1

Aulis Aarnio
Professor, Dr.
The Tampere Club, Finland

The Hermeneutic Circle of Legal Realism

Some Thoughts on Alf Ross and the Legal Science

Alf Ross (1899 – 1979) was, without a doubt, one of the greatest European legal

thinkers of the 20th century. He was like a tall oak, in the shadow of which only a

few smaller plants could persist. In the Nordic countries he is on an equal

standing with Axel Hägerström, whose philosophy also provided Ross with his

first intellectual home. In this spirit, Ross took a strong stand on the normative

nature of morality (and valuations) in his writings of the early 1930’s. According to

him, things such as moral obligations are only expressions of a sense of

duty.Because of this background, we have good reason to connect Alf Ross with

the intellectual legacy of so-called Scandinavian realism.

Still, to understand Ross and his contemporaries, we must remember that Ross

also wrote in the pragmatic tradition beginning in Denmark from Oerstedt and

was carried on in the works of, for instance, Ussing and Vinding Kruse. When

seen like this, Alf Ross was a deeply Nordic thinker, and connecting him to, for

example, American realism (specifically: pragmatic instrumentalism) is a

complete distortion of the background of his learning and his basic legal scientific

commitments.

As regards Alf Ross and his thoughts on the doctrinal study of law are
 2

concerned, we must be careful to separate his “two persons”. The first, is the

analytical Alf Ross, whose findings were in Nordic countries elaborated by the

so-called analytical school of the doctrinal study of law. The second “person”,

under closer scrutiny here, is the (legal) realist Alf Ross. This side of Ross’

thinking has never really found its place in Nordic countries.

What Alf Ross was interested in as a realist was the nature of the scientific

doctrinal study of law. In order to be a part of science, the statements of it should

(at least in principle) be true or false in the same way as those of science in

general. Ross himself formulates his scientific program as follows:

“The leading idea of this work is to carry, in the field of law, the empirical
principles to their ultimate conclusions. From this idea springs the methodological
demand that the study of law must follow the traditional patterns of observation
and verification which animate all modern empirical science; and the analytical
demand that the fundamental legal notions must be interpreted as conceptions of
social reality, the behavior of man in society, and as nothing else”.

Alf Ross was not a realist only due to the influence of Axel Hägerström, even

though Hägerström’s shadow was cast on the young Ross. Actually, Alf Ross’

path to Hägerström went through Hans Kelsen. Ross’ book ”Theorie der

Rechtsquellen”, published in German, is strongly marked by Hans Kelsens’

thoughts, even to the point that Ross´s countryman Vinding Kruse, among

others, criticized him for being too Kelsenian. It is apparently this very criticism

that led to Ross finishing his dissertation in Uppsala under the guidance of

Hägerström.
 3

This is also how Ross was drawn to the camp of Scandinavian realism, even

though he did this in a characteristic way. It is also interesting to note that in the

introduction to the work Theorie der Rechtsquellen (The Theory of the Sources of

Law) Ross actually disassociates himself from the central arguments of his book

and, at the same time, from the whole Kelsenian thinking. The introduction is

thus something of a pathway from the young Ross to a more mature character.

The most significant and influential work Ross published on the theory of the

doctrinal study of law is ”On Law and Justice” (Om ret og retfaerdighed).

Aleksander Peczenik as my collaborator, I have tried to make some corrections

to these inaccuracies in Ross’ favor, so this theme will not be discussed at any

greater length in this context. However, in this contribution the original text has

been used to as large extent as possible.

”On Law and Justice” is no longer a pure example of the Uppsala school. Its

author has also seen the phase of logical empiricism and the so-called Vienna

circle, which is why the work must be evaluated against the scientific background

of its time. Before we get better acquainted with the realist-Ross of this phase,

there is reason to make a brief excursion into legal theory in general.

The words ”legal theory” refer literally to theory about law. When interpreted like

this, legal theory can be (1) theory about legal systematics, that is, the structures

norms, and concepts), (2) theory about the application of law (adjudication), (3)

thery on legislation, or (4) theory about legal research (the doctrinal study of law).
 4

Ross often talked about legal philosophy instead of legal theory. This concept

should not be taken verbatim. In current terms, Ross’ thought was purely about

legal theory, not about a part or area of general philosophy. Seen from this point

of view, legal philosophy could be, for example, philosophy about the origin of

law, its foundation, its social role or other basic questions like this. This is only to

refer to the conversation about whether philosophy is a science and about the

role it has if it is not.

