The Commonsense Conception and its Relation to Scientific Theory terug (Nederlands)

In: Philosophical Explorations 2001/1, pp. 17-30                                    back (English)


In this paper I discern two concepts of meaning: meaning 0 – which is assigned by us on the basis of our commonsense conception in order to constitute our own daily reality – and meaning 1, which we assign when we interpret reality scientifically. Authors who contend that the commonsense conception is nothing but a kind of scientific theory, do not see that the two fields of life have their own concept of meaning. Commonsense and science are not separate from each other, however: though both have their own practices, the way we interpret reality scientifically presupposes our commonsense conception.


The Commonsense Conception and its Relation to Scientific Theory

Henk bij de Weg

When we explain an action in terms of propositional attitudes, such an explanation follows the same lines as an explanation of a physical fact or event; only the variables are different. Where physical explanatory variables are found in an explanation of a physical fact or event, ‘mental’ variables are found in a ‘mental explanation’ of an agent’s action. Moreover, there is no difference whether an explanation has been formulated by someone (e.g. a scientific researcher) who has observed an agent or by the agent himself or herself.

When referring to mental variables, one usually thinks of an agent’s belief or desire, and for that reason I shall call a mental explanation a ‘belief-desire explanation’ (or BD explanation for short).1 An approach to BD explanations as just formulated is defended by, for instance, Churchland, Dretske and Jackson. To make clear the similarity between a physical explanation and a BD explanation, Churchland first shows the logical form of the way the strength of a current is determined by means of its tension and the electrical resistance, then states: ‘Action explanations, and intentional explanations in general, follow the same pattern. The only difference is that here the domain of abstract objects being exploited is the domain of propositions, and the relations displayed are logical relations’ (Churchland 1992, 115; cf. 4-5). Churchland’s ‘action explanations’ and ‘intentional explanations’ denote what I have called ‘BD explanations’. For Dretske, the similarity between BD explanations and explanations of physical phenomena is not their logical form but the way the reality is represented. Whether one explains why the thermostat turned the furnace on because the room temperature dropped or why the duke stood up because the queen entered the room, in both cases basically the same model of representation can be used (Dretske 1988, 50 and passim). For Jackson, BD explanations are a kind of dispositional explanation which can be compared with physical dispositional explanations: someone falls through the ice and this is explained by referring to a concrete cause (the person was skating on the ice; a pike had broken the ice during an attempt to seize the skater) plus the brittleness of the ice. A BD explanation approaches matters in the same way: ‘If my movement is explained by a desire for coffee, it must be true that I desire coffee, that the desire for coffee caused the movement, and that the movement is, or is a causal consequence of, part of the kind of causal chain that makes it true that I desire coffee’. It is the last part of the quotation that refers then to a disposition or an analogue of a disposition of an agent in the same way that the brittleness of the ice as a disposition of the ice is one of the factors that caused the agent to fall through the ice (Jackson 2000, 88-89; quotation on p. 89).

In these and other cases where BD explanations are considered to be similar to or derived from physical explanations, the essence is a methodological if not also ontological physicalism. But are BD explanations really nothing but physical explanations with different variables? This is the question I examine in this article. But in doing so I want to generalize the problem, for explaining is only one, albeit an important aspect of both what is called the commonsense conception – or our commonsense views or ‘folk psychology’2 – and of science or theory (conceived in a narrow sense3). My question will be: is there a difference between our commonsense conception and scientific theory? And since I shall answer this question affirmatively, I want to examine also the relation between both. In order to make my criticism concrete, I focus especially on the views of Churchland and Jackson.4

1. Meaning 1 and meaning 0

I shall start with an example that I have taken from Stoutland 1976. A women’s head nods. Sometimes we take this movement as just physical: the person nods because she is falling asleep. In other circumstances, however, we take it as an action: the person is giving a sign or is greeting someone. But why is the nodding movement sometimes taken as just physical and in other cases as an action? This becomes clear when we ask the person why she nodded. If she says, for instance, that she did so because she saw a friend (a belief) and that she wanted to greet him (a desire), then we call it an action. But if the nod was an expression of her falling asleep, then she cannot give a reason and a desire for it: the nodding movement happened to her. We can also put it this way: the nodding movement that was used by the person as a greeting contained – unlike the nodding movement of falling asleep – a meaning for the agent.

