnik (2006) advocates. For Kacelnik, philosophers and psychologists are concerned with one kind of rationality because they focus on the rationality of beliefs, while economists Biological Theory 5(1) 2010

4. One can define “irrationality” as the failure to act according to one’s best interest, which is recognized in the literature on weakness of will. There is an attempt to argue that weakness of will is actually rational. But there is a growing literature, aided by the rise of behavioral economics and behavioral decision theory, recognizing that weakness of will is an aberration from rationality. 5. New findings suggest some exceptions to the Weissman barrier. For instance, Steele et al. (1998) have found that the immune system response to new viral invasions, which involves somatic genetic mutations, can be passed on to the germline, i.e., to progeny. If true, this would be one instance of the Lamarckian thesis of inheritance of acquired characteristics. 6. Roughgarden (2004) celebrates the diversity of populations, which she calls the “rainbow.” But she seems to conflate the diversity of a trait, such as the coexistence of beaks with differing sharpness, with the number of a gender in a population. The latter is not about the diversity of some trait, but rather about the complexity of specialization of the process of reproduction. 7. Such a possible “inefficient” selection optimization (i.e., sub-optimal equilibrium) differs from the notion of “inefficient” rationality optimization (i.e., irrationality) along the same line that differentiates these two types of optimizations, as discussed here. 8. In economics, some authors have also pointed out situations of market stickiness that resemble the stickiness that hinders evolutionary selection optimization as pointed out by Gould (1977, 2002). As a result of non-ergodic feedbacks concerning innovation, market niches, and other variables, the potentially more productive firms do not even appear (Khalil 2000).

References

63

Are Plants Rational?

Cosmides L, Tooby J (1994) Beyond intuition and instinct blindness: Toward an evolutionary rigorous cognitive science. Cognition 50: 41–77. Darwin C ([1880] 1989) The Power of Movement in Plants. New York: New York University Press. Darwin C ([1871] 1998) The Descent of Man. Amherst, NY: Prometheus Books. Darwin C ([1859] 2006) The Origin of Species. Variorum edition (Peckham M, ed). Philadelphia: University of Pennsylvania Press. Dawkins R (1976) The Selfish Gene. New York: Oxford University Press. Dawkins R (1982) The Extended Phenotype: The Long Reach of the Gene. San Francisco: Freeman. De Jong G (2005) Evolution of phenotypic plasticity: Patterns of plasticity and the emergence of ecotypes. New Phytologist 166: 101–117. Dennett DC (1995) Darwin’s Dangerous Idea: Evolution and the Meanings of Life. New York: Simon and Schuster. Detrain C, Deneubourg JL, Pasteels JM, eds (1999) Information Processing in Social Insects. Basel, Switzerland: Birkh¨auser. Dewey J ([1922] 1957) Human Nature and Conduct: An Introduction to Social Psychology. New York: Modern Library. Dewey J, Bentley AF (1949) Knowing and the Known. Boston: Beacon Press. Dewey J, Tufts JH (1908) Ethics. New York: Holt. DeWitt TJ, Scheiner SM, eds (2004) Phenotypic Plasticity: Functional and Conceptual Approaches. New York: Oxford University Press. Dosi G, Freeman C, Nelson RR, Silverberg S, Soete S, eds (1988) Technical Change and Economic Theory. London: Pinter. Dudley SA (2004) The functional ecology of phenotypic plasticity in plants. In: Phenotypic Plasticity: Functional and Conceptual Approaches (DeWitt TJ, Scheiner SM, eds), 151–172. New York: Oxford University Press. Eichinger L, Pachebat JA, Gl¨ockner G, Rajandream MA, Sucgang R, Berriman M, Song J, Olsen R, Szafranski K, Xu Q, Tunggal B, Kummerfeld S, Madera M, Konfortov BA, Rivero F, Bankier AT, Lehmann R, Hamlin N, Davies R, Gaudet P, Fey P, Pilcher K, Chen G, Saunders D, Sodergren E, Davis P, Kerhornou A, Nie X, Hall N, Anjard C, Hemphill L, Bason N, Farbrother P, Desany B, Just E, Morio T, Rost R, Churcher C, Cooper J, Haydock S, van Driessche N, Cronin A, Goodhead I, Muzny D, Mourier T, Pain A, Lu M, Harper D, Lindsay R, Hauser H, James K, Quiles M, Madan Babu M, Saito T, Buchrieser C, Wardroper A, Felder M, Thangavelu M, Johnson D, Knights A, Loulseged H, Mungall K, Oliver K, Price C, Quail MA, Urushihara H, Hernandez J, Rabbinowitsch E, Steffen D, Sanders M, Ma J, Kohara Y, Sharp S, Simmonds M, Spiegler S, Tivey A, Sugano S, White B, Walker D, Woodward J, Winckler T, Tanaka Y, Shaulsky G, Schleicher M, Weinstock G, Rosenthal A, Cox EC, Chisholm RL, Gibbs R, Loomis WF, Platzer M, Kay RR, Williams J, Dear PH, Noegel AA, Barrell B, Kuspa A (2005) The genome of the social amoeba Dictyostelium discoideum. Nature 435: 43–57. Fodor J (2007) Why pigs don’t have wings. London Review of Books 29(20): 19–22. Franks NR, Mallon EB, Bray HE, Hamilton MJ, Mischler TC (2003) Strategies for choosing between alternatives with different attributes: Exemplified by house-hunting ants. Animal Behaviour 65: 215–223. Fudenberg D (2006) Advancing beyond advances in behavioral economics. Journal of Economic Literature 44: 694–711. Ghiselin MT (1974) The Economy of Nature and the Evolution of Sex. Berkeley and Los Angeles: University of California Press. Ghiselin MT (1992) Biology, economics, and bioeconomics. In: Universal Economics: Assessing the Achievements of the Economic Approach (Radnitzky G, ed), 55–104. New York: ICUS Publications. Gigerenzer G (2005) I think, therefore I err. Social Research 72: 195–218. Gigerenzer G (2006) Bounded and rational. In: Contemporary Debates in Cognitive Science (Stainton RJ, ed), 115–133. Oxford, UK: Blackwell. Gigerenzer G, Selten R, eds (2001) Bounded Rationality: The Adaptive Toolbox. Cambridge, MA: MIT Press. 64

