terug (Nederlands)

back (English)

                                                                                              Table of contents in Dutch



The book can be obtained for 9,90 euro (plus postage) by sending
an e-mail to the author:



Henk bij de Weg

De betekenis van zin voor het begrijpen van handelingen, Kok Agora, Kampen 1996.   

(English title: Meaning, action and understanding)


In studying what people do two points of view can be distinguished: We can choose the perspective of the actors themselves (the actor’s perspective), or we can look at what is going on from the outside, from a distance (the researcher’s perspective). Regarding the relation between both points of view three standpoints have been defended:

– both perspectives exclude each other;
– they complement each other;
– one perspective can be reduced to the other one.

The question which of these standpoints is right, has led to a long lasting controversy, the so-called "Controversy on Explanation and Understanding" (CEU). This controversy forms the background of the present study. However, unless it is useful in some way, I do not get into the arguments for the different positions, nor is it my intention to try to disprove those with which I do not agree. It is a tenet of this study that there are two points of view to look at what people do and that both points of view complement each other:

– by taking the researcher’s point of view we can explain what people do;

– by taking the actor’s perspective we can understand what people do.

While the explanation point of view has been elaborated in an extensive methodology, this has not been the case with the understanding point of view. Moreover, the discussion on the understanding (or "verstehen") point of view has stagnated. It is therefore the main objective of this book to reflect on the foundations of understanding what people do, their actions, in order to make a contribution to the progress of the methodology of understanding actions.

I do not start my reflections from scratch. The starting-point of my analyses is Von Wright’s practical syllogism. The reason is not only that the discussion stagnated here, but also that in its succinctness and explicitness this practical syllogism shows the essence of the verstehen point of view. Besides that I pay much attention to the scheme for the rational explanation of actions developed by Schwemmer. This scheme is (partly) a further development of Von Wright’s practical syllogism. It is specially made for the social sciences, which are central in my analyses as well.

I do not begin my analyses, however, with a discussion of both schemes. First of all a theoretical and methodical frame has to be made. After the introduction in chapter I, I begin with the development of such a frame in chapter II. In the early eighties Habermas made the remark that the issue, whether there are different methods for the social sciences on the one hand and the natural sciences on the other, had been solved. At the same time Apel (who’s way of thinking about science in many respects agreed with Habermas’s) stated that the discussion about this issue had started all over again. Why this difference in opinion? This becomes clear as soon as we consider what both authors mean by "verstehen". Then we see that, at the time both authors made their remarks, "verstehen" meant for Habermas something else than Apel. In this light both authors are right. It is my contention, however, that there are not just two but three kinds of verstehen. I have called them respectively "ontic verstehen", "theoretical verstehen" and "verstehen as a method".

In chapter III is explained what the three concepts stand for. How we must understand the concept of ontic verstehen has been clearly formulated in Giddens’s theory. For Giddens understanding is the mediation of frames of meaning of different people, or what is more, of people with a different background or culture. This is exactly what ontic verstehen means. Giddens calls this understanding a condition of human society, but I think it is better to regard it as the way of being of human society. This mediation of frames of meaning is what in German is called "Verständigung", or, as Apel explains it, "both the understanding of meaning and the coming to agreement".

Even though verstehen as ontic verstehen is the way of being of human society, it is neither necessary to make it a starting-point of an analysis of human reality nor to give explicitly expression to it in such an analysis. If we keep away from doing this, while using a theoretical framework with concepts like "cause" and "behaviour", then we explain what people do. However, in case we do take the ontic verstehen as a point of departure for our analysis of what people do, while using a theoretical framework with concepts like "reason", "meaning" and "action", then verstehen is a theoretical perspective. I have called this "theoretical verstehen".

Classical examples which use verstehen as a theoretical perspective are the "verstehende Soziologie" as part of Weber’s analyses and the work of Schütz. I use both sociologies not only to illustrate the theoretical verstehen but also, and this goes, especially for the sociology of Schütz, to develop my own theoretical framework. After having analysed Weber’s well-known definition of "sociology", I conclude that he recognises a reality that is meaningful to the people participating in it and that he uses this meaningfulness for his analyses of what people do. By quoting a fragment of Der Aufbau der sozialen Welt I show that this is Schütz’s approach, too. Schütz’s book also provides me with some concepts which play a key role in my analyses of the works of Von Wright and Schwemmer and in the development of my own theory and scheme for understanding action. For Weber the difference between action and behaviour is that action has meaning while behaviour has not. According to Schütz this is not correct. He states that action as well as behaviour have meaning, both being a form of experience for the subject. The difference is that action is planned while behaviour is not. I do not agree with Schütz that behaviour is a form of experience, as becomes apparent in chapter IV, though I accept that the difference between action and behaviour is the planned character of action.