Hesse showed, however, that such a traditional view according to which natural science data do not have a meaning, unlike data in the human sciences, has been superseded. In every empirical assertion, also in pre-theoretical descriptions of observations that are starting points of investigation and theory, ‘we employ concepts that interpret the data in terms of some general view of the world or other ... There are no stable observational descriptions, whether of sense data, or protocol sentences, or "ordinary language", in which the empirical reference of science can be directly captured’. It is true also for natural science that ‘what count as facts are constituted by what the theory says about their interrelations with one another’ and that ‘meanings in natural science are determined by theory’ (Hesse 1980, 172-173; italics Hesse). This standpoint is endorsed by Churchland and he has developed it from a neurocomputational point of view (cf. Churchland 1998b, 19-21). It implies that natural data have a meaning as well. Thence it seems that physical phenomena that are explained in natural science, like the nodding of a person when she falls asleep, do not differ in this respect from mental phenomena that occur in BD explanations, like the intentional nod of greeting.

However, Hesse tends to neglect that there really is a difference between the meanings we give to physical phenomena and those we give to mental phenomena (and generally to social phenomena). This was already implied in our example. Mental phenomena are given meaning by the agent himself or herself; physical phenomena get their meanings from us from the outside. Therefore, in order to be certain, we have to ask the nodding person whether she meant something with her movement and, if so, what she meant. On the other hand, not only acting persons themselves make BD explanations for themselves (often, however, they do it implicitly and formulate explanations only if requested). Also scientists try to ‘catch’ what agents do in explanations. This is why Giddens comes to the conclusion that: ‘… the meaning of scientific concepts is tied-in to the meaning of other terms in a theoretical network; moving between theories or paradigms involves hermeneutic tasks. The social sciences, however, imply not only this single level of hermeneutic problems, involved in the theoretical metalanguage, but a "double hermeneutic" because social-scientific theories concern a "pre-interpreted" world of lay meanings’ (Giddens 1977, 12; italics Giddens). This made Habermas distinguish between two levels of meaning: level 1 and level 0. The former is the level all sciences are faced with when they theoretically interpret their objects; the latter level is characteristic of those sciences that have to deal with an object that has been given a meaning by the people themselves; in our example, the nodding movement as a greeting (Habermas 1982, 162-163). I therefore think it makes sense to draw a distinction between two kinds of meaning. I call them meaning 1 and meaning 0. The former is the concept of meaning that is used on the first level. It is the meaning a scientist gives to an object, either physical or social in character; in other words, it is the scientist’s theoretical interpretation of reality. Meaning 0 is the concept of meaning for the underlying level 0. It is the meaning the people who make up social reality give to this social reality or to parts of it themselves; in other words, it is their interpretation of their own lived reality.

The existence of a double meaning from the scientist’s perspective does not imply, however, that the interpretation of reality by the people themselves is different in character from the theoretical interpretation of it by the scientist. As just formulated, the thesis of double hermeneutics says only that the interpretations of the social reality by the agents themselves are single and that those by the scientist are double. As we have seen, Churchland states that the interpretations of their reality by the ‘common people’ have the status and structure of scientific theories. Churchland also says it expressly: ‘Our self understanding … is no different in character from our understanding of any other empirical domain’ (Churchland 1992, 112). Here, ‘our understanding of any other empirical domain’ refers to the theoretical interpretation by the investigator. For my theme – BD explanations and generally the commonsense conception (understood as the interpretation of social reality in general) – it is important that according to Churchland the common people make generalizations that in no respect can be distinguished from scientific generalizations, i.e. that on the whole they are causal and nomological in character. And just as one can use causal and nomological relations to give explanations and make predictions, one can use commonsense relations in the same way. One can even use them to control and manipulate the behaviour of others. But Churchland has to admit that the commonsense theory is a rather ramshackle one: predictions often do not come true, and generalizations are vague and sloppy or festooned with ceteris paribus clauses, and the commonsense theory shows hardly any progress. But that does not alter the fact that we are talking here about a theory (id., 111-116).