Glimcher PW, Dorris MC, Bayer HM (2005) Physiological utility theory and the neuroeconomics of choice. Games and Economic Behavior 52: 213– 256. Goff BL, Shughart WW II, Tollison RD (1997) Batter up! Moral hazard and the effects of the designated hitter rule on hit batsmen. Economic Inquiry 35: 555–561. Goldstein DG, Gigerenzer G, Hogarth RM, Kacelnik A, Kareev A, Klein G, Martignon L, Payne JW, Schlag KH (2001) Group report: Why and when do simple heuristics work? In: Bounded Rationality: The Adaptive Toolbox (Gigerenzer G, Selten R, eds), 173–190. Cambridge, MA: MIT Press. Gould SJ (1977) Ontogeny and Phylogeny. Cambridge, MA: Harvard University Press. Gould SJ (2002) The Structure of Evolutionary Theory. Cambridge, MA: Harvard University Press. Gould SJ, Lewontin RC (1979) The Spandrels of San Marco and the Panglossian paradigm: A critique of the adaptationist program. Proceedings of the Royal Society of London B 205: 581–598. Griffin DR (1984) Animal Thinking. Cambridge, MA: Harvard University Press. Griffin DR (1991) Animal Thinking. Scientific American 265(5): 144. Griffin DR (1992) Animal Minds. Chicago: University of Chicago Press. Griffith TM, Watson MA (2006) Is evolution necessary for range expansion? Manipulating reproductive timing of a weedy annual transplanted beyond its range. American Naturalist 167: 153–164. Grime P (1997) Mechanisms of vegetation response to climate change. In: Past and Future Rapid Environmental Changes: The Spatial and Evolutionary Responses of Terrestrial Biota (Huntley B, Cramer W, Morgan AV, Prentice HC, Allen JRM, eds), 195–204. Berlin, Germany: Springer. Hall BK, Pearson RD, M¨uller GB, eds (2004) Environment, Development, and Evolution: Towards a Synthesis. Cambridge, MA: MIT Press. Herman LM (2006) Intelligence and rational behaviour in the bottlenosed dolphin. In: Rational Animals? (Hurley S, Nudds M, eds), 439–467. Oxford, UK: Oxford University Press. Hodgson GM (2002) Darwinism in economics: From analogy to ontology. Journal of Evolutionary Economics 12: 259–281. Huber H, Stuefer JF (1997) Shade-induced changes in the branching pattern of a stoloniferous herb: Functional response or allometric effect? Oecologia 110: 478–486. Hurley S, Nudds M, eds (2006) Rational Animals? Oxford, UK: Oxford University Press. James W (1893) The Principles of Psychology. New York: Holt. Kacelnik A (2006) Meanings of rationality. In: Rational Animals? (Hurley S, Nudds M, eds), 87–106. Oxford, UK: Oxford University Press. Khalil EL (1993) Neo-classical economics and neo-Darwinism: Clearing the way for historical thinking. In: Economics as Worldly Philosophy (Blackwell R, Chatha J, Nell EJ, eds), 22–72. London: Macmillan. Khalil EL (1996) Economic action, naturalism, and purposefulness. Research in the History of Economic Thought and Methodology 14: 119–140. Khalil EL (1997a) Economics, biology, and naturalism: Three problems concerning the question of individuality. Biology and Philosophy 12: 185– 206. Khalil EL (1997b) Buridan’s ass, uncertainty, risk, and self-competition: A theory of entrepreneurship. Kyklos 50: 147–163. Khalil EL (2000) Survival of the most foolish of fools: The limits of selection theory. Journal of Bioeconomics 2: 203–220. Khalil EL (2003) The context problematic, behavioral economics and the transactional view: An introduction to “John Dewey and Economic Theory.” Journal of Economic Methodology 10: 107–130. Khalil EL (2007) The problem of creativity: Distinguishing technological ´ action and cognitive action. Revue de Philosophie Economique 8: 33–69. Khalil EL (2008a) Why natural selection cannot explain rationality. Ms. Biological Theory 5(1) 2010

