Philosophy and Ideology. Z. A. Jordan 1963
The questions what does the truth of a belief consist in and how can we recognise that it is true, have given rise not only to various theories of truth but also to various philosophies. For as soon as the question of truth is raised, a whole intricate complex of issues appears in its wake and the examination of the concept of truth is transformed into a dispute between rival philosophical theories. Some of the most influential modern philosophical doctrines, for instance, the instrumentalist philosophy of Dewey or logical empiricism, have the criticism of the traditional concept of truth as their starting point.
Marxist-Leninist philosophy claims that it has a theory of truth of its own, the socalled (materialist) theory of objective truth. Its fundamental importance for the whole doctrine was emphasised by Lenin. ‘To regard our sensations as images of the external world’, Lenin wrote, ‘to recognise objective truth, to hold the materialist theory of knowledge – these are all one and the same thin’ . Lenin could not have wished to say that all these beliefs are ‘one and the same thing’ in virtue of having the same meaning, for they do not mean the same. What he probably had in mind was the logical equivalence of the three views which being different in their respective meaning entail each other. Since according to Lenin the materialist theory of knowledge and materialism are also logically equivalent beliefs, these two together with the copy theory of perception and that of objective truth should be considered from the Marxist-Leninist viewpoint as a coherent whole in which each entails the other and none of which could be false if the other is true. This seems to be clear also from what Lenin’s followers in Poland say on this matter.
In spite of the importance attached to the concept of truth in Marxist-Leninist philosophy, no larger work dealing with this concept ever appeared either in the Soviet Union or elsewhere. Schaff’s study Some Problems of the Marxist Theory of Truth, published in 1951, should be, therefore, considered as by far the highest single achievement of the Marxist-Leninist school in Poland. It is the first and – its title notwithstanding – an exhaustive monograph on the subject, notable for its wide scope. It is half analytic and half polemical, the second half being concerned with the conception of truth of empirio-criticism, conventionalism, pragmatism and logical empiricism, including operationism.
There is very little in the works of Engels and Lenin which Schaff could draw upon in his examination of the materialist theory of truth (this might have been one of the reasons why no other monograph of this kind was published elsewhere), and he is heavily indebted to non-Marxist philosophy. In particular he made considerable use of various concepts and views worked out in Poland, above all, by Twardowski, Kotarbiński, Ajdukiewicz and Tarski.
Schaff did not adopt the achievements of the Warsaw school indiscriminately. He subjected them to a scrutiny from his own viewpoint and adapted them to his own purpose. But even when this was not necessary and his direct philosophical debt is unmistakable, he was inclined to exaggerate the differences or to find them where there were none, for the Warsaw school thinkers were at that time condemned for their alleged idealism, conventionalism and other disreputable tendencies. His critical examination of and polemical observations on other people’s views, those of foreign thinkers in particular, to whom he owes a great deal, do not always excel in fairness and objectivity. He often fails to acknowledge the value of the contributions made by thinkers with whom he disagrees and misconstrues in his account the views which he wishes to criticise. A harsh and self-righteous note of accusation pervades even criticism which is otherwise justified. Mach confessed in Analysis of Sensations that he never felt it necessary to insult people whose opinions differed from his own. In this matter, as in many others, Schaff felt differently from Mach.
Marxist-Leninist philosophy rests on what might be called the ‘principle of dichotomy’. There are two and only two possible standpoints concerning the nature and constitution of the Universe, those of materialism and of idealism (in the Marxist-Leninist sense). There are two and only two possible standpoints in epistemology – the materialist and the idealist theory knowledge. In the Marxist-Leninist universe of discourse materialism or dialectical materialism and the materialist theory of knowledge are classes with a single element, and all the other possible views, whether like or unlike, comparable or incomparable, constitute their respective complementary classes. The materialist conception of truth also rests on the principle of dichotomy.