I do not want to deny that people use their daily experiences and their commonsense views based on them to predict what other people will do, or possibly to control or manipulate them. However, this does not imply that what we have here is a scientific theory. To demonstrate this, I shall use the example of a grammar. Jackson – who also defends the view that the commonsense views we use for explaining and predicting actions are a kind of theory – employs this example to show how, according to him, our commonsense conception is related to our brain states. The example is, however, very useful for providing insight into the relation between a commonsense ‘theory’ and a scientific theory, and for making clear the differences between them. As native speakers of a language, we have in our heads a (usually implicit) knowledge of what are correct and incorrect sentences in our language, what are correct and incorrect forms, etc. We can call this our ‘implicit grammar’. One thing grammarians do is make this knowledge explicit and systemize it (cf. Jackson 2000, 92). The ‘explicit grammar’ that is so made seems to be a reflection of our implicit grammar. In a certain sense that is true, but there are differences. The explicit grammar has been formulated in terms and rules native speakers will never use as such. For example, such a grammar distinguishes between substantives and adjectives and indicates when inversion is applied, things native speakers usually are not aware of when they speak. Moreover, although it is correct to say that the explicit grammar is a reflection of the implicit one, the reverse is not true: the latter functions as a norm for the former. If a native speaker no longer keeps to the current grammar on a certain point and more native speakers begin to speak in the same deviant way, after some time we no longer say that these speakers make grammatical mistakes, but that the language has been changed. However, if on the other hand the explicit grammar does not correctly reflect the implicit one, the language of the native speakers does not change, even if the grammarians maintain that they have described it correctly. If the native speaker does not want to conform to the rules of the explicit grammar, then the latter simply describes the implicit grammar in a wrong way. The language as spoken by the people themselves (i.e. the implicit grammar) is the reality that patterns the explicit grammar.

There is an analogous difference between our commonsense conception and a scientific theory. I shall skip the details5, but the essence is this. When authors like Churchland and Jackson conceive the commonsense conception (‘folk theory’) as a theory or impute a theoretical character to it, they mean not only that the commonsense conception has the same structure as a scientific theory (cf. above), but also – and especially – that certain commonsense views are used to explain and predict the behaviour of other people by those who have these views. As for Churchland we have seen this already. Jackson says, for instance: ‘If we [the folk] can make detailed predictions and explanations of behaviour, we must have a theory good enough for the task. The predictions and explanations do not come from no-where … and that is what "theory" means here’ (Jackson 2000, 91-92; italics HbdW). For Jackson, it is not so much the form that is important for the theoretical character of the commonsense conception, but the fact that the commonsense views are founded and that explanations and predictions are not guesses (ibid.). On the other hand, Churchland stresses particularly the similarity in structure between the commonsense conception and a scientific theory (Churchland 1992, 115-116). What is ignored in such views, however, is that explaining and predicting on the basis of commonsense views is different in character from explaining and predicting on the basis of a causal or nomic theory. To show this, I shall take the case of predicting.

We can make predictions based on a causal and nomic theory and on our commonsense views. If a scientific theoretical prediction does not come true, we say that the prediction is false. We then doubt the correctness of the theory behind the prediction or at least the methods used. Predicting from our commonsense views is different. Of course it may well be that our commonsense views are false and that for this reason our predictions do not come true. But even if our commonsense views are correct and have been proved so in the past, it is still possible that our predictions do not hold good in new cases. For example, we expect that somebody will do something from habit but he does something different: the agent has changed his opinion. We think that it is normal that the persons we deal with keep their appointments, but Peter sometimes does and sometimes does not. We call him ‘unreliable’. Or we watch a football match and it strikes us that the offside rule is not applied. Later a friend tells us that the rule was inoperative. The match was an experiment. However, such cases are not falsifications of our commonsense views, nor do they call for adjustment of them. The fact is that the relation of our commonsense conception to reality simply is different from the relation of a scientific theory to reality. If in the case of a scientific theory the reality is not as supposed, it has repercussions on the theory. But if the reality does not correspond to our commonsense views, it tells us something about the reality itself. A person has changed his opinion because from now on he behaves differently and because our former predictions no longer come true. This is what we call ‘having changed one’s opinion’, and it cannot be that somebody says he has changed his opinion while he does not behave that way (such a person is ‘untrustworthy’). Likewise a person is unreliable just because our predictions of his behaviour are sometimes correct and sometimes not. And the football match where the offside rule is not applied (though the rule has not been abolished for normal matches) is called by us an experiment just for this reason (cf. von Wright 1983, 43-44).

These examples show that our commonsense views as such can be true while they may not always hold good. But then it is not the truth of these views that is at issue, as in the case of a scientific theory, but what Apel calls their applicability (cf. Apel 1979a, 197). That a commonsense ‘theory’ sometimes is not applicable does not need to affect its truth. Even more, we have just seen that the fact that our commonsense views are not applicable may constitute the social reality in a certain way, because we then apply the predicates ‘having changed his mind’, ‘unreliable’ or ‘experiment’.

