We argue that evolution has supported a combinatorial explosion in the possibilities for action in man as a result of evolutionary pressure from three distinct sources: communication, learning, and mediated action. The latter is intimately related to language use and planning and hinges on the incorporation of plans conveyed by others into once own actions. The individual development of sophisticated planning ability is a complex process of gradual internalization of sociocultural entities that come increasingly to mediate actions. The most significant research question concerns the process of mediation rather than the structure of plans or other mediating entities and therefore propose that mediated actions, rather than plans, occupy center stage in psychology.
1.1 Bridgeman (1992) defines consciousness as developing and executing plans of a certain level of complexity and extension in time. According to this view, consciousness is more than just another module in the system. It is an active process: Many parts of the brain take part in the planning. A plan itself is viewed as an image or a more abstract schema of actions. More formally, we could define this schema as a partially ordered set of actions. Actions are ordered both serially and hierarchically, so that a plan (or a subplan) consists of sequences of (possibly overlapping) subplans. In Bridgeman's view, consciousness deals with making, storing, "unpacking" (articulating), and monitoring plans. One may question whether it is useful to define consciousness in a way that departs markedly from general usage in psychology and philosophy. In this commentary, we will leave this issue aside. Instead, we will argue that the central concept of plan, as defined above, is still not adequate as a unit of analysis in psychological theorizing. Following Vygotsky (e.g., 1978, 1986), we propose that in order to do full justice to the dynamics of human ontogenetic and phylogenetic development we must focus on the broader concept of mediated action.
1.2 An important step in Bridgeman's target article is to view the ability to use language as an evolutionary "double" of the ability to make plans. Somewhere in evolution there emerged two parallel forms of behavior -- planning and language use -- that share many important characteristics such as serial and hierarchical structure. In fact, a schema can often be articulated both as a plan (e.g., a sequence of motor actions) and as a sentence describing that plan. This is important because it introduces the possibility to communicate plans to others. Plans developed by person A can be incorporated into plans by B. And the result of the incorporation may in turn be communicated to A. For example, Andy gives directions to Bert on how to get to a good fishing spot. Bert may, in following the directions, find a shortcut to the water. This elaborated version of the plan may be used by Andy himself again. We could in this case speak of a form of "heterogeneous recursion of planning." More directly, Andy may retrieve several plans from memory and combine these into a new plan: a form of "autonomous recursion of plans."
2.1 Bridgeman's view is reminiscent of Vygotsky and Luria's theory on higher mental processes. Almost fifty years ago, Vygotsky (e.g., 1978, 1986) proposed that the processes that characterize human thought (which can for the most part be described as forms of sophisticated planning) are the result of the convergence of two different lines of development: "the elementary processes, which are of biological origin, on the one hand, and the higher psychological functions, of sociocultural origin, on the other... The developmental roots of two fundamental, cultural forms of behavior arise during infancy: the use of tools and human speech" (Vygotsky, 1978, p.46). Both practical intelligence (simple planning) and elementary speech use are transformed into a complex of higher mental functions. This transformation takes place through interaction with persons in the direct environment. Through this (mainly verbal) interaction the child is supplied with a wide range of "mental tools." Many of these tools can be characterized as abstract plans. In the course of development the use of mental tools takes on a more and more important role in thinking and acting.
2.2 In his target article, Bridgeman (1992) mentions Vygotsky, but not with reference to mediated action. He merely assumes that the (autonomous) recursion that is the result of internalization of speech is sufficient to make available the "enormous power of human thought" (par.3.7). As it appears, this is only a superficial description of what is going on in the development of higher mental processes.
2.3 Vygotsky and Luria hypothesize that the child's development of higher mental processes follows the path from interhuman to intrahuman processing. This is similar (but by no means identical) to Bridgeman's idea that abstract plans can be conveyed to us by others. Young children (and animals) are extremely limited in their ability to incorporate abstract plans into their own. The crucial point, therefore, concerns how the child learns to make use of abstract plans.
2.4 Roughly speaking, a child acquires a sense of how to use abstract plans by simply taking part in them. In early stages this usually occurs in interaction with the mother. She gives meaning to "spontaneous" actions by the child. An example of this is the transformation of a baby's grasping behavior, which occurs in three stages. (1) A baby tries to reach for a toy. (2) The mother picks it up and hands it to him. (3) The baby's grasping action is transformed into a simple pointing action. In a very limited sense, this is an example of incorporating a simple abstract plan ("mother picks up something for me") into an existing action ("I try to pick up something"), transforming both into something qualitatively different ("I try to get mother to pick up something for me"). Many such qualitative transformations finally result in a full-blown system of higher mental processes.
2.5 As the higher mental processes develop further and further, more abstract plans can be incorporated in one's actions. These plans and mental tools may range from the very simple (e.g., a knot in a handkerchief) to the extremely complicated (e.g., superstring theory), but they all serve the same basic function: mediating actions. The plans exert an influence on the action (e.g., an efficient versus a clumsy algorithm for calculating the square root), eventually to transform it into something qualitatively different. In this way, through continuous (mainly verbal) interaction, a significant part of the vast reservoir of knowledge embodied in the sociocultural environment of the child becomes incorporated into of sophisticated planning and other higher mental processes. Precisely how this occurs is still largely unknown and remains a major research object in psychology.
2.6 The important point here is that the focus of research shifts from the structure of the plans to the question of how plans and other mental tools conveyed by others (come to) mediate our own actions. This is not to say that we should disregard the internal structure of plans, but we must view a plan's structure in relation to its mediating function in activity. The question then becomes: Why are we able to have our actions mediated by the plans (and mental tools) of others? This is partly a developmental problem concerning why young children lack this ability; and it is partly a genetic problem concerning why animals lack this ability. The Vygotskian tradition has mainly pursued the developmental question in the manner outlined (very briefly) above, even though Luria and Vygotsky (e.g., 1992) have also stressed the importance of the genetic line.