In this, he is similar to the later Ludwig Wittgenstein who emphasized that for him

it was not important to practice philosophy, but to analyze philosophically

interesting problems. Philosophy is not a science; it is therapy that frees man

from unnecessary problems. In the same spirit, Alf Ross writes: Not legal

philosophy, but legal-philosophical problems (ikke ”retsfilosofi”, men

”retsfilosofiske problemer”). Yet, what is a legal-philosophical problem for Ross?

To encapsulate, it is the language of the doctrinal study of law. In this as well, he

was a child of his age. As Ross himself mentioned, logical empiricism, still very

much dominant at the time of ” On Law and Justice” set as its programmatic

statement the thesis: The task of philosophy is the logical analysis of the

language of science. Therefore, ”On Law and Justice” strives for the analysis of

the scientific nature (status) of the legal statements presented in the doctrinal

study of law.
 5

Alf Ross´ realism is literally legal realism. Everything he writes about law is

written through the doctrinal study of law. As we are about to see, this attitude is

a source of both Alf Ross’ strengths and weaknesses. Let us first take a look at

of a few starting points. Ross’ main target since his early works was natural law.

In his Kelsenian phase he might be best characterized as a legal positivist,

whereas the thoughts of the later Ross are more related to philosophical

positivism. Both of these attitudes meant a polemical attitude against natural law.

Therefore, one possible way to read ”On Law and Justice ” is choosing the

author’s negative orientation toward natural law as an interpretative horizon.

Another characteristic springs from Ross’ aforementioned motive of creating a

theory for scientific the doctrinal study of law. What follows from this is an attempt

to define what science is followed by the evaluation of whether, and with which

qualifications, the doctrinal study of law fits this definition. If it does not, it is

something other than science, for example influencing, manipulation or

persuasion. In this, Alf Ross is a true representative of logical empiricism, which

is no surprise when taking into account the strong position of the movement in

the years following World War II. In this setting, we can count, for instance,

Theodor Geiger and Hans Reichenbach among Ross’ kindred spirits.

According to logical empiricism, relevant scientific arguments are either (1) based

on perceptions or can at least (2) be returned to them. Ross concurs with this
 6

statement:

”It is a principle of modern empirical science that a proposition about reality …

must imply that by following a certain mode of procedure, under certain
conditions certain direct experiences will result … This mode of procedure is
called the principle of verification”.

The real content of an empirical proposition is composed of those actual

consequences that the proposition either expresses directly or that can be

logically derived from proposs that express direct experiences. Only in this way,

science fulfills the requirements for truth (sandhetskvalitet). In case that a

proposition does not fulfill this requirement, it does not belong to science even

though it might be interesting, aesthetically pleasing or well suited for invoking

feelings. Thus, only empirical propositions could belong to Ross’ scientific world-

view.

This can be called the empiricism hypothesis. On the other hand, also

statements (logically) derived from empirical sentences (verifiable implications)

belong to science. Let us call this the logicality hypothesis. Scientific theories are

complex combinations of these two elements. To encapsulate: They are

empirical and logical, to put it in one word, logico-empirical.

As a legal realist, Alf Ross saw the law produced by the legislative act is not valid

in any other way but formally. This formally valid law only becomes valid law

(geldende ret) once it is realized in society. Ross writes: “I reject the idea of a

specific a priori validity which raises the law above the world of facts, and
 7

reinterpret validity in terms of social facts”. From this viewpoint, the problem is in

what the “social nature” of law actually means.

Danish professor in civil law, Knud Illum, contemporary and colleague of Ross,

emphasized that law is valid if and only if the citizens have experienced it as

valid. According to Illum, the “legal conviction” of the people is the measure for

validity. In this context. To a certain extent, Ross accepted this view. Peoples’

conceptions (borgernes retsakter) define the content of legal life (retslivet) in its

broad sense. Otherwise, Ross didn’t accept Illum’s view .For Ross, people´s

conception about legal issues is not the criterion of validity.

Like American realism (pragmatic instrumentalism) he took off from the point

that only the law applied by officials is ”living”, or valid. Thus, the legal

phenomena as the counterpart of the norms must be the decisions of the courts,

and the validity of norms can therefore sought solely in the judicial application of

the law, and not in the law in action among private individuals.