To return to our language example. Words are added to a language, others are dropped and language rules change, and this determines what the native speaker’s language is. Grammarians describe this language, and for them it is important that their descriptions are in agreement with the native speaker’s language, otherwise these descriptions are incorrect. We see something like this also in the case of our commonsense conception. If somebody fails to keep an appointment with me, I may call him ‘unreliable’, a friend of mine may call him ‘absent-minded’ and another person will perhaps say: ‘He’s young, so he’s a little playful’. In other words, we give our social reality a certain meaning based on our commonsense views and on certain expectations (predictions) that result from them, and it is this that determines what the reality is for us. Those who consider our commonsense conception to be a theory analogous to a scientific theory do not see that the meaning we assign on the basis of our commonsense conception is different from the meaning we assign to reality if we interpret it theoretically. They ignore the difference between meaning 0 and meaning 1. The former is the meaning we assign in order to constitute our own reality to which we have a subjective relation; the latter is the meaning we assign in order to get to know a reality (including a social reality) that already exists by describing and explaining it. As social scientists, we have or try to have an objective relation to this reality, just as grammarians have an objective relation to the implicit grammar they describe. This is exemplified by the fact that a scientific theory usually employs a terminology that is not employed by agents as such when they act, just as in the case of an explicit grammar. Phrased in terms of the difference between meaning 1 and meaning 0: those who think that the commonsense conception is a theory, consider commonsense views to be meaning 1 assignments by understanding them only as views by which we explain and predict, while in fact they are meaning 0 assignments.

2. Practice as mediation of meaning 0

Seen in this way, the commonsense conception in conjunction with the meaning 0 assignments based on it is of course nothing but a practice (cf. recently Baker 1999). According to Churchland, however, this does not make a theory different from the commonsense conception: ‘…learning the theories peculiar to any discipline is not solely or even primarily a matter of learning a set of laws and principles: it is a matter of learning a complex social practice, of entering a specialized community with shared values and expectations, both of the world and of each other… The contrast drawn between a "theory" and a "practice" is therefore a false and misleading one’ (Churchland 1998a, 11).

In fact, also science is a practice and scientists assign a certain meaning 0 to what they do (bij de Weg 1996, 122-139): they define what a theory is for them, and what explaining and predicting is for them. Since Kuhn, we have known that what counts as a valid interpretation of reality is subject to change. What is not subject to change, however, is the object of interpretation as such. Since Darwin developed his theory of evolution, the origin of life has not moved up to the past, but we have moved it up to the past. But the practice of a scientist is not the reality that is studied with the help of a theory; that reality is a reality the scientific investigator is objectively opposite to. The scientific practice is a practice that makes this objectification possible and determines the rules for it. That these rules can change does not alter this fact. However, in the case of the commonsense conception the bearer of this conception (the counterpart of the scientist) is himself a part of the reality to which this conception refers. The bearer of the commonsense conception is subject in this reality and the practice coincides with the reality to which it refers. A gesture, for example (as a gesture and not merely as a movement), cannot be seen in isolation from the commonsense view of the bearer of this view; for example, a gesture that to the Dutch is a signal to go away, is to the Greeks a request to come nearer.

That there is a difference between a practice that is the basis of a scientific theory and the practice of the commonsense conception becomes clear from the occurrence of the so-called ‘Merton effects’, i.e. the self-fulfilling and self-denying prophecies. It cannot be that one makes a prediction on the basis of a scientific theory and that the prediction is fulfilled or otherwise by the object of the prediction because of the existence of the prediction. Stars do not explode because we have predicted it: they explode and we predict it by means of our theory. On the other hand, stars do not ‘decide’ not to explode because they ‘know’ we have predicted that they will. But such effects do occur if people know that we have predicted or foreseen that they will behave in a certain way. For example, if the organizers of a demonstration expect many people to participate, this may come true because on hearing this expectation, nobody will fear that they will be part of a minority and everybody will want to be able to say that they participated. In another situation, fewer people than expected may turn up, because many may think: ‘If so many people are going, they won’t need me.’

That there is a fundamental difference between the practice of science and the practice of commonsense can also be concluded from a thought experiment described by Apel (Apel 1976, 172-173; 1979a, 275). The basic idea is as follows. Suppose a group of psychologists investigates how certain stimuli from the environment are causal factors that make agents believe certain things and behave accordingly. Then it is basically possible that these psychologists also consider what their research colleagues do as causally ‘motivated’ by those environmental factors. Let us suppose that one of the members of the research group really does so, and that she takes the way her colleagues in the investigation develop their scientific beliefs as causally determined by relevant stimuli from the environment. Then this psychologist does keep to the scientific theoretical practice in the sense that she (let us suppose) draws valid conclusions on the base of a valid theory according to scientifically accepted norms, but it will be clear that by taking what her colleagues do as causally determined the psychologist can no longer have a communicative relation with them.6 Apparently the scientific practice presupposes another non-theoretical practice in which colleagues are not conceived as objects of a theory but as co-subjects.