Weber distinguishes between "actual understanding" and "understanding out of a motive", but according to Schütz actual understanding is subjective and not useful to an "understanding" sociology. Again I do not agree with Schütz, for the distinction made by Weber points to a basic distinction in understanding action, i.e. the one between the understanding of the meaning of an action and the understanding of its purpose. This distinction has proven to be very important for the method of verstehen, together with the one made by Schütz between "because-motives" and "in-order-to-motives". A because-motive is, according to Schütz, an actor’s coherence of experiences from the past which are relevant to an action, an in-order-to-motive is his or her expectations about the future while acting. In my opinion Weber’s and Schütz’s distinctions are important, for in conjunction they indicate the three basic questions for a method of understanding of action: 1) the question what an action is; 2) what an action is for; and 3) why an action is performed.

In the last part of chapter III I do not describe in detail what verstehen as a method is; this is done in the following chapters. However, if verstehen is a method, what does it mean that a certain approach is a method? It is surprising that a good definition of "method" does not exist. For this reason I have developed a definition myself: a method is a set of logical and philosophical principles and rules for correctly answering the central question of a theoretical framework of analysis in order to acquire scientific knowledge, in so far as these rules and principles refer to the structure of the answer of the central question. Because this definition is vague in some respects and leaves room for arbitrary interpretation, I have also developed five criteria of method, which are treated in extension in chapter IX.

If there is a difference between the explanation and the understanding or verstehen of what people do, what then is this difference? This is the central question of chapter IV. A standard answer in the CEU was put in terms of the structure of both methods. But is the essence of the difference between both methods really to be found in their structures, or rather, as usually understood, the causal respectively non-causal character of each method? In order to answer this question I have looked at the objects of both methods, behaviour in the case of explanation, and actions in the case of understanding. Both objects have sense, but the difference between an item of behaviour and an action is that an action does not only have sense for its observer (sense 1) but also for the subject of the action, who puts a sense in it (sense 0). It is due to this "double hermeneutic" (Giddens), or, as we can also put it, this incorporation of a sense 0 in a deed, so an action, that the understanding of an action is principally different from the explanation of an item of behaviour. Its non-causal structure does not account for the difference. When we try to understand what actor does, it is the (subjective) sense of the action for actor we want to get to know and not its (objective) sense for the researcher. In trying to explain an item of behaviour it is the other way around.

In chapter V I return to the question what the relation is between explanation and understanding. Apel has defended the thesis that explanation and understanding are complementary forms of knowledge. As leading knowledge-interests they supplement each other, exclude each other while they cannot be reduced to each other. In principle I agree with Apel, but in his analyses he makes a serious mistake. On the one hand he puts verstehen (as a method) and the method of explanation side by side. On the other hand he claims that verstehen (as communicative understanding) precedes the method of explanation. In this view the methodical understanding is the scientific continuation of the prescientific communicative understanding. This is not correct. The kind of verstehen that precedes the method of explanation is something different than the method of verstehen that is used for understanding human action. It is the communicative understanding that justifies the possibility of understanding as a method. I have called it "ontic verstehen"; in German it is called "Verständigung". After having disentangled this confusion in Apel’s description of his complementarity-thesis I develop a new, three-dimensional model for the relation between explanation and understanding.

Using the theoretical and methodological framework I developed thus far I begin with an analysis of Von Wright’s practical inference for "explaining" (i.e. understanding) action (chapter VI). Is it really possible to understand ("explain") an action with this syllogism, as Von Wright pretends? I think it is not, for if we fill out this scheme with an example, we know what actor acted for, that is we know his or her intention, but we do not know why actor acted in the way he or she did: we do not know his or her reasons for acting. In terms of Schütz, we know actor’s plan in acting, his or her expectations about the future, be we do not know actor’s past experiences which made him or her to act the way (s)he did. Von Wright’s practical syllogism is incomplete, because it does produce only a partial understanding of actor’s reasons. It therefore has to be extended.

