A life decided beforehand

This is why Bratman defines intentions as the building blocks of plans. Once a decision has entered the network of our intentions, it creates new constraints and problems. Some possibilities take on greater significance, whilst others fade into the background. If I have decided to leave the car with my son and go to the library by public transport, the idea of driving to the library in that very car is no longer among the acceptable alternatives. Previous plans thus function as a filter for the admissibility of possible options. In other words, they narrow down the range of alternatives worth considering.

In economics, choice theory takes as its starting point a pre-defined set of alternatives and asks which of these will be chosen. Bratman’s position takes us a step back, as a preliminary matter. What is the origin of these alternatives? Why do I consider these and exclude all the others? Part of the answer lies in the history of decisions already made. My previous plans determine what deserves my attention today. Rationality thus takes on a historical or procedural dimension, as some would say. What makes sense to do today depends, at least in part, on what I reasonably decided to do yesterday. Of course, a previous decision can be reviewed and abandoned. The world changes, new information comes to light, we discover errors, circumstances shift. We update our assessments and beliefs on the basis of new data. A theory that transformed every intention into an irrevocable constraint would describe a form of rigid obstinacy, not a form of rationality. Bratman, on the other hand, attributes a more subtle property to intentions: that of stability or inertia. ‘Every intention (…) concerning the future is mutable or revocable,’ he writes. ‘But revocability does not necessarily imply an actual change of mind. In the absence of new considerations, I will normally simply maintain my intention right up until the moment of action. Maintaining my previous intention and failing to revise it constitute, so to speak, the ‘default option’ (1987, p. 17). Once I have decided to go to the library this afternoon, the matter tends to remain settled. The decision is retained without being continually subjected to fresh judgement. The arrival of relevant information may reopen the matter, but in its absence, deliberation proceeds by taking the previous decision as part of the background. It is this relative stability that allows the intention to endure over time. The interesting thing is that inertia performs a cognitive function. Every decision temporarily closes a door and, precisely for this reason, allows us to deal with the doors that remain open.

Let’s imagine we’ve decided to pursue a PhD. The decision still leaves almost everything unresolved: where to study, which topic to choose, who to work with, and which modules to take. And yet it already provides a framework. When I choose the university or the topic of my thesis, I will keep the broader plan in mind. Bratman describes plans precisely in these terms: partial and hierarchical. A general plan contains gaps that will be filled in when the time comes. Their incompleteness is an asset. A highly detailed plan for the next ten years would quickly become useless because the world changes and because today we are unaware of many of the circumstances in which we will have to act tomorrow. A completely open-ended plan, on the other hand, would offer too little guidance. Planning occupies this middle ground. Bratman’s favourite examples are very broad and general. They range from pursuing an academic career to raising a family or campaigning for social justice. These are projects that can span decades and which, precisely for this reason, must remain radically incomplete. They give structure to a life without writing every page in advance. The ability to plan, therefore, does not simply consist of having preferences about the future. It consists of being able to leave certain decisions open because others are sufficiently stable. And this is where the philosophical question once again begins to become a question of justice. To construct a plan, it is not enough simply to be able to choose. We must be able to rely, within reasonable limits, on the persistence of the conditions within which those choices will take on meaning. We can then better understand what happens when we make decisions.

Deliberation gives rise to an intention. The intention becomes part of a network of plans. That network shapes subsequent deliberations. Some issues are deemed resolved, whilst others become urgent. New problems concerning the means emerge. Some alternatives are ruled out because they are incompatible with previous commitments. Yesterday’s decisions become, within certain limits, the framework for today’s decisions. In this regard, Bratman distinguishes between two dimensions of the commitment inherent in an intention. The first is volitional: when the time for action arrives, an intention that is still active tends to govern behaviour. The second is centred on reasoning: in the interval between decision and action, the intention tends to persist, guides the search for means and influences the formation of subsequent intentions. It is the combination of these two dimensions that explains how a deliberation carried out at one moment can exert organised effects long afterwards.