This is what we can conclude from these cases. The occurrence of Merton effects shows that the practice upon which the development of scientific theories is based is different from the practice of our commonsense conception. In the first practice, the subjects of knowledge, who are the bearers of the practice at the same time, are separate from the objects of knowledge. In the commonsense practice, the ‘objects’ are at the same time co-subjects and therefore also co-bearers of the practice. What the case of the investigating psychologists shows is that the practice of the commonsense conception precedes the practice of scientific theory.

Meaning assignment takes place each time we approach reality. This was what Hesse concluded. This way of assigning meaning (assignment of meaning 1) is reflected in a certain practice. Churchland was right in arguing that. Our commonsense conception, however, is not opposite to a reality to be approached and it does not precede the reality (in Popper’s sense that we must first have a theory before we can investigate the reality). Our commonsense conception is a part of the reality. As such it is continuously confronted with other commonsense conceptions. Therefore, assignment of meaning 0 is not a one-sided process, as it is in the case of theoretical interpretation where the scientist is opposite to the reality, and where the subject of knowledge is opposite to the object of knowledge. It is a process of mediation with other commonsense conceptions of other subjects of knowledge. Is the person I called ‘unreliable’ (and others called ‘absent-minded’ or ‘playful’) because he fails to keep his appointments really unreliable? Perhaps he simply has other views about what an appointment is because he has a different cultural background. The confrontation of both commonsense conceptions requires mediation in order to understand each other and in order to make clear what the problem is and how it must be solved. More exactly, it requires mediation between the two frames of understanding. But this example also shows what mediation between commonsense conceptions is: the way that living together takes place and being social comes about. In other words, what is called ‘folk psychology’ – i.e. the explaining, predicting, manipulating, controlling, etc. of other people on the basis of our beliefs and desires and such like – is by its interaction with the ‘folk psychology’ of other people the foundation of social reality.

What this mediation is, is well expressed by the German word ‘Verständigung’, the word employed by Apel for this purpose. Verständigung encompasses, as Apel explained to an English-speaking audience, ‘both the understanding of meaning and the coming to agreement’ (Apel 1979b, 7; italics HbdW). Both aspects are important for the process of mediation that the assignment of meaning 0 on the basis of our commonsense conception is, for as Apel said just before the quotation: ‘Children could not learn the meaning of words if they did not learn to use them at the same time as coming to agreement with other members of the language community about paradigmatical evidences of experience and about paradigmatical rules of behaviour’ (ibid.). And I want to add that this, of course, applies not only to children, but also to adults.

3. The commonsense conception and its relation to scientific theory

Science holds its position within the reality constituted by the mediation between different commonsense conceptions. By mediating between our commonsense views and those of other people, Verständigung induces a certain practice: the commonsense practice. This practice includes the beliefs and desires that determine our daily actions and all our other views about our daily reality that found what we do: what food is, what going to school is, what being polite is, what a question is, what working is, what art is, what justice is and also what science is. Within and also on the basis of this practice, certain subpractices can split off, e.g. art, law and science. Each of these subpractices has norms and rules of its own for its particular activities and constitutes its own subreality, e.g. painting, jurisdiction, and scientific research. Characteristic of the scientific activities is the pursuit of abstraction in the form of objectification by the application of certain methodical rules. Or formulated conversely: certain social activities (e.g. painting, jurisprudence and science) require certain rules and norms that form subpractices on the basis of the general commonsense practice. Verständigung, and therefore the commonsense conception, are the precondition of the scientific practice and other subpractices. These subpractices are then the connection between the commonsense activities and particular activities. The interface between the general commonsense practice and the subpractices can be either more or less smooth or rather strict. Characteristic of the scientific theoretical practice is that it is rather strictly separated from the commonsense practice, though this does not alter the fact that the former supposes the latter (see Figure 1).



level 1



scientific theory (and research)







Level where the scientific interpretation of reality takes place (assignment of meaning 1)



level  0 

and level 1




‘Staircase to science’, where the norms and methodical rules for approaching the reality and for what counts as valid knowledge, etc. are determined


level  0


commonsense conception (daily practice)


Ground level, where the daily interpretation of reality and Verständigung take place (assignment of meaning 0)