One dimension of this idea is based on law always being the use of coercive

force. The machinery of officials realizes the use of coercion, as the judges

define which deed fulfills the threat of sanction. In Rossian terms, law, coercion

and official activity thus form the core of valid law.

Nevertheless, Ross rejected the idea that law’s coercive nature means the
 8

upholding of legal norm with the support of coercion. In the adjudication the

judge uses power but this does not meant that the validity of law depends on

power and coercion. Power is not something “behind” law although law is, in

simple terms, an instrument of power.

A key to Ross’ concept of the doctrinal study of law can be found in his theory on

the validity of norms. I shall illustrate this idea with certain ideas introduced by G.

H. von Wright on the same subject, since they are even better suited for

analyzing the basic concepts about validity than the analyzes made by Ross. von

Wright writes that it is not often natural to talk about the existence of a (legal)

norm. Instead, we can state that there exists a duty to do something. What, in

this case, does it mean that individual i has the duty to do, let us say, O?

According to von Wright, this means the same as: i belongs to group C of which

it is (very) probable that if a member of group C does not do O, then some A

regularly does R. Here it is presumed that A is an authority an something that is

unpleasant from the viewpoint of a member of group C, for example, some “bad”

or “unwanted” consequence. von Wright goes on to say that the unpleasantness

caused by R is, from C’s viewpoint, greater than the benefit or advantage gained

from not doing O. If individual i has, in these circumstances, the duty to do O, the

norm N “Every C has to do O” is valid. A norm like this can be called a primary

norm. Its “target” is precisely the behavior of the citizen (C) and its validity

depends on the reaction of the officials, in the aforementioned way.

 9

The legal order also includes norms that oblige officials (A) to put into effect

consequence R if an individual i of group C neglects to do O, that is, does not

obey the norm N. It is this secondary norm that guarantees the realization of the

sanctions set in the primary norm. The validity of the secondary norm thus

implicates the validity of the primary norm, but not the other way around, which is

why the validity of the secondary norm presumes the existence of a “Hartian” rule

of recognition or something like that.

Let us now return to the Rossian view about legal validity. Ross, like many other

legal theoreticians, used as his example the game of chess. He asked: What

does it mean that a certain rule of chess is, during the game, a valid chess norm?

The answer to that question opens the nature of a (chess) game as a normative

phenomenon. Once this example is moved into law, we can outline an answer to

the classic problem: What is law? In Ross’ opinion, the validity of a chess norm

can be analyzed only by separating the external and internal viewpoint. Ross

himself did not use exactly these terms, later presented, for instance, by H. L. A.

Hart, but he had by all accounts adopted the same view. Actually, it may be so

that in jurisprudence Alf Ross would earn the “paternity” of this distinction

because he dealt with it so early as in the beginning of 1950´s.

The distinction is crucial in Ross’ case. If someone who observes a game of

chess from the outside tries to understand the players’ moves, he must connect
 10

this activity to certain rules. The player acts in the way he does because he has

to, since if he does not, he will be excluded from the game. The game is

conceptually defined by the rules. The same goes for any other game. External

understanding thus presumes a previous, ”internal” understanding of the subject.

Ludwig Wittgenstein defended exactly the same idea. Without an internal point of

view, it would never be possible to understand an entirely foreign culture, since

we cannot know anything about rules that are completely strange to us. The

same hold true as far as the chess game is concerned. The rules of game make

the player´s actions intelligible to someone on the outside. The rules express an

abstract normative idea content, which can be used as an interpretative or

explanatory scheme.

In case of law, that interpretative scheme makes it possible to understand legal

phenomena (legal life in Rossian terms) as a meaningful totality, and at least to

some extent to predict the future legal phenomena (legal life).

From the internal point of view, the rule is no longer an interpretative scheme. As

distinct from the outsiders, the player does not try to understand his own or the

other player’s actions. He simply plays and tries to do it in accordance with the

rules in order to guarantee his participation in the game. The chess-game

example provides a model to understand the validity of a legal norm as well.