Nevertheless, even if we know actor’s reasons for acting and his or her intention in acting, it may still be true that we do not understand the action in question. Maybe the action is cruel or actor had a choice between different actions and we do not know why (s)he chose the one (s)he actually did. Therefore we must also know the meaning an action has for actor. More in particular we should know whether and why the actor thought the action to be appropriate, or in case actor had the choice between different actions, which action actor considered to be most appropriate and why. This shows that it is also important to ask what an action is for actor. Von Wright asked for the meaning of an action, but he did not see it in terms of its appropriateness. Moreover he used a separate syllogism for "catching" the meaning. In my opinion it is better to integrate this syllogism for the understanding of meaning in the scheme for understanding actor’s reasons and intention. If we add propositions for the understanding of actor’s reason for and meaning of an action to the practical syllogism as described by Von Wright, we can construct what I have called a scheme for the understanding (or verstehen) of actions.

Von Wright developed his theory and method for the "explanation" (understanding) of action mainly in his Explanation and Understanding. Later he partially reconsidered what he had written in this book. This change of view could have consequences for my theory and method of verstehen developed so far. In chapter VII I therefore analyse parts of Von Wright’s later theory of action. One modification in his later work is that he now allows for different variants of his practical syllogism. This is no serious threat for my scheme, because it already implies the possibility of variants. More important is that Von Wright now distinguishes between external and internal reasons. An internal reason is what he used to call an intention. An external reason is a symbolic challenge like an order or a request. In this chapter I do not deny that this distinction makes sense, but I doubt its usefulness from a methodological point of view. In an extended analysis I show that the distinction between an external and an internal reason does not affect the structure of my scheme for the understanding of action as developed in the preceding chapter. An external reason in the sense of Von Wright can be either what I have called a reason or an intention and this is why the distinction between external and internal reasons is methodically not relevant.

In his later work Von Wright realised, he says, that there is a difference between "to intend to do" and "to do intentionally" and that it cannot be said that we intend to do everything we do intentionally. I agree with Von Wright in so far as we do not always have conscious intentions in acting, but I think his remarks do not hit the essence of the method of understanding action, even though this method explicitly asks for the intention in acting. The reason why we ask for actor’s intention is not that we assume that (s)he has an explicit intention while acting; we ask for his or her intention because we assume that under normal circumstances actor can make his or her intention explicit, even though at the moment of acting such an intention was implicit and not consciously formed.

In chapter VIII I turn to the theory and scheme for the rational explanation of action developed by Schwemmer. However, I begin this chapter with a discussion of Schwemmer’s later theory of action. The reason for this is that Schwemmer later severely criticised schemes for the understanding of action of the kind he himself had developed before. The essence of his criticism is that a rigid scheme cannot represent the fluent stream that acting is. I reject this criticism, because it confuses representation and approach of what actors do. We represent what people do by means of a theory, we approach it by means of a method. Schemes such as those developed by Schwemmer or by myself are condensed forms of such methods.

However, the refusal to reject Schwemmer’s scheme for the rational explanation of action on fundamental grounds does not necessarily mean that this is a correct scheme for understanding action. Although it is in some ways better than Von Wright’s practical syllogism, I have four arguments to think it catches actor’s sense of an action in a imperfect manner. Firstly, it neither allows for the possibility that actor has several actions to choose from nor does it incorporate the question what an action is for actor. Secondly, Schwemmer claims that his scheme is an improvement of Von Wright’s practical syllogism, but in fact he replaces the question for actor’s intention by the question for actor’s reasons. Moreover, because Schwemmer considers it impossible to catch actor’s subjective purposes in action he asks for actor’s "objective" purposes instead, objective purposes being those that can be formulated in a theoretical language. At that point, and that’s my third argument, he is no longer understanding what actor does, but explaining it. What I finally do not agree on is Schwemmer’s contention that in social sciences a deductive-nomological way of explaining what people do is not possible or at best within temporary limits. In chapters IV and V I explained that there are two ways of analysing what people do and that these two ways cannot be reduced to one another. This is also true for the social sciences.

In chapter III I developed five criteria for a method. In chapter IX I ask how they are satisfied by the method of verstehen. The so-called cognitive schema-theory of reading comprehension is very important in the discussion of several criteria. So, after having shown that actions can be analysed as texts, I give a short summary of relevant points of this approach.

The first criterion is that a method must indicate what it means that we can say that we have acquired valid knowledge, or rather, in our case, what it means that we can say we have understood an action. According to the schema-theoretic view comprehending means to have a scheme that one can apply to a relevant situation. So for an actor we can say that (s)he comprehends his or her action if (s)he has a scheme describing the action. Our methodological scheme for understanding action tries to describe that same action. Therefore I conclude that we can say that we understand an action if the methodical scheme corresponds with actor’s scheme for the action in question.