A theory of truth should provide a definition of the common property of true beliefs, that is, of some property which can be ascribed to all true beliefs without exception. If this condition is not fulfilled, a theory of this sort would be considered to be deficient in this respect. Schaff tacitly makes this assumption. From this viewpoint, all definitions of the common property which belongs to true beliefs can be divided into two main types, called, respectively, the correspondence or the materialist and the idealist conception of truth. Since the first one is derived from Aristotle Metaphysics, in Poland it has also been called the classic, and the second, which does not make use of the correspondence relation, the non-classic definition of truth. Schaff often refers to the latter as the ‘theory of subjective truth’, set over against that of ‘objective truth’. ‘Subjective’ and ‘objective’ are to indicate respectively that there is nothing beyond the belief in virtue of which a statement is true or false, or that there is something extraneous, as it were, on which the truth of the belief depends. In the first case what makes a belief true does, and in the second does not involve the mind of the individual who holds the belief, and, consequently, a statement is true irrespective of whether it has ever been actually made or is made by anybody at any time.
The materialist theory of truth assumes that there is a reality outside the mind and independent of it, which the mind can know, and the idealist theory does not need any of these assumptions. According to Schaff, the idealist theory of truth implies more than that. The common characteristic of the idealist conceptions of truth is their denial that objective reality is knowable or that it exists at all, and, consequently, they define truth as the conformity of thought with a principle or a norm. Some examples of the idealist conceptions of truth are provided by the theories which identify truth with self-evidence, consensus communis, coherence, economy of thought or utility.
The correspondence or materialist conception assumes that a belief is true if it agrees or accords with objective reality. The definition implies that there is an objective reality, that it existed before man appeared on the surface of the Earth and that it exists irrespective of whether anybody perceives it or not. This provides the criterion for distinguishing a pseudo-correspondence theory of truth from a genuine one. In Schaff’s opinion, when this criterion is applied most of those who profess the correspondence theory turn out to be concealed adherents of the idealist conception of truth.
Thus, Aristotle definition in Metaphysics is not an instance of the correspondence theory of truth sensu stricto, that is, a materialist theory of truth. Aristotle’s definition is a brilliant intuition of a man of genius which in Marxism-Leninism has been raised to a higher level of rational perfection. The Thomist definition of truth verbally refers to Aristotle’s formulation but it reduces the adaequatio rei et intellectus to the accord of man’s thought with the thought of God. Most of the great thinkers of the past – Descartes, Leibniz, Kant – paid tribute to Aristotle but the manner in which they conceived what thought must correspond to in order to be true, led them away from the correspondence theory sensu stricto.
In modern philosophy thinkers who define correspondence as agreement of thought with facts must be distinguished from those who define it as agreement with experience. The latter adhere to idealism, for they are driven to the conclusion that only experienced events occur and that it is meaningless to speak of events which no one experiences. This imposes mutilating restrictions upon knowledge and identifies existence with cognition: esse est cognoscendi. On the other hand, those who speak of the correspondence of thought with facts, among whom are Kotarbiński, Ajdukiewicz and Tarski, respect the letter but not the essence of the materialist theory of truth. They are left somehow suspended between materialist truth and idealist error. Only the Marxist-Leninist conception provides a perfectly consistent definition of truth as well as a comprehensive theory that is able to give a completely satisfactory answer to every question concerning truth and falsehood. The materialist theory of truth comprises the claim of being itself an incontrovertible truth.
The superiority which the materialist theory of truth claims for itself is based on two arguments. A satisfactory solution of the question of what the truth of a belief consists in should define the terms between which the relation of correspondence holds. Moreover, the nature of the correspondence relation should also be defined. The supporters of the materialist theory of truth assert that they are able to accomplish both these tasks with a greater precision and material adequacy than any other theory.
So far as the terms of the correspondence relation are concerned, the materialist theory assumes that the predicates ‘true’ and ‘false’ refer primarily to beliefs, and derivatively to sentences in which they are expressed. Whatever a belief might be, it should not be considered as a psychological phenomenon (the act of believing). When a belief is said to be true or false, what is meant is that someone believes that something is or is not the case or that that which is believed is true or false.
