Analytical Philosophy and Philosophy of Science was and is mainly concerned with the procedures of research and theory-building in the realms of natural science. The so-called humanities or "moral sciences", since the times of Auguste Comte and John Stuart Mill, were mostly considered to be immature and underdeveloped parts of the monolitic venture of modern science, which was built on the bases of mathematics and physics. The members of the Vienna and Berlin Circle rigorously excluded them from their ideal of unified science as "metaphysical poetry" (Begriffsdichtung) and accepted only parts of them, e.g sociology, economics and later on linguistics, which seemed apt to be moulded in the forms of the paradigm of mathematical theory-building.

Problems of truth, falsity and verisimilitude held always the center of the attention of philosophers of science. Philosophy and sciences, since their beginnings, were understood to be cultural institutions for investigating the truth, to combat and eliminate falsities and to purge probabilities in such a way as to get them nearer to or to convert them into truths. The dominant conception of truth, from Aristotle up to our days and obviously also in Analytical Philosophy, was and is the correspondence model or the so-called picture-theory of truth. Truth, in this conception, is the relation of a picture to its pictured object, whatever the nature of the picture or the object might be. It is the "adaequatio rei et intellectus" (equality of the thing and the mind) in terms of scholastic Aristotelians, and it is the one-to-one correlation of a syntactic formula to its semantic object. Falsity then is also a correlation, which obviously always introduces itself as a true picture until it can be "falsified". Falsification then is normally held to be a demonstration, that the alleged picture is in deed no picture at all, and that it depicts therefore nothing. But besides the unsolved problem, whether "nothing" might perhaps be depicted by true pictures (as e.g. by the pictures of absolute darkness or silence), there are good reasons at hand to argue, that a false picture is certainly not the picture of its supposede semantical object, but of an other object, depicted by another true picture. Verisimilitude, truth-likeness or probability than is something in-between: an unclear picture containing traits of conjectured but unproven truth and supposed but equally still non-demonstrated falsity.

Everybody knows, that this clear-cut model of truth has come into some troubles. Popper insisted that truth can be conceived of only as an ideal, which science never attains, science being only conjecture and thus probable knowledge, always in need of tests of falsification1. Feyerabend spoke out "Against method"2, which was indeed a call for abolishing all truth questions regarding knowledge. Thomas S. Kuhn historized truth questions into paradigmatic but pluralistic and uncommensurable world views3. But notwithstanding all these troubles, the "normal scientist" of the scientific community has not the time to wait for better solutions here, but continues to use the traditional truth conception of the picture-theory, taking his theories to be true pictures of his research objects and trying to clarify all those parts of his theories which do not yet fit adequately into his world-picture.

Now, talk of "interpretation" is quite common in the scientific community. The scientist interprets the phenomena of nature themselves, theories about these phenomena and not the least (mathematical) formulas exacting these theories. Behind this custom, there prevails the old neoplatonic tradition (of the cabbala, Augustinus, Galilei and so many others) of considering the whole of nature as the creation of a divine creator and as a second "book of revelation" parallel to the first one, the bible, wherein everything is written in mathematical and geometrical characters, which need decifering by the scientist, who then can detect the real sense and meaning of this creation and its author. Later times have replaced God by the human genius, who continues to speak in a language, mainly a mathematical one, which is not perspicable for the laic, but only for the congenial interpreter. To grasp the sense and meaning here - of the phenomena of nature, of scientific theories, of mathematical formulas, of logical concepts, of texts and books and man-made documents - is the task of hermeneutics. And obviously this task has likewise to do with truth, falsity and probablility.