All this means that legal rules have a dual function, the external and the internal
 11

ones, depending on the point of view one takes. In this regard, the legal norms

are a part of the normative ideology (“judicial ideology”), internalized by the

judge. In this way, Ross independently represented a similar conception of

understanding activity (actions) as H.L.A. Hart. According to Hart, officials have

to accept certain judicial rules of recognition (as well as rules of amendment and

execution) as common norms concerning themselves.

As regards the doctrinal study of law Ross establishes as his point of reference

the sentence:

A (assertion) = D is valid (Danish) law

In which, for example
D = the Uniform Negotiable Instrument Act s.62.

A similar argument could naturally be made about any other possible statute in

any other national law. The assertion A is, according to Ross, a typical legal

statement. On exactly which conditions this assertion is true in the Rossian

sense? Ross reminds first, that a legal norm, just like a chess rule, forms an inter

subjective meaning and motivation context regulating the behavior as a shared

ideology adopted by the majority of judges. Rules are followed because they are

experienced as binding. In this regard, Ross was a typical Scandinavian realist.

He thought that a rule, which is experienced as legally binding provides the judge

with a model for behaviour. The model motivates the judge as does the rule of

the chess game as far as the player is concerned. The rule which serves as an

interpretation or explanation scheme for an outsider, is for the judge the basis of
 12

motivation.

The motivation to follow the rule is not based on need-calculations or other such

matters as interests, Instead, the normative ideology of judges gives impulses to

a person who experiences that ideology as a kind of social pressure. This makes

him to internalize the ideology spontaneously so that he commits himself to the

norm at issue independent of or even against external expectations.

The judicial ideology is not well articulated, but assumed as a kind of ”silent

commitment”. However, it gives an institutional support for the recognition of the

validity, similar to the Hartian rule of recognition. That rule of recognition makes

very seldom it explicit which norm is valid at a given time. The same holds true

as regards the judicial ideology. It is a complex totality with the result that

different judges may have and also actually have different conceptions on the

validity of norms. Nonetheless, it does not make the judicial ideology as a rule of

recognition dispensable or lessen its significance as the foundation of law.

Ross is careful not to confuse what is with what ought to be, and he thus in no

way belittles the strength of Hume’s guillotine. One cannot draw something that

ought to be from something that is. Hence, the judges are not supposed to draw

normative (ideological) consequences from the their actual work. Ross only

emphasizes that what is truly felt as binding must be internalized. The same

problem was pondered by G. H. von Wright in his theory of action. For him, one

of the forms of activity is “internalized” action, for example our socially

 13

internalized habit of stopping at a red light. It is unnecessary to ask in those

cases what an individual intended (wanted) to do. The reference to internalized

behavior is enough.

However, Ross was more interested in the method of the doctrinal study of law

than the deep structure of law. Therefore the core of his theory concerns exactly

the question: What is the assertion ”A = (D) is valid (Danish) law ” about? This

question is a crucial one as regards the scientific status of the doctrinal study of

law. In the simplest possible terms, Ross’ answer is: The assertion A is a

prediction. The real meaning content of the assertions on valid law is a

prediction concerning future social events.

The doctrinal study of law never looks at the past, to history, but at the future,

and therefore the legal research should never be interested in the court´s past

behavior de facto. This is the case despite the fact that it has to base its

predictions to a large degree on the previous behavior of officials.

The focus of the doctrinal study of law is always on the judge´s commitment to

follow a certain norm. This is another matter than “the activity of courts” since

the court is, according to Ross, an institution, and it is inaccurate to say that the

institution ”commits to a shared ideology”. In the Rossian theory the judge and

only him is the key actor.

Ross does not make a rigid difference between the predictions of the doctrinal

study of law and legal-political estimations (sentential ferenda). This is due to the
 14

fact that future social events are never strictly determined. Each prediction might

also affect the future, shaping it. To a certain degree, the prediction might thus

also be self-fulfilling.

Notwithstanding the above difficulties, an assertion A presented in the doctrinal

of law is as follows:

A: The norm N (for example §28 of the law on bills of exchange) is

currently valid law.

Independent of the linguistic formulation of A, its meaning content can be

reformulated in the form of prediction:

A1: If proceedings are instituted in a court so that the facts and

circumstances fulfill the essential requirements of 28§ of the law on
bills of exchange, it follows that the norm expressed by 28§ of the
law on bills of exchange will form the decisive foundation of the
court’s decision in the case (integrerende bestanddel af
dombegrundelsen), and all this on the precondition that there are no
factors between the prediction and the decision that change the
foundations of the prediction A.