The view that only beliefs are primarily true or false is not a new doctrine. In Poland it was held universally and specifically by Twardowski, Kotarbiński – by the latter with certain reservations – and Kokoszyńska, and abroad by numerous thinkers. The only difference between them and Schaff is that they accept the existence of some simple beliefs which are not necessarily expressed in words and Schaff does not. Furthermore, the commonly accepted interpretation of the correspondence theory asserts that the property of truth and falsehood does not depend on any internal characteristic of beliefs, but on the relation of the beliefs to something else outside thought. About what this something else is, there are a number of views, as many thinkers, among them Marxist-Leninists also, rightly emphasise.
The most wide-spread opinion maintains that the truth and falsehood of beliefs depends on their relation to the events or states of affairs or facts. In Schaffs view, it depends on their relation to objective reality, which is an unambiguous expression, while the others are not. This is a moot point, if not a highly doubtful one (unless ‘reality’ simply means ‘the sum of particular events and objects’). It is not more but less difficult to define the terms ‘event’, ‘state of affairs’ and ‘fact’ than ‘objective reality’. If anything, the first three start with some advantage of concreteness which the fourth lacks. We can observe events and directly apprehend states of affairs and facts, but objective reality is an abstract term. Having recorded a measurement or a count, having applied any recognised empirical procedure or used the evidence of our senses, we cannot say that we have measured or counted or investigated objective reality. The intension of this concept might be more or less clear to us, but its extension is not. ‘Objective reality’ seems to be a term about which endless disputations have been carried on and are bound to continue in the future. For this reason it has been abandoned or avoided as much in Poland as abroad. It has not always been used in a metaphysical sense, but its metaphysical associations make it misleading.
The view according to which ‘true’ and ‘false’ are primarily predicated of beliefs and derivatively of sentences eliminates a certain metaphysical doctrine which for the sake of brevity will be called ‘the hypothesis of propositions’. According to this hypothesis a proposition is not a linguistic entity, for it is not a part of any language at all, but a postulated or inferred entity, obtained by abstraction from language. This ideal entity is what a sentence in a given language and its translations into other languages have in common, the ‘content of meaning’ expressed by a declarative sentence whatever form the words used in it might have. It is assumed that propositions do not exist in the way material objects exist. Nevertheless, they are objectively real and are apprehended by the mind when the mind asserts or judges. These timeless entities, known only to some logicians, mathematicians and philosophers, are said to be in some cases-namely when the truths of mathematics and logic are considered – one of the terms between which the correspondence relation holds. On this view a sentence is true if it represents a true proposition, and false otherwise.
In Poland, it was Łukasiewicz and Ajdukiewicz who seemed to have supported the hypothesis of propositions without ever being committed to it. On the other hand, Leśniewski, Kotarbiński, Czeżtowski and Tarski clearly rejected it. Kotarbiński devoted much of his energies to the criticism of this and similar hypotheses concerning the existence of abstract entities. Schaff mentioned Bertrand Russell as the chief contemporary supporter of the hypothesis of propositions, which is incompatible with the standpoint of materialism and, consequently, inadmissible for the materialist theory of truth. He was not satisfied with its criticism from the viewpoint of nominalism, carried out in detail by Kotarbiński, for the nominalist approach is purely negative and barren. In general, nominalism is as much inadmissible as Platonic realism.
Kotarbiński rejected Frege’s differentiation between particular judgments and their objective contents, between sentences as syntactical entities, series of printed or uttered signs, taken in abstraction from their meaning, and propositions which are named or represented by sentences. These distinctions separate what actually can be differentiated only in thought. For in fact there are no judgments or assertions either in the psychological or in the logical sense, there are only individuals who make judgments or assertions. When verbalised, the meaning of the latter is inseparable from the form of words or other symbols by means of which they are expressed. A sentence in Kotarbiński’s sense is a series of sounds or inscriptions together with the meaning which they convey. Kotarbiński did not object to the use of the expression ‘judgment’ or ‘sentence in the logical sense’, provided that it was clearly understood that it was an abbreviated expression which did not name anything and by means of which the meaning of a sentence could be concisely referred to. This is the present-day un-hypostatised use of the term ‘proposition’ which replaces the word ‘sentence’ whenever we are not concerned with a particular indicative sentence but with any sentence irrespective of the language in which it is expressed and which has the same meaning. Schaff rejected this kind of propositions too. It is not clear why he should have wished to dispense with them and to raise thereby almost insuperable difficulties in accounting for the existence of objective truths, i.e. propositions whose truth is not dependent upon the mind.