Elias L. Khalil

Khalil EL (2008b) The roadblock of culturalist economics: Douglass North and the retreat of economics. http://mpra.ub.uni-muenchen.de/1045/ 1/MPRA paper 1045.pdf Khalil EL (2008c) Moral outrage: A two-moment theory of nationalism, entrepreneurship and other internal motivations. Ms. Khalil EL (2009) Natural selection and rational decision: Two concepts of optimization. Journal of Evolutionary Economics 19: 417–435. Khalil EL (2010) The Bayesian fallacy: Distinguishing internal motivations and religious beliefs from other beliefs. Journal of Economic Behavior and Organization, forthcoming. Doi 10.1016/j.jebo.2010.04.004 Khalil EL, Marciano A (2010) The equivalence of neo-Darwinism and Walrasian equilibrium: In defence of Organismus economicus. Biology and Philosophy 25: 229–248. Doi: 10.1007/s10539-009-9172-2 Kreps DM (1990) A Course in Microeconomic Theory. Princeton, NJ: Princeton University Press. Landa JT (1986) The political economy of swarming in honeybees: Votingwith-the-wings, decision-making costs, and the unanimity rule. Public Choice 51: 25–38. Landa JT, Wallis A (1988) Socio-economic organization of honeybee colonies: A transaction-cost approach. Journal of Social and Biological Structures 11: 353–363. Levins R, Lewontin R (1985) The Dialectical Biologist. Cambridge, MA: Harvard University Press. Lewontin RC (1984) Adaptation. In: Conceptual Issues in Evolutionary Biology: An Anthology (Sober E, ed), 235–251. Cambridge, MA: MIT Press. Llin´as RR (2001) I of the Vortex: From Neurons to Self. Cambridge, MA: MIT Press. Luria D (1996) Why markets tolerate mediocre manufacturing. Challenge 39: 11–16. McFarland DJ (1977) Decision making in animals. Nature 269: 15–21. Marciano A, Khalil EL (2010) Organomics contra neo-Darwinism. Ms. Marsh B, Kacelnik A (2002) Framing effects and risky decisions in starlings. Proceedings of the National Academy of Sciences of the USA 99: 3352– 3355. Matsuda R (1987) Animal Evolution in Changing Environments with Special Reference to Abnormal Metamorphosis. New York: Wiley. Maynard Smith J (1978) Optimization theory in evolution. Annual Review of Ecology and Systematics 9: 31–56. Mayr E (1982) The Growth of Biological Thought: Diversity, Evolution, and Inheritance. Cambridge, MA: Harvard University Press. Mousavi S, Garrison J (2003) Toward a transactional theory of decision making: Creative rationality as functional coordination in context. Journal of Economic Methodology 10: 131–156. M¨uller GB, Newman SA, eds (2003) Origination of Organizational Form: Beyond the Gene in Developmental and Evolutionary Biology. Cambridge, MA: MIT Press. Nelson RR, Winter SG (1982) An Evolutionary Theory of Economic Change. Cambridge, MA: Harvard University Press. Ng Y-K (1996) Complex niches favour rational species. Journal of Theoretical Biology 179: 303–311. Nooteboom B (2000) Learning and Innovation in Organizations and Economies. Oxford, UK: Oxford University Press. North DC (2005) Understanding the Process of Economic Change. Princeton, NJ: Princeton. Orzack SH, Sober E (1994) Optimality models and the test of adaptationism. American Naturalist 143: 361–380. Orzack SH, Sober E (2001a) Introduction. In: Adaptationism and Optimality (Orzack SH, Sober E, eds), 1–23. Cambridge, UK: Cambridge University Press. Orzack SH, Sober E (2001b) Adaptation, phylogenetic inertia, and the method of controlled comparisons. In: Adaptationism and Optimality (Orzack SH, Sober E, eds), 45–63. Cambridge, UK: Cambridge University Press. Biological Theory 5(1) 2010