It is therefore easy to see why Bratman focuses on the famous paradox of Buridan’s ass. Placed exactly halfway between two perfectly equivalent piles of hay, a hungry donkey would starve to death because there is no reason to go to the right rather than to the left. What may seem like a caricature of a decision-making problem actually raises a serious issue. In life, we constantly encounter almost equivalent alternatives. Two routes to reach the same town, two identical packets on the shelf, two equally convenient timetables. An agent capable of planning possesses the ability to decide even under conditions of equal desirability. Once the decision has been made to take the A1, that decision creates a new problem concerning how to join it. From that moment on, some turns become appropriate whilst others are incompatible with the plan. The act of deciding has produced a practical structure that was previously lacking.

Here, the distance from the image of the agent as a mere bearer of desires and beliefs becomes evident. Bratman wishes to accord intention an autonomous status. Desires and beliefs explain a great deal about our behaviour, but they leave in the shadows that capacity through which we provisionally set a course of action and allow it to organise what comes next. It is precisely this function that makes a life extending over time possible. And it is perhaps here that Bratman’s argument begins to approach the problem that has been with us for several weeks. So far, we have encountered ‘noitude’, as we have called it, by following different paths. With Margaret Gilbert, we analysed the emergence of reciprocal obligations from joint commitment. We followed Raimo Tuomela in distinguishing between acting in ‘I-mode’ and acting in ‘we-mode’. Then there was John Searle and his social ontology, which underpins institutional reality. Bratman adopts a different perspective. He begins with the architecture of individual action and observes a property of plans that we have already encountered almost without realising it. Plans of action are not merely an individual matter. In fact, they also have social significance because they help us to co-ordinate our actions.

First and foremost, plans bring a person into harmony with themselves. The plan drawn up today makes it easier to predict what that same person will do tomorrow. But it is precisely this predictability that can influence other people’s plans. If I know you’ve decided to be at the library at four o’clock, I might decide to meet you there. Your intention becomes a condition of my plan; what I decide to do may, in turn, alter yours.

All it takes is a slight increase in complexity for a new problem to arise. Two researchers decide to write a paper together. Both want it to be published, but that shared desire still leaves almost everything to be worked out. They need to divide the work, establish a sequence, and agree on a methodology. Each person’s plan must leave room for the other’s and, at the same time, depend on it. If one of them decides to hand in the analysis on Friday, the other can organise their own work on the assumption that the analysis will be ready by Friday. A decision made by one person thus begins to shape the future of another. But are two plans that fit together already a shared plan? Not necessarily. I might decide to leave the house at eight because I know that at that time my neighbour vacates the parking space. My plan depends on his; our actions are coordinated, and yet we are not doing anything together. Mutual predictability can coordinate individual plans without turning them into a joint activity. What, then, is needed for your plan and mine to become parts of something that we can properly say we are doing together? This is the problem that Bratman will address as he moves from individual planning to shared action.

First, however, let us return to the question of justice with which we began. The ability to shape one’s own future depends not only on the stability of our intentions, but also on the ability to rely on the decisions of others and on the relative stability of the social and institutional environment. Moreover, not everyone occupies the same position within this network of plans. Some make decisions upon which others must build their own futures; others must continually adapt their plans to decisions taken elsewhere. An employer decides whether to renew a contract, but that decision becomes part of the worker’s family plans. An administration decides whether to maintain a service, but that decision affects the plans of those who have organised their lives relying on that service. We thus encounter, from a different angle, a problem we have already faced. There is the power to choose between alternatives. There is the power to decide which alternatives may be chosen. And there is, even further upstream, the power to construct the plans within which others will be called upon to act. Inequality can also concern this: how much we can plan and how much we must adapt; how much others must take our plans into account and how much we, in turn, must take theirs into account. Bratman does not build a theory of justice on this. But his theory of planning allows us to formulate the question. Before asking ourselves whether a shared plan is just, however, we must first understand what truly makes it shared.