The hypothesis of substantial propositions is not the only theory which makes possible the objectivity and permanence of the truths of logic and mathematics, believed to be true before anybody thought of them and irrespective of whether anybody actually thinks them or not. This time-independent meaning of the term ‘true’ is accepted by Marxist-Leninists. An objective or absolute truth is ‘not dependent upon man and mankind’ .
The objectivity of the truths of logic and mathematics does not require that these truths be independent of minds for their existence; there would be no propositions if there were no men to think of them and it is actually a matter of accident whether anybody happens to think of them or not. The essential point is that a proposition of logic and mathematics should not depend upon minds for its truth, that is, that what makes it true should not involve the mind which happens to think of it. We don’t need the hypothesis of substantial propositions if some other theory would account consistently for the fact that though the propositions are created or constructed by man, their truth or falsehood is not. For such a theory can maintain that though propositions are minddependent, their truth-value does not depend on anybody’s whim, and that there are true propositions which have not been thought and might never be thought by anyone.
A puzzling feature of the materialist theory of truth is its adherence to the objectivity and permanence of true propositions while it simultaneously asserts that only particular judgments or thoughts and sentences are true or false. For if only judgments made and uttered by particular persons at a particular time were true or false, objective truth would be unaccountable for. At every moment there are countless propositions of logic and mathematics which nobody happens to think. Consequently, the permanence of objective truth, its independence of man and mankind’, would vanish. If only particular judgments are true and false, every true proposition is a mind-dependent truth.
Schaff follows Kotarbiński and assumes that judgements in the psychological sense are not entities of some sort but cognitive experiences of particular persons. Moreover, thinking is always done in sentences and to sever judgments from sentences is impermissible. We cannot really speak of judgments and sentences as separate units. There are only ‘judgment-sentences’. We may speak of true and false sentences but the sentences are true or false in virtue of the judgments associated with them, which impart to the former their truth-value.
This, however, makes the acknowledgment of objective truth still more difficult to account for. What is judged, the content of judgment, becomes private to every mind and we have no means of finding out whether what one person thinks is also thought by another. If thoughts are to be exchanged between people, they must possess a modicum of verifiable identity. Apart from syntactical rules, the meaning of the sentences in which thoughts are expressed is usually considered to constitute the bridge of communication between private worlds of thought and experience. This bridge is destroyed if the understanding between people is said to be ultimately based on thoughts themselves, which are not public and directly accessible for comparison. The possibility of communication by means of language becomes extremely doubtful and incomprehensible if the common meaning of sentences that establishes the understanding is replaced by the putative identity of private thoughts.
The ‘simple phenomenon of reflection’ provides, according to Schaff, a satisfactory answer to those objections. The same objective reality is reflected by different minds and at different times; that which accounts for the identity of mental reflections lies outside the individual in the external world. In virtue of the common reference, thoughts have not the private character which is ascribed to them. The objective reality also safeguards the objectivity of truth, its independence of someone’s happening to think it.
This seems to be, however, the same error as Engels’ when he spoke of the ‘properly made and properly used perceptions’ as a means of finding out what the external world is like. The materialist theory of truth starts at the wrong end; what it is supposed to establish, it assumes as its premiss. The assumption is clearly not capable of being falsified. It can be maintained, if that is somebody’s wish, but it is not a justifiable and rational assumption. The concepts of objective reality and mental reflection, as used in the materialist theory of truth, are metaphysical in the clearly pejorative sense of this term. They apply to objects and processes which we have never experienced and which in the nature of the case it is impossible that we should ever experience. Consequently, the explanation which they purport to provide is hollow in significance.