The prediction A1 can be taken as accurate (anses for sand) if and only if there

are good reasons to estimate that the prediction will be realized in the activity of

the judge. Still, it must be emphasized that the “truth-value” of sentence A1 does

not depend on whether, for example, the evidence of the fact can be estimated

well enough in an actual case. The predictive nature of sentence A1 does not

depend on procedural evidence, but on the prognosis whether a certain norm will

be included in the judges´s reasoning or not, that is, whether a norm will be

applied in given circumstances.

 15

The prediction concerns always the normative ideology internalized by the judge.

The norm is valid in a certain legal order if and only if it is a part of that judicial

ideology. Further, the assertion A1 about the validity of a norm is true if and only

if the norm is a part of a binding judicial ideology.

In one sense, Ross’ thoughts prove to be quite problematic. He thinks that the

probability by which the assertion A1 proves to be accurate also defines the

validity of norm N. From this it follows that norm N can never be either valid or

invalid. It is only valid with a certain probability that changes on a scale of

(nearly) 1 – 0. Ross himself points out, that the validity depends on the varying

degree of probability concerning the predictions of the future behavior of the

judges. i

For its part, the degree of probability depends on the experience-material

(erfaringsmateriale), the prediction is based on. Ross calls this material legal-

source - material (retsskilderne). Therefore, a prediction deals with the probability

of a certain norm belonging to the judicial ideology shaped by the sources of

law. As this is the case, there appear problems in two different dimensions. First,

it seems to be impossible to apply a probabilistic model to the treatment of this

kind of material. Legal source -material (or judicial ideology) does not often give

one the chance to estimate probability quantitatively (for example as a numerical

prediction: a probability of 75%). There are no quantitative measures for the

predictor to use. Thus, the prediction must be based on a qualitative analysis.

 16

The “probability” seems to be more about a kind of ”credibility” “acceptability” or

”arguability”, that is, about how well founded justification of the future activity of

courts can one form. “Arguability” is something else than a mathematical-

theoretical “probability” based on statistics, other quantitative data or on the use

of empirical observations. This is connected to the fact that “predictions” of the

doctrinal study of law – unlike the predictions of science, for example – are based

on data, the ontological status of which is problematic. This can be seen as one

ponders the way a norm, a court decision or a custom “exists”. As we saw

before, these data are ontologically difficult in the same way as the “existence” of

language. They are partly dependant on the subject and his interests.

It is obvious that the perceptions concerning the scientific predictions (meter

readings, pictures on an electron microscope etc.) also need interpretation, but

the difference to the ”perceptions” that support legal predictions is clear. As the

interpreter of his “perceptions”, one engaged in the doctrinal study of law also

belongs to the community that has produced the interpreted data in the first

place. This person evaluates the data from an internal viewpoint, and is not an

objective observer in the same sense as a physicist observing the readings on a

meter. A lawyer has already learned the language where and in relation to which

the interpretation is produced.

This can be called epistemic internal viewpoint. In a sense, it is a key both to the

criticism of Ross and the theory of argumentation represented in this

 17

contribution. The “epistemic internal” viewpoint combines the intellectual

approaches of the scholar and the judge toward law, for example, toward

sources of law and their way of use. Without an adequately shared epistemic

viewpoint the scholar and the judge will not reach an agreement on the content of

law, and in extreme situations might not even understand the language the other

uses. This adequate and shared viewpoint also separates the scholar and the

judge from an epistemic external viewpoint, which is represented by, for instance,

a sociologist following an empirical method.

On the other hand, the scholar and the judge have an organizationally different

position in relation to law. Only the judge is “inside” the organization of

jurisdiction. He is the applier of coercive force, why the position of the judge can

be called organizationally internal to make a distinction with the organizationally

external viewpoint of the scholar.

A genuine prediction always aims at the future even though it is based on past

facts. This raises a problem: What, in the end, is the target of the prediction?

There are multiple answers to this question.

(1) The prediction is targeted toward the end result of the case at hand (e.g.. the
Supreme Court will affirm the consequence X)

(2) The prediction strives to make a future decision understandable, to give it an

interpretative background (e.g.. the Supreme Court’s decision X in situation Y is
understandable because Z)

(3) The scholar tries to anticipate the (public) arguments the court announces to
 18

have based its decision on

(4) The scholar predicts the norm the decision (probably, credibly etc.) is based
on, even though the norm is not written out.