The second citerion asks for the conditions of application of a method. Taylor described three such conditions for the understanding of an action, which I partly modify and supplement with a fourth condition. The essence of the conditions is that the method of understanding can only be applied when dealing with, what Taylor called, "subject-referring properties".

The third criterion asks for the relation between the subject and object of knowledge. Habermas describes the attitude of the subject of knowledge to his or her object as "performative". Consequently the object of knowledge is not merely an object but at the same time a subject. For Habermas the performative attitude implies that the researcher as a subject of knowledge not only describes actor’s reasons but also evaluates them. This view is not tenable, however, as it confuses the sense the subject of knowledge (researcher) gives to an action with the sense actor gives to it.

The fourth criterion asks for the generalizibility of knowledge acquired by a method, in this case the method of verstehen. Often the method of verstehen is seen as idiographic and the method of explanation as nomothetic while verstehen is regarded as subjective and explanation as objective. Although as a matter of fact the objective character of generazibility does pose a problem when trying to make theories for understanding what people do (cf. Bourdieu), the dichotomy does not take into account that actors, too, have generalising views which (can) guide their actions. Moreover, such generalising views are often not individually defined but characteristic for a group or even for a whole society. In terms of the cognitive schema-theory such generalising views are schemes that formulate for actors what to do in situations which are more or less like those schematically formulated, and as I just said, such schemes can be individual or shared by more people. If this is true then we can also make generalizing "understanding" theories, for such theories are nothing else than schemes developed by a researcher that tries to catch – the individual or shared – abstract, generalising views (i.e. schemes) used by acting people.

The fifth methodical criterion asks for the nature of the relations in a methodic scheme, in this case a scheme for the understanding of actions. Traditionally the question about the nature of the relations in a methodic scheme is answered in terms of their causal or logical character. It is obvious that the nature of the relations of the scheme I have developed in chapter VI cannot be causal, since causality refers to the way a researcher perceives these relations; under normal circumstances it cannot be said that an actor perceives his or her actions as caused by his or her reasons etc. An actor does not make reasons from which the actions follow automatically (as experienced by the actor); an actor has reasons and acts on account of those reasons.

A more fundamental problem, though, is the following: when we ask for the nature of the relations in my methodic scheme what exactly does that mean? In this scheme we do not have just one single relation, as for instance in Von Wright’s practical syllogism, but it is complex and contains several relations. Maybe we can disentangle those relations and then call the relation between intention and action in a way verificational (cf. Von Wright). However, this cannot mean that an actor has to act, once (s)he has an intention, because the verificationality of the intention-action relation is methodical and it exists for the researcher but not necessarily for the actor. As De Boer and later Von Wright showed, an intention does not always have to be followed by an action. And for actor’s reasons we cannot even say that. An actor can have reasons for an action, but this does not imply that actor will act on account of these reasons. And an actor can give a meaning to an action but this fact in no way determines its execution. My conclusion is accordingly that the nature of the relations between the "determinants" of an action and the action itself is contingent.

However, if the relations in a scheme for the understanding of actions are contingent, then such a scheme cannot be used for predicting actions. Of course we can fill out a scheme, predict that an actor will act in such and such a way and, if the actor acts accordingly, conclude that we filled out our scheme correctly and that the underlying theory has not been falsified. But what if actor does not act as was predicted? Then two things are possible: it may be so that we did not fill out the scheme correctly and that our theory is false. But it is also possible that we have independent information that makes the truth of both our theory and scheme probable. In that case our conclusion must be different: not our theory and scheme are false but actor has changed his or her opinion. The coming true of our prediction is not a measure for the correctness of our theory and scheme but a criterion of whether or not an actor has changed his or her reasons, intention and meaning.

In my last chapter (chapter X) I look back upon the previous ones. In a short summary I demonstrate how I answered the question how the method of verstehen is possible. I answered it in two ways, firstly by showing how we can catch the subjective self-interpretation of actors and secondly by showing how we can do this methodically. In this chapter I also refer to the relations between the understanding of actions and ethics and the philosphy of mind. If there is, as I maintain (unlike for example Dennett or Churchland), a difference between a first person view and a third person view, then my study shows how it is possible to understand, and not to explain, actions such as a murder, the posting of a letter, or the Battle of Heiligerlee.

terug (Nederlands)

 back (English)