Figure 1: The relation between the commonsense conception and scientific theory


Churchland argues that ‘both the content and the success of FP [folk psychology] have not advanced sensibly in two or three thousand years. The FP of the Greeks is essentially the FP we use today, and we are negligibly better at explaining human behavior in its terms than was Sophocles’ (Churchland 1992, 8). As we have seen, Churchland has a misconception about what folk psychology (or the commonsense conception) is. It is not a theory from which predictions can be deduced that can falsify the theory, and it is not a theory that can be replaced by a better one. Moreover, it is not true that the ‘folk psychology’ of the Greeks is essentially the same as ours. Ancient Greek society was a class society that distinguished between free citizens and slaves. Among the former, only adult males had the right to vote (at least that was the situation in Athens). This society was very different from today’s Western European society. Much of what belongs to our present commonsense conception would be completely useless in ancient Greek society, not only because the social organization has changed but also because many daily social conventions are now entirely different. And should we not consider it to be progress ‘in the content and the success’ that today human rights are generally accepted and applied to everybody irrespective of class, sex, origin, etc. (however poorly they may be observed), while in ancient Greece they (or what counted as human rights in those days) obtained only for free citizens? And what about the feudal system? Were not even then the rights different according to which estate one belonged? Is it not to be called progress that there are generally valid human rights? And is this not true irrespective of whether the commonsense conception is conceived as mediation or as a theory?

That the commonsense conception has to be conceived as mediation does not mean, however, that the process of mediation and the results of mediation (commonsense views, beliefs, desires, etc.) cannot be the object of study. I do not deny that there cannot be a commonsense theory – or, if you like, ‘folk psychology’ – but it is different from what Churchland, Jackson and others have in mind. The theory I mean is not the one of the bearers of the commonsense conception (the ‘folk’), but the theory that has this commonsense conception as object and tries to investigate it. And then, when we approach the commonsense conception theoretically as object, just as before we must distinguish between on the one hand the theoretical practice and on the other hand the theory proper and the research.

A good example of commonsense views taken as objects of research is Willis 1978. Willis shows in his study why a group of pupils calling themselves the ‘lads’ fails at school by investigating not the objective characteristics of the lads, but the counterculture as worded by the lads themselves and as expressed in their actions, and by investigating how this counterculture collides with the dominant views of the school and the teachers who represent those views. Willis shows how the lads have their own purposes at school, which deviate from the official purposes of the school and those of the conforming pupils, and makes it clear how the lads develop ways of acting which they use as instruments for achieving their purposes. He goes on to show how it is possible to fit the ways of acting and the views of the lads in a theoretical frame with the help of such concepts as ‘penetration’ and ‘limitation’, and how it is possible to characterize them that way and make them understandable. However, such a theoretical analysis is not an analysis the lads themselves use to describe their situation – and never will as long as they remain ‘lads’ – although the theoretical analysis starts concretely from the lads’ culture and is an expression of that culture. Thence, we cannot say that Willis’s analyses are the commonsense views of the lads: his analyses make them the object of these analyses. In short, Willis analyses the beliefs (the counterculture) and desires (purposes) of the lads and the actions that result from them, and fits them in a theoretical frame. In this way, what the lads do and why they do it is objectified.

Examples of the practice on which the theoretical approach to commonsense views as object is based are von Wright’s practical syllogism (von Wright 1977) and what I call the belief-desire model (BDM) as developed by Davidson. The authors develop what beliefs and intentions (von Wright) and desires (Davidson) are, show how they are related to actions resulting from these beliefs and desires or intentions, and go into the presuppositions that are often implicit or only vaguely indicated. I shall take a look at the BDM.7

Davidson does not take notice of the beliefs and desires of particular persons and how these result in actions, as Willis did in his investigation of the lads, but raises the questions what counts as a belief, what counts as a desire, what counts as an action, and what the nature of the relations between belief, desire and action is. For example, in his "Action, Reason, and Causes"8 Davidson develops the following postulates (Davidson 1989,4):

‘1. In order to understand how a reason of any kind rationalizes an action it is necessary and sufficient that we see, at least in essential outline, how to construct a primary reason.

  2. The primary reason for an action is its cause.’

In this quotation, a ‘primary reason’ is the agent’s ‘pro attitude’ (e.g. a desire) and an accompanying relevant belief. Davidson then shows that what the agent’s reasons are depends on the way what the agent does is described; that acting for a reason implies acting intentionally; that the agent’s intention must be known in order to know what his primary reason was; and how it is to be conceived that a primary reason is the cause of an action. As for the last point, Davidson argues that it is not necessary to know the laws connecting the primary reason and an action in order to be able to give a valid explanation (id., 4-19). In "Agency", Davidson reasons that there is a difference between what somebody does and what happens to him or her, and he further develops his ideas about causation. In "Intending", Davidson discusses the concept of intention and shows that, for example, it is not necessary that an intentional action is preceded by an intention.