Taking into account Ross’ view of the doctrinal study of law, the alternatives (1) –

(3) can be ruled out. If the prediction would solely concern the end result of a

certain case, it would not fulfill the basic requirements of the doctrinal study of

law. Contrary to the task of an attorney or other advisors involved with certain

cases, the task of legal research is not to anticipate the results of singular

decisions.

As regards the alternative (2), it is difficult even to talk about a prediction. Making

future cases understandable is not prediction at all. It is an explanation, and this,

in the case of alternative (2), can be at most a reference to an “explanatory

conformity” or the motivational foundation of the judge. The alternative (3) is no

more promising. The fact that we have some way of anticipating the arguments

used by the court does not necessarily tell us what the content of valid law is. As

Ross remarks on many occasions – like a true realist – arguments can also be

nothing more but a façade. The decision may be justified with entirely different

arguments than with those used de facto by the judge, or they may mean

something entirely else. In some cases, the decision may be made even without

a single rational argument. On the other hand, we can never find out what the

judge really thinks, for we only receive the arguments, having thus to manage

only with the facade. Ross did not think this enough deeply, since even the realist
 19

doctrinal study of law does not predict mere “facades”, it reaches to something

deeper, something more genuine. It predicts normative information. What is this?

The answer is bound to the alternative (4), combined with alternative (1). For

Ross, the genuine prediction concerns the premises used by the judge, that is,

the judicial ideology.. He points out, that if we are able to predict the legal

premises of the case, we also succeed to predict the conclusions. In this way,

Ross places the judicial ideology in the centre of his prediction theory. The

prediction attempts to reach the (de facto) effective normative ideology or that

one which can be thought to be effective.

The prediction does not concern the individual arguments used by the court, but

whether or not the norm belongs to the normative ideology accepted by the

judge. He follows that very norm due to the fact that he has internalized the

normative ideology in an uncompelled way. For this reason, the norm forms a

genuine motivational foundation for the judge´s decision. The motivational

foundation of the judge and his knowledge of the facts are what define the

decision he makes. This is why it is important for the predictive doctrinal study of

law to know which (normative) motivational foundation the judge commits to.132 In

this regard, Ross has to be clearly separated from American realists.

Here, the doctrine of the sources of law forms an essential part of the normative

ideology. According to Ross, the content of the normative ideology is defined by

the sources of law, which are the essential constituent of valid law. The
1
 20

normative ideology consists, in its turn, of directives which do not unambiguously

regulate the final decision but gives a guideline or a standard according to which

the judge is able to orientate further in order to find, in the end, the decisive

directive. Into this normative ideology belong not only the norms of law but also

those of customs and morality.

Ross calls his conception ideological behaviorism. It aims at anticipating

behavior, but selects as its target the ideology that guides the behavior. Here

Ross is drawn far away from rough stimulus-response-behaviorism, and he thus

also takes distance from vulgar court realism. Still, the theory of prediction faces

other challenges as well.

First the usefulness of the prediction theory can be questioned. A judge is not

interested at all in knowing the prediction about his own actions. His interest

concerns the way he should act, that is, what is the content of the binding

normative basis in his case. There is, however, an even more serious deficiency

in the prediction theory. It does not make possible the critique of the court’s

activity.

The doctrinal study of law is expected to evaluate the interpretations of the

adjudication, and to give constructive feedback to the courts. The social task of

the doctrinal study of law is partly normative, and accepting this normative task it

necessarily becomes critical toward the judicial practice. The doctrinal study of
 21

law gives reasoned recommendations on what it is to act in a right way in certain

typical situations. The doctrinal study of law that is solely predictive is only a

lackey of the courts, not their supervisor.

Ross rejected this critique. For him, the prediction theory does not make the

doctrinal study of law a slave to practice (slave af praksis). His answer is firmly

attached to his original theoretical hypotheses. Ross says that predictions “live”

constantly. When there is new information about the judge’s actions, the

prediction is adjusted. The prediction A1 in case Y is not probable if it, all things

considered and including the former decision X, is probable that case Y will not

be decided like X. The predictive doctrinal study of law is thus all the time

dynamic. It takes into account all the new information concerning the changes of

the normative ideology. Still, there is left an open question: What knowledge is

the prediction based on?