Pal C (1998) Plasticity, memory and the adaptive landscape of the genotype. Proceedings of the Royal Society of London B 265: 1319–1323. Peltzman P (1975) The effects of automobile safety regulation. Journal of Political Economy 83: 677–725. Pepperberg IM (2006) Intelligence and rationality in parrots. In: Rational Animals? (Hurley S, Nudds M, eds), 469–488. Oxford, UK: Oxford University Press. Pesendorfer W (2006) Behavioral economics comes of age: A review essay on advances in behavioral economics. Journal of Economic Literature 44: 712–721. Plotkin HC (1994) Darwin Machines and the Nature of Knowledge. Cambridge, MA: Harvard University Press. Price TD, Qvarnstr¨om A, Irwin DE (2003) The role of phenotypic plasticity in driving genetic evolution. Proceedings of the Royal Society of London B 270: 1433–1440. Raby CR, Alexis DM, Dickinson A, Clayton NS (2007) Planning for the future by western scrub-jays. Nature 445: 919–921. Raff RA (1996) The Shape of Life: Genes, Development, and the Evolution of Animal Form. Chicago: University of Chicago Press. Ramsey G, Bastian ML, van Schaik C (2007) Animal innovation defined and operationalized. Behavioral and Brain Sciences 30: 393–407. Richards CL, Bossdorf O, Muth NZ, Gurevitch J, Pigliucci M (2006) Jack of all trades, master of some? On the role of phenotypic plasticity in plant invasions. Ecology Letters 9: 981–993. Rieskamp J, Busemeyer JR, Mellers BA (2006) Extending the bounds of rationality: Evidence and theories of preferential choice. Journal of Economic Literature 44: 631–661. Rosen S (1981) The economics of superstars. American Economic Review 71: 845–858. Roughgarden J (2004) Evolution’s Rainbow: Diversity, Gender, and Sexuality in Nature and People. Berkeley and Los Angeles: University of California Press. Rumbaugh DM, Washburn DA (2003) Intelligence of Apes and Other Rational Beings. New Haven, CT: Yale University Press. Samuels R (1998) Evolutionary psychology and the massive modularity hypothesis. British Journal for the Philosophy of Science 49: 575– 602. Samuels R (2000) Massively modular minds: Evolutionary psychology and cognitive architecture. In: Evolution and the Human Mind: Modularity, Language, and Meta-cognition (Carruthers P, Chamberlain A, eds), 13– 46. Cambridge, UK: Cambridge University Press. Samuelson PA (1938) A note on the pure theory of consumers’ behaviour. Economica 5: 61–71. Schaap P (2007) Evolution of size and pattern in the social amoebas. BioEssays 29: 635–644. Schlichting CD, Pigliucci M (1998) Phenotypic Evolution: A Reaction Norm Perspective. Sunderland, MA: Sinauer. Schoener TW (1971) Theory of feeding strategies. Annual Review of Ecology and Systematics 2: 369–404. Schuck-Paim C, Pompilio L, Kacelnik A (2004) State-dependent decisions cause apparent violations of rationality in animal choice. PLoS Biology 2: 2305–2315. Simon HA (1957) Models of Man. New York: Wiley. Simon HA (1977) Models of Discovery. Dordrecht: Reidel. Smith EA, Winterhalder W (1992) Natural selection and decision-making: Some fundamental principles. In: Evolutionary Ecology and Human Behavior (Smith EA, Winterhalder B, eds), 25–60. New York: Aldine de Gruyter. Sober E (1983) Parsimony in systematics: Philosophical issues. Annual Review of Ecology and Systematics 14: 335–357. Sober E (1984) The Nature of Selection: Evolutionary Theory in Philosophical Focus. Chicago: University of Chicago Press. 65