Concepts like ‘action’, ‘primary reason’ and ‘cause’ as employed by Davidson are fundamentally different from concepts like ‘penetration’ and ‘limitation’ as employed by Willis. The latter are used by Willis in order to interpret particular ways of acting and to characterize them theoretically. On the other hand, a concept like ‘action’ (as opposed to what happens to someone), for example, as used by Davidson, is a concept that indicates what belongs to the frame of an analysis of action. It is used for making explicit the praxis of a theory of action. What usually is implied by Davidson, though prominently present in the background, is that an analytical approach to action always deals with the agent’s beliefs and desires.9 Beliefs and desires are not objective motives that a researcher ascribes to an agent on the basis of a psychoanalytical examination and that were unknown to the agent, but are reasons as worded by the agent himself or that can be worded on being asked. Generally, it is a feature of an approach that seeks the agent’s commonsense conception that the views as conceived by the agent himself are central. This, what I want to call, ‘commonsense approach’ takes its departure from the agent’s perspective. On the societal level, I am thinking of those approaches that analyse the culture and habitus as they are or can be worded by the members of the society. As such, a commonsense approach is opposite to one that looks for the unconscious motives and (on a societal level) the structural forces that, according to the analyses made by the researcher, objectively move agents or objectively bring about social developments and take place ‘behind the backs of the agents’ (and that can be known only by the agents if they are pointed to them by a third person or if they themselves take the third person’s viewpoint). Such an approach is based on the researcher’s perspective.

The agent’s perspective cannot be reduced to the researcher’s perspective; it has its own methodological problems. I shall skip these problems here10, but in this context it is important that being a specific approach, the approach that is based on the agent’s perspective has a practice of its own, alongside though not unconnected to the approach based on the researcher’s perspective. But the commonsense practice is not only the object of the former approach: as a scientific approach, it presupposes the commonsense practice just like any other scientific approach (see Figure 2).




level 1


agent’s perspective:

theory (and research)


agent’s perspective:




theory (and research)





Level where the scientific interpretation of reality takes place (assignment of meaning 1)




level  0

and level 1


‘Staircase to science’, where the norms and methodical rules for approaching the reality and for what counts as valid knowledge, etc. are determined


level  0


commonsense conception (daily practice)


Ground level, where the daily interpretation of reality and Verständigung take place (assignment of meaning 0)


Figure 2: The relation between the commonsense conception and scientific theory (improved version)

Thus, there are two ways we can approach our objects scientifically: from the agent’s perspective and from that of the researcher; or as a person or as a thing. Which approach is adequate depends on the nature of the object and our scientific questions, and often we have no choice. Inanimate objects do not have subjective viewpoints; actions (as actions, thus opposite to behaviour) can be grasped only from the agent’s point of view.

By making a distinction between the agent’s perspective and that of the researcher, I appear to be taking a dualistic standpoint. However, it is no more dualistic than the supposition that mind and body are two aspects of man, even though the two are certainly connected. To call that dualism shows a lack of philosophical imagination, i.e. an inability to see commonsense (‘Verständigung’) as the foundation of action. What is then neglected is that the agent’s perspective and that of the researcher are not unconnected to each other, but are related by means of the commonsense conception. Science is what counts as science. That is no datum but what science is is determined by subjects that are guided no less by their scientific and rational decisions than by their commonsense views, as has been made clear by such studies as Latour 1987: science is a social process based on commonsense just like any other social process. On the other hand, the commonsense views on which science is based are not immune to what is brought about in science. Scientific rationality and scientific results are not isolated domains. Beliefs about what rationality is and results from science have become part of our commonsense conception and affect daily life and how we motivate our actions. One only need look at what has happened in the past few decades in the field of education. And that again is reflected in the development of theories from the agent’s perspective. This in its turn has an influence on commonsense, for example, by using the agent’s perspective to make people aware of their motives for action (which are partly affected by science), in case a problem arises or at another moment. And these changed commonsense views can have repercussions on the practice and theory in the researcher’s perspective.