One can easily see in the history of hermeneutics and in the history of science, that the prevailing conception of truth was and is the same for both realms. It is the correspondence model. It was not lent or transmitted from the (quadrivial) sciences to the (trivial) humanities, but it rose form the common origin as a lasting legacy of occidental neoplatonism. When in our times some few analytical philosophers began to assess the procedures of hermeneutical research in the humanities - as Toulmin, Davidson, Searle, Hirsch, Freundlieb, Albert and Lenk did4, to name some of them - there is nothing astonishing in the fact, that they presupposed the correspondence model of truth and falsity also in this realm of the text- and document-interpretation. Nor is it surprising that they met with the same persuations of the more traditional representants of the "theory of Geisteswissenschaft" from Schleiermacher to Dilthey, Heidegger and Gadamer5, namely that there is meaning and sense as an object in the texts, which is to be pictured in and by the corresponding interpretation. Schleiermacher and Dilthey propounded the opinion, that the task of interpretation consists in a "psychological and congenial reconstruction of the original thought of an author"6. Whatever the author of a text constructed in his mind, the interpreter should be capable of re-constructing it in his own mind. Gadamer spoke of bridging the gap between different horizons of meaning (of different times and /or cultures)7.

In order to clarify things, let us depict the hermeneutical situation in terms of a pot-and- cup model of sense and meaning. The pot stands for texts and documents and represents the material side of the relevant object. The real and interesting object, however, is its content, so to say the "soup" of sense and meaning in the pot. It is the product of an author, be it a God, a genius or a simple human being, and it bears all the idiosyncratic flavours of talents, conditions, and intentions of the author. Understanding and interpreting sense and meaning is usually depicted as "putting out" ("auslegen") or "taking out" ("ausschöpfen") this metaphorical soup into one's own cup of understanding. The common terminology itself hints here at the metaphor of the pot. And it goes still further: whence the author is divine or a genius, the soup is considered to be quite "inexhaustible" ("unerschöpflich"), whereas the soup of mundane concoction is said to be fully and completely exhaustible by the interpreting consumer.

Now, the soup in the pot and the soup in the cup are different. What they have in common as the same soup represents their both being sense and meaning. But the soup in the pot stands for the sense and meaning in the text, and the soup in the cup stands for the meaningful interpretation of that sense and meaning in the pot. Correspondence then is based, according to mathematical equations, on the substantial identity (the so-called "value" or "meaning" of Frege) and at the same time on the difference of the expressions of that meaning by different "senses" (according to Frege's terminology).

What now about false and probable interpretations in the pot-and-cup metaphor? False understanding and interpretation compares with a quite different soup in the pot and in the cup, the latter not steming from the relevant pot, but from a different one, and poured into the cup by an evil spirit in guise of prejudices and malevolence. Probability in understanding and interpretation must then result, when it is an open question, whether the soup in the cup comes from the right pot or from a different one.

It is well-known, that this realistic "gnoseology" in the foundations of the correspondence-theory of truth and falsity in hermeneutics has severe shortcomings. Firstly, it presupposes the object as a "thing in itself" (as Kant has called it) and as existing quite independently from its being known or not known by a knowing subject. Secondly, speaking about truth of knowledge in terms of the correspondence model, there is no means of comparing the object with the knowledge-picture, because all one could say about the "object in itself" is already knowledge and hence belongs to the picture itself. Even analytical philosophers have come up the long way to acknowledge this "dialectical" situation. There is much talk nowadays of theory-comparing, of identifying true theories by falsifying concurrent ones. But the correspondence.theory of truth and falsity and accordingly also the pot-and-cup modell does no longer work here, and there seems to be some preparedness to drop it and to switch to the other theory of truth and falsity, which also has a long tradition behind it: the so-called coherence-theory of truth and falsity.

The coherence-theory has its own shortcomings in the fact, that one can easily stylize theories in coherent fashion by logical and mathematical means, so that there are usually too many coherent theories concurrencing in order to explain the same object (think of the "duality" of particle- and wave-theories in microphysics). Even if some of them might be shown to be the same theory in different "ideal languages", so that theory-comparing results in simple translation from one symbolic medium into another, there remains always a rest of really different if not contrary theories of likewise coherent structure about the same object. And this shows, that coherence alone will not do, but must be supplemented by an additional criterium. It seems to me, that Quine, who notably propounded non-translatability of theories in general, was on the right track with his idea of holism. Holism, adapted to our truth problem, must then mean comprehensiveness - besides coherence - as a criterium for true theories. This means, that a true theory must not only explain a certain object in a coherent manner, but that it must fit coherently into encompassing other theories about a whole field of studies and further on into transdisciplinary horizons.