According to an orthodox interpretation of the Rossian theory, the doctrinal study

of law is based on empirical research dealing with the behavior of judges. This is

why the doctrine of the sources of law is not, according to Ross, normative as to

its nature. Instead, the doctrine is descriptive and presents the content of judicial

ideology in actual use, and this material can be observed empirically.

However, let us remember what Wittgenstein said: One must know something

before asking the name. This holds true also as regards the predictions. In order
 22

to predict the future behavior of the judges, one must already have some idea of

the judicial ideology prevailing in the court as well as of its role in judge´s

decision-making procedure. Yet, this is not enough. For a prediction one needs

also information about the effective normative ideology. This is where Ross gets

into trouble: On which grounds can a conception about judicial ideology be

formed in practice?

Interviewing the judges is out of the question. It produces something like the

legitimation of what the judges feel to be the right way to act rather than

information about the effective normative ideology. When asked essential

questions, man tries to give a best possible image of himself, and judges are only

human beings.

There remains only one possibility. The prediction must based on what one

knows about the judges thinking in general. The scholar has to presume that the

judges’ thoughts about the content of law are similar to the thoughts of other

lawyers, including the scholar himself. The predictor must have the same

epistemic internal viewpoint as the judges. Without this, the scholar has no way

of understanding the judge’s actions.

This means, that the scholar has to be ”inside” the same normative world as the

judge, that is, he has to internalize the same sources of information – and vice

versa. The question is not only about the scholar somehow “finding” the
 23

descriptive sources of law. He also interprets, understands and puts then into the

order of preference. The scholar gives the sources of law a content he assumes

to correspond to the legal community’s general way of thinking about sources of

law. According to this principle, the scholar and the judge have to use (nearly)

the same

(1) sources of law,

(2) legal standards of reasoning, and
(3) somewhat similar principles of rational discourse.

To put it in a nutshell, the scholar necessarily has to assume that the judge has

adopted a similar, not necessarily identical, but family-resembling way of thinking

as he himself. From this it follows that legal “prediction” is essentially similar to

any kind of argumentation based on sources of law. There is no other way of

“anticipating” the future activity of the courts than to use the source material

accepted in the legal community in general.

The Rossian prediction theory does not take this dilemma seriously. That is why

his theory leads to mere pseudo-predictions. The scholar might have, of course,

a predictive intention, but as a matter of fact, he must base the predictions

exactly on the same source material as the non-predictive doctrinal study of law.

This conclusion is not made different only by naming the statements as

predictions. As far as their content is concerned, they are nothing else but a

doctrinal interpretation, or weighing, filling normative gaps, using contra legem

deliberation, or eliminating contradictions.

 24

It seems to me, that Ross was aware of the problems inherent in pseudo-

predictions even though he did not draw the necessary conclusions from these

difficulties. He even noted that the predictions of the doctrinal study of law are

linked with the same self-fulfilling characteristic as “predictions” about society in

general. When a prediction is made public, it tends quite often to bring about

exactly those consequences which are anticipated in the prediction. This is so

because both the “predictor” and his target are intertwined with the same societal

system, and bound to the same normative basis.

All this means, that Alf Ross could not consistently hold to his empiricist thesis.

His barely noticeable separation from logical positivism is a necessary

consequence of rejecting purely behaviorist legal theory. The essential element

of his prediction theory is the notion of judicial ideology, which in its turn is a

normative phenomenon. As Ross was quick to deny that the task of the doctrinal

study of law is to examine the judge’s motivational process by the tools of

psychology, there was actually nothing more left than the acceptance of a non-

positivist viewpoint to a certain degree. Ross´ theory is both behaviorist as well

as idealistic, why Alf Ross was a positivist by words, not by deeds. Instead of

empiricism, Ross actually represented a hermeneutic, that is, a non-positivistic

approach. He defined the validity of law on the grounds of the judge’s internal

motivation, and doing so he had to refer to a normative material, i.e., to the

sources of law. This is why the Rossian doctrinal study of law cannot be only

description or explanation of the judge´s behavior, not even its prediction. The
 25

doctrinal study of law is the rational reconstruction of how the judge as well as

the citizen should act.

26
i