4. Conclusion

With the help of the distinction between the agent’s perspective and that of the researcher, I can now answer the question whether or not mental explanations (BD explanations) are nothing but physical explanations with different variables, which was the starting point of this article. In fact, it is not important whether the variables in an explanation are mental or physical; that is only a secondary problem. What is essential is whether one wants to explain from the viewpoint of an acting person – i.e. from a first-person perspective – or from the viewpoint of an observer, i.e. from a third-person perspective. That is, what matters is whether we are interested in the meaning given by an agent himself to his actions (meaning 0) or that given by the researcher (meaning 1). And this distinction implies that mental explanations are not physical explanations, but that each has a logic of its own.11 It also determines what the relation is between the two kinds of explanation, and generally what the relation is between the commonsense conception and scientific theory.


Apel, Karl-Otto (1976), "Causal explanation, motivational explanation, and hermeneutical understanding. (Remarks on the recent stage of the explanation-understanding controversy)", in Contemporary aspects of philosophy, Gilbert Ryle (ed.), Stocksfield etc.: Oriell Press etc.

Apel, Karl-Otto (1979a), Die Erklären:Verstehen-Kontroverse in transzendentalpragmatischer Sicht, Frankfurt am Main: Suhrkamp Verlag.

Apel, Karl-Otto (1979b), "Types of social science in the light of human cognitive interests", in Philosophical disputes in the social sciences, S.C. Brown (ed.), Brighton etc.: Harvester Press etc., 3-50.

Baker, Lynne Rudder (1999), "What is this thing called ‘Commonsense Psychology’?", Philosophical Explorations 2 (1), 3-19.

Churchland, Paul M. (1992), A neurocomputational perspective. The nature of mind and the structure of science, Cambridge, Mass. etc.: The MIT Press.

Churchland, Paul M. (1998a), "Folk Psychology", in Churchland/Churchland, 3-15.

Churchland, Paul M. (1998b), "Theory, taxonomy, and methodology: A reply to Haldane’s ‘Understanding Folk’ ", in Churchland/Churchland, 17-24.

Churchland, Paul M., Patricia S. Churchland (1998), On the contrary. Critical Essays, 1987-1997, Cambridge, Mass., etc.: The MIT Press.

Davidson, Donald (1989), Essays on actions and events, Oxford: Clarendon Press.

Dretske, F. (1988), Explaining Behavior. Reasons in a World of Causes, Cambridge, Mass., etc.: The MIT Press.

Giddens, Anthony (1977), Studies in social and political theory, London: Hutchinson of London.

Habermas, Jürgen (1982), Theorie des kommunikativen Handelns. Band I. Handlungsrationalität und gesellschaftliche Rationalisierung, Frankfurt am Main: Suhrkamp Verlag.

Hesse, Mary (1980), Revolutions and reconstructions in the philosophy of science, Brighton: The Harvester Press.

Jackson, Frank (2000), "Psychological explanation and implicit theory", Philosophical Explorations 3 (1), 83-95

Latour, Bruno (1987), Science in action, Cambridge, Mass.: Harvard University Press.

Stoutland, Frederick (1976), "The causal theory of action", in Essays on explanation and understanding. Studies in the foundations of humanities and social sciences, Juha Manninen, Raimo Tuomela (eds.), Dordrecht etc.: D. Reidel Publishing Co., 271-304.

Weg, Henk bij de (1996), De betekenis van zin voor het begrijpen van handelingen, Kampen: Kok Agora.

Weg, Henk bij de (1999), "Dretske and the causality of reasons", on website: (English) and (Dutch).

Willis, Paul E. (1978), Learning to labour. How working class kids get working class jobs, Westmead, Farnborough: Saxon House.

Wright, Georg Henrik von (1977), Explanation and understanding, Ithaca, N.Y.: Cornell University Press.

Wright, Georg Henrik von (1983), Practical Reason. Philosophical papers. Volume I, Oxford: Basil Blackwell.



1In this article the word ‘mental’ refers exclusively to propositional attitudes and explicitly not to such sensations like pain, grief, fear etc.

2In this article I do not make a difference between these descriptions.

3Thus excluding the humanities and arts and their theories.

4For my criticism on Dretske, see my 1999.

5See my 1996, IX.6.3

6For an elaboration of this point, see Apel 1976, 172-173.

7An extensive discussion of von Wright’s practical syllogism can be found in my 1996.

8See Davidson 1989 for this article and the other articles to which I refer.

9Explicit for example in 1989, 4

10See my 1996 and cf. Apel 1979a.

11See my 1996 for an exposition of the logic of mental explanations, or rather of ‘action explanations’ as they are called there.


terug (Nederlands)

back (English)