Let us see now how coherence and comprehensiveness as truth-conditions work in the realm of hermeneutics.

I would like to propose here another model showing the relevant features in the hermeneutical process. It is the mirror-model of interpretation. A mirror, as everybody knows, shows all things lying before it in such a way, that they seem to ly behind it (or in it). But we all know for sure, that what we see in a mirror is not "in it", but before it. And we know the mirror-reflection to be an "appearance", because we can compare the mirror-picture with the real things before it. And doing this, we always find that the mirror-picture converts the right and the left side, which then serves as a criterium for the illusionary character of the mirror-picture. Consequently, whatever we know about the real nature of the things we get from the things before the mirror and not from their pictures in the mirror.

What is the metaphorical impact of this mirror-model on hermeneutics? I would like to point out, that texts and documents work alike optical mirrors, and that they are, metaphorically speaking, hermeneutical mirrors. Looking into a text or a document cheats us in the same way as the optical mirror does, by presenting us with the illusion that we see something in it, that is not there, but that we must bring with us. It is the sense and meaning, which we ourselves "see" into the texts (and certainly not with our real eye, but with a "spiritual" one, as Plato said). It is a significance we hope to find in the text, and which we find there according to the degree by which we apply it to the texts. The more we bring to it, the more we find in it. It begins with elementary traits. For instance, knowing the letters or characters or other symbols of a document and their meaning, knowing words and sentences of a language, having ideas and reminiscences about things, situations and persons, and going on to whole cultural horizons. For him, who does not command such knowledge, the text obviously is not only silent, but it is no text at all.

What does the mirror-metaphor teach us about understanding in terms and as a critique of the above mentioned correspondence-theory of truth? Cheating us to "see" the meaning in the text, we try to mirror it by our interpretation. The true interpretation, as we said, should then "correspond" to the original meaning in the text. In order to correspond, there must be two different meanings to begin with: one in the text, the other in the interpretation. But how can they be distinguished to enter into a correlation, since they are not material things, which could be split up and attributed to either pole of the correlation? There is only one means to achieve this. And this is, what the mirror does by converting the right and left side of every picture, thus indicating the falsity of the illusion: We must declare every interpretation, which is ment to be a corresponding picture of an original sense and meaning in the text, to be a false interpretation. Only when we suppose to be two different systems of meaning and sense, one in the text, the other in the interpretation, then only can we differentiate between the meaning of an interpretation from its alleged object, the meaning in the text. And the latter being different from the former, the interpretation must be false.

So the mirror-model may help us to take into account and to unveil the illusions of "sense and meaning in the text-machine" (as Ryle might have said). It depicts the situation, which Goethe once coined into the popular saying: "Beim Auslegen seid frisch und munter; legt ihr's nicht aus, so legt ihr's unter" - which may be translated in the present context as saying: "since you can't lay out the sense, you have to lay it in or under, that is: do suppose it". But that leads to the question, how is it that a text as a material machine may be capable to instigate a thinking mind to evoke and assemble ideas in such a way, that their combination and structure might be called an interpretation of the text? What can the mirror-model of interpretation suggest here?

There are certainly many traits and qualities in mirrors which have served well in metaphorical explanations about what thinking really is. The word "speculation" and its connotations are testimony to that in the history of philosophy. The "monadological mirror" of Leibniz was not the only one, but perhaps the most prominent metaphor for a thinking being. It became the cornerstone of all later pluralistic world-view conceptions as well in the philosophy of science as in hermeneutics and ideology-critique. Let us name only Johann Martin Chladenius in the 18th century8, Wilhelm von Humboldt in the 19th9, and Mauthner up to the later Wittgenstein1o and Thomas S. Kuhn in the 2oth. They all expounded the conception that the world and the texts alike are to be "seen" from a certain - and for every viewer different - standpoint in a mirror-like distortion of world- and sense-pictures. But here the mirror was identified with the thinking and knowing subject, and the object of the mirroring was presupposed as being the real thing in itself. We propose, however, to exchange these parts: The object (as the text) is taken to be the mirror, and the subject (the interpreter) receives his own thoughts and his knowledge as reflected and shaped into a certain order and structure by the text-mirror.

