View through Three Arches of the Third Storey of the Colosseum by Christoffer Wilhelm Eckersberg, Denmark 1815
The setting
I will focus now on how prehistoric humans might have managed their purposeful interactions of governance without knowledge of the real intentions of their counterparts. Prehistoric individuals, unlike their primate ancestors, did already have the cognitive ability to infer that the intentions of other humans are likely in greater or lesser degrees to differ from their own intentions. This ‘objective’ cognisance of intentions in interactions allowed humans to develop skills and knowledge to identify or imagine and compute probable patterns and variabilities of intentionality.
Furthermore, the intentions even of young children are shaped by consciousness that they belong to a group, and, therefore, that the intentions of others in the same group may in essential ways be quite similar to their own. This awareness that other humans have intentions, and that those who are close to you are more likely than strangers or outsiders to have similar intentions, will have helped prehistoric humans to make the concessions and compromises required for collaboration.
There nevertheless was a fact about ‘intentionality’ that impeded human interaction — the impossibility of knowing the true intentions of others. This unavoidable predicament has been one of the principal motivators of the human enterprise of governance. Ultimate and ineluctable ignorance about the interior of every potential counterparty’s mind — irrespective of repeat experiences, proven reliability, and the close contact upon which reputation and trust can be built — propelled societies to find mechanisms that substitute for being unable, and indeed not desirous, to know the multifarious and potentially unpleasant individual intentions of all its members. The great projects of building society did not want this knowledge, and governance would have been thoroughly overloaded if it had been forced to account for it.
Mechanisms that partially solved the problem of not-knowing individual intentions can be divided into three main types: a) explicit rules to regulate the standards of action within a society; b) a genuinely shared belief in the impersonal common good or in a supernatural being, either of which may be a surrogate third party mechanism for policing intentions by compensating for the absence of reliable understanding about intentions; and c) systems of reciprocity operating intuitively without central rules. These three mechanisms are likely to have existed simultaneously given that each was still incipient and in its state of evolutionary origin. It was safer to combine the three methods than to run the risk of selecting between them.
I further assume that each of these mechanisms will have represented a considerable stimulus to human neural networks. Combined efforts in all three should have contributed to encephalization. I find that all contending claims about the drivers of brain size (ecology, sociation, and knowledge acquisition) can be encompassed under the general heading of ‘governance’ insofar as governance pertains to three prehistoric mechanisms — rules, third parties, systems. Brain growth drivers included cooperative and individualised decisions about subsistence, about social order, and about the preferred mechanisms of decision making. Given such range and inclusivity of activity, and the inescapability of participation in prehistoric governance decisions, it is doubtful that many prehistoric persons would have had the luxury of free riding on the skills of their companions (which has been alleged to decrease brain size).
I will separately examine specific decisions that created prehistoric rules (‘The Second Decision’). Here I discuss intentionality in interaction and the role of third parties insofar as they complement a systems perspective, and offer preliminary comments about rules that potentially help or hinder systems and third party surrogates.
ABC No Rio by Dan Witz, USA 2011
A game theory of contingent intentionality
I propose ‘contingent intentionality’ as a concept that helps explain the emergence of systems when dualistic one-on-one conditions of interaction are informal, uncertain, and unregulated or unmediated by a third agency. I suppose a transaction of favours, services, advantages, or possessions. An individual acts intentionally if the action is deliberate with a definite purpose. He acts contingently when the action can be followed through to its end only if certain circumstances prevail or can be anticipated to prevail. When one party to the exchange does not know the true intention of his counterparty there is contingency about intention. Presupposing absence of transaction rules, if there is contingency about intentions there is freedom to act differently and to modify the intentions without any resulting sanction.
The critical precondition for un-contingent action would be that one or both actors have sure knowledge and understanding of the intentions and the meaning of the intentions. The negotiation and commitment to action would be unproblematic. One intentional actor could just accept or reject the other’s proposed intention. However, because true intentions are unknown, the transaction, or the decisions made to clarify the terms of a transaction, depends in reality on how the potential partners can feel their way to tolerance and understanding about inevitable uncertainties. In such conditions it may be possible to modify intentions, give greater credibility to the intentions, publicise the intentions to an audience in order to spread commitment and risk, or subject fulfilment of intentions to preagreed sanctionable conditions. The starting point is clear — there is inevitably a doubling of intentions and, in the absence of a relevant framework of rules, an inevitable doubling of contingencies.
This is how a transaction of A and B would first appear:
A begins with his intention, a goal-oriented action whose achievement depends on B’s corresponding reciprocal action with acquiescence or consent. A’s intention in his interaction with B is initially contingent on a selection he must make among perceivable action alternatives, conformant with his intention. B’s response to A will be contingent on A’s selection but is also contingent on B’s own reactive and corresponding, compatible, correlative give-and-take selections.
The doubling-up of contingency will lead to a chain-reaction of secondary intentions adapted on the spot to overcome uncertainty. Because the outcome of A’s action (his success in attaining his goal) is contingent on B’s reaction to what A does, A necessarily reorients his intentions to B’s probable response, but also to how A thinks B may have interpreted A’s expectations of how B will respond. Yet at the same time the resultant mass of interactive intentions has somehow to be communicated as a generalisation that is understood by A and B but is not identical for either A or B.