3.1 From a genetic perspective, almost all forms of communication are strongly favored. If we stretch the meaning of communication to let it range from, say, molecular interaction to signal exchange in animal populations, we might even say that communication is the most important prerequisite for evolution. Low creatures such as sponge cells can be said to communicate. This very basic form of communication results in the formation of autocatalytic cycles of morphologically different elementary cells mutually enhancing each others conditions of survival. It has been argued that aggregates of communicating entities (in the sense of sponge cells) may have an evolutionary advantage, sometimes to the extent that they become partitioned and start functioning as separate units in the ecosystem (e.g., Depew and Weber, 1989).
3.2 In this view, which derives from nonequilibrium thermodynamics, the internal complexity of organisms and organs, consisting of many communicating elements, is seen as a direct embodiment of some precursor ecosubsystem. It is important here that the original functioning of the elements is transformed as a result of their new position. The mass action of a system of many elements gives rise to emergent properties that are singled out by evolution. The more stable variant systems survive (which appears as a consolidation of the system). The further evolution of these systems is mainly directed by their emergent properties. When such a partitioning occurs, the requirements of the elements' functioning will change, for example, to ensure greater stability of the overall system. With communication, an elementary type of evolutionary pressure is thus introduced. The ensemble of elements further evolves into a system with increasingly complex (emergent) properties that derive from low-level interactions of communicating elements.
3.3 Learning introduces yet another type of evolutionary impetus. Evolution may constrain learning in individual animals by providing the brain with an initial structure that favors the emergence of certain cognitive functions while suppressing others (Murre, 1992). If an animal must learn a particular skill in order to survive, there will be a strong evolutionary pressure to equip the newborn animal with as much of the initial learning capability for this task as possible. This is a major evolutionary difference between nonlearning creatures (plants) and learning creatures (animals). Evolution "on its own" (without learning) is not able to determine many forms of complex behaviors. One argument is that the DNA cannot carry enough information to encode many complex actions (Changeux and Danchin, 1976); another is that evolution is not able to find isolated, deep attractors in the evolutionary search space (Hinton and Nowlan, 1987). We may thus argue that evolution broadly prestructures the brain and that learning imparts a finer structure on this. Depending on the learning tasks, the more successful initial brain structures will have a higher chance of reproduction. Learning can therefore be said to guide evolution in a particular direction. It adds extra power to evolution by greatly enlarging the complexity and flexibility of actions that can be produced by an organism.
3.4 Humans are able to mediate their actions by plans conveyed by others. This invokes a third type of evolutionary pressure. Newborn babies do not learn only by themselves; they also incorporate a huge range of existing plans (and other mental tools) into their own actions. The better they succeed in having their actions mediated by ready-made plans, the better they will function and the better their chances of survival. As argued above, this is by no means an automatic procedure. The process whereby external plans become internalized encompasses all of the development of higher mental processes. Only through this developmental process can an individual gain the high level of self-control necessary to overcome the barriers of sophisticated planning. For this to occur, speech (or some comparable form of sophisticated sign-use) is crucial. But as was also argued above, for language to be effective as a means of conveying abstract plans, higher mental functions must be developed to a certain minimal degree.
3.5 We thus see that there is an important mutual dependency between "the reservoir of plans and mental tools" (which is in essence culture) and sophisticated planning and other higher mental processes: The development of one enhances the effectiveness of the other. The focal point of this mutual enhancement is mediated action. From a genetic perspective, we must hypothesize that language (as opposed to the simple speech found in young children) and sophisticated planning have co-evolved, not because they share a common structure, but because they both perform a necessary function in mediated action. Simple speech use and practical intelligence may have co-evolved because of a common structure, as is argued by Bridgeman (1992), but these are only the precursors of language and planning (or higher mental processes).
3.6 In summary, we have argued that evolution has supported a combinatorial explosion in the possibilities for action in man as a result of evolutionary pressure from three distinct sources: communication, learning, and mediated action. The latter is intimately related to language use and planning and hinges on the incorporation of plans conveyed by others into once own actions. The individual development of sophisticated planning ability is by no means an automatic or self-contained process. On the contrary, it is a complex process of gradual internalization of sociocultural entities (such as plans) that come increasingly to mediate actions. The most significant research question concerns the process of mediation rather than the structure of plans or other mediating entities. We therefore propose that mediated actions, rather than plans, occupy the center stage of psychology.
Bridgeman, B. (1992). On the Evolution of Consciousness and Language. PSYCOLOQUY 3(15) consciousness.1
Changeux, J.P., and A. Danchin (1976). Selective stabilisation of developing synapses as a mechanism for the specification of neuronal networks. Nature, 264, 705-712.
Depew, D.J., and B.H. Weber (1989). The evolution of the Darwinian research tradition. Systems Research, 6, 255-263.
Hinton, G.E., and S.J. Nowlan (1987). How learning can guide evolution. Complex Systems, 1, 495-502.
Luria, A.R., and L.S. Vygotsky (1992). Ape, primitive man, and child: essays in the history of behaviour. Hemel Hempstead: Harvester Wheatsheaf.
Murre, J.M.J. (1992). Categorization and learning in modular neural networks. Hemel Hempstead: Harvester Wheatsheaf (forthcoming in September).
Vygotsky, L.S. (1978). Mind in society: the development of higher psychological processes. Cambridge, MA: Harvard University Press.
Vygotsky, L.S. (1986). Thought and language. Cambridge, MA: MIT Press.