What then can the mirror-model suggest for our topics and teach us about truth, falsity and verisimilitude in the hermeneutical process? 1. that there is no sense and meaning in the text itself, and what we take to be such a thing is only existent in the interpretation itself. 2. The text, or rather generally all literature (and further all documents), are not an object of our research for meaning, but they are "cultural machines" to concretize and modify our own ideas and knowledge. 3. The optical laws of reflection and side-converting of real mirrors have their analogon in the hermeneutical features of clarifying our own ideas and of substituting them as to be not our's but those of the author of the text. 4. The resulting picture is not a picture depicting an object whatsoever, but it is a pittoresque construction - like the "pictures" of modern art - using perhaps pictorial elements, but only as a means to make the interpretation understandable.

This mirror-model conveys some corollaries, namely: 1. the more the interpreter knows, the more and the better he understands his text or his document. 2. He who has no ideas to bring to the text, cannot interpret the text. 3. He who concludes that there is nonsense in a text, has already interpreted a part of the text as if it were no text at all. 4. And he who comes to the conclusion of a "non liquet" vis-à-vis a text, meaning, that he cannot understand the text, confesses only, that he has no relevant knowledge at his disposal.

Applying our criteria of coherence and comprehensiveness upon the question of truth in hermeneutics, we can say that the whole business of constructing a true interpretation of a text consists in assembling all available knowledge concerning a text or document and shaping it into a certain order. Just by doing this, we at once transgress the immediate text and go over into contexts and further superstructures of ranges of artefacts of cultural determination. What von Humboldt rightly remarked, namely that in a single word of a language there resounds the whole of the relevant language, that is also true for interpreting singularities concerning a text. Its symbols - letters, characters, artificial signs - must be recognised as symbols proper and therefore belonging to systems of signs of a general character. The same holds true for single words of a text-language and further for patterns of linguistic expressions, for contexts speaking about things, persons and situations, for expert literature on problems and themes. We have to know all that beforehand, and if we don't, we need to get hold of such competent knowledge from whatever discipline, which seems relevant to the task. This going up from the singular to the general and down from the general to the singular has often been described as the "hermeneutical circle". But it is not something specific in the hermeneutical procedure, but that holds true for every kind of theory-building. It is the logical way of inductive and deductive assertation of coherence of all parts of a theory.

Falsity in interpretation can only be measured against a true interpretation. The false interpretation is then called a misinterpretation. But with the exemption of cases of intentional misinterpretation of texts and documents in the law-practise and in the making of history, it seems to be extremely rare that an interpreter offers false interpretations intentionally. This business, as all kinds of lying and fraud, is too strenuous, because one has to know the true interpretation beforehand and must dissimulate it cautiously. Therefore, habitually, interpre-tations are always ment to be true ones. And under this presumption they also must strife to fulfill the criteria of coherence and comprehensiveness at once. The first test of falsity therefore consists in establishing incoherences and inconsistencies, which means to show that the allegedly false interpretation suffers from inner contradictions. But this does not make the whole interpretation false. Contradictions, as is well known, show up in those parts of a theory, where the same topic is affirmed and negated at the same time, and than either the affirmation or the negation must be true. In order to assess which part of a contradiction is true and which is false, one has to compare them with the true interpretation. Only then the false part of a contradictions may be identified and measured against the truth.

It seems to me of utmost importance to underline this fact, that inner contradictions never falsify the whole of a theory or an interpretation, because logicians traditionally take contradiction to be the formal indicator of falsity point-blank. If a contradiction arises, there must - by definition - also be a part of truth in it, the other part being then the really false one, althought it may often be difficult to establish, which part is the true and which is the false one. One can easily imagine, that many truths in interpretations (as also in theories) have been rejected and eliminated as falsities by applying a too simplistic and so to say all-over criterium of contradiction.