This looks complicated when written out as a game scenario. Yet, the variables of exchange in a prehistoric context would not have been more cognitively challenging than devising a plan for subsistence or for strategic hunting. The reactions and counteractions to declarations of intent between consenting adults could have been instantaneous and instinctual. And the assimilation and calculation of this long chain of proposals should not have been beyond the cognitive competence of prehistoric post-Neanderthal individuals, since they had in most cases already achieved what is referred to as a ‘level-five order of intentionality’. If, as it is credibly suggested was indeed the case, they were competent enough to manage a construction such as “D believes that C supposes that A wants B to believe that E can persuade B to agree with A” then I see no reason why they should not also have been able to take active control of a dual dialogue such as the one enacted above — namely a negotiation between A and B that incorporates several modifications of A’s original proposal to B on the basis of several additional well-informed counterproposals.
Thus far I have presented the amicable low-level routine of puzzling through the various unknowables of intentionality. But there is a latent danger of conflict arising if two parties cannot lift the veil of contingent intentionality sufficiently high to see the way forward. We should, therefore, realistically think of doubled-up contingent intentionalities as a potentially universal hazard of conflict caused by the indeterminability of meaning and understanding between the social units of interaction. A primitive difficulty like this one, multiplied over many individuals and exchanges, could threaten the maintenance of social order and material subsistence.
Here is an alternative way of conceiving the ‘A’ and ‘B’, incorporating observation and, potentially, intervention by a third party called ‘C’:
In the interaction of two motivated individuals, A does not want to act in pursuit of an objective without a relative certainty of knowing how B will react. But B, who is equally similarly motivated to pursue goals, will not wish to react to A without relative certainty of knowing A’s real meaning, intentions, and expectations. B can ignore the uncertainty, or may be persuaded to think there is really no uncertainty.
But when B does react to A, A will then face precisely the same dilemma in reverse and in return. In other words, when B reacts to A he turns the tables on A. A will then be in the position that B previously was, and the contingencies of interactive intentions should be plainly evident on both sides. C, an interested though neutral observer of A and B, can see how this doubled-up uncertainty about intentions is influencing the interactional behaviour of A and B. Regardless of how the precursors and outcomes of the interaction are modelled and calculated by A and B, C could be in the position of an arbiter or mediator. Failing this, the very fact that a neutral party C is watching and objectively appraising the transaction might exert a moderating influence on the behaviour of A and B, and facilitate a resolution to the contingencies.
Street Scene in Windy and Rainy Weather by Christoffer Wilhelm Eckersberg, Denmark 1846
Third Parties Real and Artificial
A ‘third-party’ in the prehistoric pre-law context is either a) a subliminal behavioural code standing in for the priority interests of the whole society, or b) a supernatural entity, embedded symbolically or perhaps to some extent subconsciously in the minds of negotiators, and that is perceived as an observer of the negotiation and a potential instructor in the use of behavioural devices that will facilitate agreement within their maze of contingent intentionalities.
The presence of the third party may neutralise sources of misunderstanding that might otherwise paralyse the interaction. It was not necessary for a third-party to be seen as a ‘power’, a surrogate controller or enforcer and punisher, though it may have been regarded as such by humans who were disposed to be superstitious. In its prehistoric form the third party simply provided binary instruction guidance, internalised in the form of a socially-constructed or socially-amplified conscience or consciousness for the purpose of distinguishing good action from bad action.
In the prehistoric group society context belief in a supernatural being was socially functional in one or a combination of five forms as — a) a shared belief that bonded society; b) a metaphysical enactment or encodement of binding behavioural rules; c) a symbolic representation of society itself; d) the precise identity of the society; or e) the imaginary third party arbiter. It was only much later that religion became important as a legitimator of political rulership. Here, very briefly, I am interested only in the aspects that might have surmounted the impediments to understanding contingent intentions.
To begin with we can benefit from Robin Dunbar’s conclusions (2022) relating to a neural ability to conceive beliefs in the existence of a god. I will attempt a summary. With first-order intentionality our primate ancestors were aware of their own beliefs about the world. With second-order intentionality they were aware that other minds similarly have beliefs about empirical facts. With third-order intentionality they could imagine a living being in a spirit world, and they could recognise the possibility that other minds might similarly imagine non-empirical worlds. Only with fourth-order intentionality could they formulate a personal belief about the existence of an intentional god, and conceive of that god as having an intentional effect on the human world. With fifth-order intentionality, appearing roughly 200,000 years ago, and permitting a language for communicating complicated symbolic ideas, modern humans were able to share their beliefs about god’s intentions.
Therefore, at an early point in the evolution of prehistoric cognition it was possible for two humans to jointly and genuinely believe in the presence of a third and potentially all-knowing supernatural being overseeing and passing judgement on the quality of their interactions. If A and B identify with the same society, and genuinely share a belief in the existence and surpassing wisdom of their society’s god-like substance, they will have begun, just as four- to six-year old children in modern society, with a group awareness that their opposite counterparty probably has a range of intentions similar to their own.
How genuine was their supernatural belief will necessarily have been a subjective judgement made on both sides about their counterparts. In fact, though, the process of assessing the reputation, trustworthiness, and credibility of their counterpart’s religious beliefs is exactly the same process that would be undergone with respect to any person’s non-religious behavioural profile. Do they honour their promises? Do they cheat and deceive? In a prehistoric post-neanderthal evaluation of orders of intentionality we should keep in mind that the well-theorised competence for complex visualisations and calculations of long chains of intentions applies just as effectively to intentions directed toward deceit and pretence.
I am still working on the secular aspect of prehistoric ‘third party’ oversight and will include it in the next post.