Are Plants Rational?

Sober E (1996) Evolution and optimality: Feathers, bowling balls, and the thesis of adaptationism. Philosophic Exchange 26: 41–57. Sober E (1998) Three differences between deliberation and evolution. In: Modeling Rationality, Morality, and Evolution (Danielson PA, ed), 408– 422. New York: Oxford University Press. Sperber D (1994) The modularity of thought and the epidemiology of representations. In: Mapping the Mind: Domain Specificity in Cognition and Culture (Hirschfeld LA, Gelman SA, eds), 39–67. Cambridge, UK: Cambridge University Press. Steele EJ, Lindley RA, Blanden RV (1998) Lamarck’s Signature: How Retrogenes Are Changing Darwin’s Natural Selection Paradigm. Reading, MA: Perseus Books. Stephens CL (2001) When is it selectively advantageous to have true beliefs? Sandwiching the better-safe-than-sorry argument. Philosophical Studies 105: 161–189. Stich S (1985) Could man be an irrational animal? Synthese 64: 115– 135. Stich S (1990) The Fragmentation of Reason. Cambridge, MA: MIT Press.

66

Tudge C (2005) The Secret Life of Trees. London: Allen Lane. Tullock G (1971) The coal tit as a careful shopper. American Naturalist 105: 77–80. Tullock G (1994) The Economics of Non-Human Societies. Tucson, AZ: Pallas Press. Veblen T (1898) Why is economics not an evolutionary science? Quarterly Journal of Economics 12: 373–397. Vermeij GJ (2004) Nature: An Economic History. Princeton, NJ: Princeton University Press. West-Eberhard MJ (1989) Phenotypic plasticity and the origins of diversity. Annual Review of Ecological Systems 20: 240–278. West-Eberhard MJ (2003) Developmental Plasticity and Evolution. New York: Oxford University Press. Wijesinghe DK, John EA, Hutchings MJ (2005) Does pattern of soil resource heterogeneity determine plant community structure?: An experimental investigation. Journal of Ecology 93: 99–112. Wilson DS, Sober E (1989) Reviving the superorganism. Journal of Theoretical Biology 136: 337–356.

Biological Theory 5(1) 2010