As concernes comprehensiveness, we said that a true interpretation must fit into the far-reaching frames of established knowledge of relevant disciplines. An interpretation may then be coherent in itself without fulfilling this criterium of fitting into those frames. Such interpretations are sometimes very ingenious, originial and fascinating and may then contribute to shake up the whole fabric of established disciplinary knowledge. But even then they are to be measured and assessed by the more comprehensive interpretation which fits coherently into the established disciplinary knowledge, and which therefore is taken to be the true one. Considered at the level of comprehensiveness, the limited false interpretation contradicts the more comprehensive interpretation in toto and must therefore taken to be false in toto.

But besides the logical and formal side of falsity in interpretations, there is still another problem of falsity involved in hermeneutics. It may often be the case - and perhaps it is the usual case in scientific and erudite literature - that we have to understand and interprete falsities as such. That is to say, we must truly understand falsity and interprete the falsity in a true way. To interprete falsities as the false part of contradictions in the text is the normal procedure here. But there are also falsities, which are not a part of contradictions. Take for instance errors or lies. The interpreter normally shyes away from this kind of interpretation, because it sounds like a critique, and critique is considered not to be the business of the interpreter, but of the researcher. The interpreter then prefers to say, that what he understands as to be false, is utter "nonsense" in the text. But nonsense cannot be understood at all, and we have reasons to say, that an alleged text suggesting the interpretative verdict of "nonsense" is no text at all. Let me hint at the very common experience of logicians, when they are confronted with quite nonsensical linguistic utterrances (which sound like speech, but are not speech at all), which they then should formalize to make them "meaningful" and to find out their truth-values. Indeed, nonsense exists, however not inside, but outside the realm of hermeneutics.

So the interpreter must be careful not to confound falsity with nonsense, and he must be prepared to understand and interpret falsities as such. The history of philosophy and of science as also the history of the humanities and letters gives abundant examples for this task, when it comes to interpret whole "philosophies" and "theories" as being partly or totally false. Indeed, the interpreter must always carefully avoid falling into the trap of confusing falsity with nonsense on one hand, or falling into the other trap of confounding meaning and sense with truth. It seems to me, that not all the famous interpreters have successfuly avoided these pitfalls.

So what about verisimilitude and probablity in hermeneutics? Probability logics is now a special field of modal logic, namely the so-called many valued logic. It regards probability as a medium truth-value between truth and falsity. And as such it is just what Aristotle called the "Third" besides the "First", namely the truth, and the "Second", namely falsity. The terms "verisimilitude" and also "probability" and "truth-likeness" suggest, that the medium truth-value is nearer to the truth or that it is just a special kind of truth. But this is only a time-honored prejudice. As the medium between truth and falsity, it is as near to truth as to falsity. And this can only mean - in my opinion - that it is both of them at once: truth and falsity combined11.

The logical form of such probable interpretations is the hypothesis or the problematical proposition. It is a suggestion for sense and meaning, but it is never an affirmation of it, nor a negation. In common language, we express such conjectures in the subjunctive way, saying: it might be that the meaning is such and such. But not knowing for certain, whether it is so or not so, we practise a kind of "docta ignorantia" (as Nikolaus Cusanus called it). Let us state it clearly: probable interpretations therefore are not at all real interpretations, but - as all hypotheses do in theory-building - mark only a preparatory step in order to focus our attention onto what can be done to achieve real interpretations. Not conducive to any obligation, they are at once the convenient means in hermeneutics to be right in any case, be it that a subsequent true interpretation "verifies" the true part in the hypothesis, or be it, that the same subsequent true interpretation "falsifies" the false part in it. And this is so, because a probable statement, that the sense and meaning of a certain piece of text "may be such and such" never excludes, that it is "not such and such".

Finally, we must distinguish between these so-called "probable interpretations" - which, as we said, are not interpretations at all - and the cases, where we interprete a text or a document as being "dialectical". We must then interpret it as invoking "double meaning", that is a true and a false one at once. But our interpretation of such "ambiguous" sense then will either be true or false - if it is not meant as a "probable" proposal or a hypothesis about the dialectics of a text.

Coming to an end, I hope to have shown that hermeneutics is a discipline in which the analytical philosopher, too, may find ample stuff and challenges for a rigorous application of his analytical and logical methodology.