It is hard to see how the theory of reflection could account for the objectivity of truth. The difficulties might not be apparent if some simple singular ‘judgmentsentences’, like ‘ Warsaw is situated on the Vistula’, are considered. But they become insuperable if the truth of such propositions as ‘two times two makes four’ or ‘all men are mortal’ is to be explained by means of the reflection metaphor. For objective reality does not include anything that these statements could possibly reflect. There is nothing in objective reality that would correspond to what they state in the same manner as the fact of the city of Warsaw extending on both banks of the Vistula corresponds to the singular statement ‘ Warsaw is situated on the Vistula’. In such cases the explanation in terms of reflections and objective reality not only lacks rationality, but also becomes meaningless. Moreover, if Lenin’s assertion ‘that an image cannot exist without the thing imaged and that the latter exists independently of that which images it’ is taken seriously, the reintroduction of the rejected hypothesis of substantial propositions seems to become inescapable. There are some good reasons to doubt the justice of Schaff’s claim that only the materialist theory of truth is capable of refuting the ‘mysticism of abstract entities’ .
The problem of the objectivity of truth cannot be more closely examined without inquiring into the second argument for the claim that the materialist theory of truth is superior to any rival theory. This argument asserts that only the materialist theory is capable of providing a satisfactory definition of the correspondence relation. According to the traditional definition, a judgment is truly made and a proposition is true if and only if there is an event or a state of affairs or a fact corresponding to what is judged or asserted, and false otherwise. Like many writers before him, Schaff points out that if the truth of a judgment or of a proposition consists in the correspondence of this kind, the term ‘correspondence’ should be defined. For neither this term nor any of its equivalents commonly used, such as ‘agreement’, ‘accord’, or ‘adequacy’, has a precise and sharply defined meaning; consequently, we are never sure whether it does or does not apply in a particular case.
It is Schaff’s contention that the term ‘reflection’ has none of the ambiguities shared by the other relational expressions. Accordingly, the materialist theory defines the term ‘true’ in the following manner ‘A judgment (‘a judgmentsentence’) is true if it faithfully reflects objective reality’. This definition, Schaff states, puts an end to all the imprecisions with which the traditional formulation of the classic definition of truth has been burdened since the time of Aristotle.
Attention has been drawn to the fact that this definition, understood literally, is a very ancient view, professed by the Greek atomists, and that the copy interpretation of the correspondence relation had been abandoned by Aristotle and Thomas Aquinas for the exact reason that it cannot apply in the literal sense. It could not be said, for instance, that the proposition ‘ Warsaw is situated on the Vistula’ is in the least like the state of affairs designated by the proposition. We can walk about the streets of Warsaw and take a swim in the Vistula, and obviously, we cannot do any of these things with the proposition. It is very odd to say that the terms in the proposition ‘ Warsaw is situated on the Vistula’ are related in the same fashion as is Warsaw and the Vistula. What is reflected in the proposition stands for but is not itself a spatial relation that binds the objects denoted by the terms ‘ Warsaw’ and ‘the Vistula’. This is, however, to say that the mental reflection resembles what is reflected in one respect and not in another. But if ‘reflecting’ is a metaphoric relational expression and cannot be taken literally, it is not an unambiguous term and, in view of its misleading suggestions, not an unobjectionable one. The term ‘correspondence’ in an unspecified sense is as good as, if not better than, that of ‘reflection’ and if instead of the former the latter is applied it is self-deceptive to suppose that a step forward has been made.
This is the conclusion which Schaff had to accept, though only by implication and by using the old terminology in a new manner. He repudiated the suggestion that the ‘theory of reflection’ implies the identity of relations between elements in each term. It is enough to assume that the relations are ‘analogous’, and by this he probably meant what has just been said, namely, that they do resemble each other in some respects and do not in some others. This means, however, that so far as the definition of the correspondence relation is concerned the materialist theory of truth has nothing new to offer.
This is also made clear by Schaff’s adherence to the semantical definition of truth which on the one hand offers a more precise form of the original Aristotelian formulation and, on the other, is neutral with respect to various epistemological attitudes. In Schaff’s view the proposition ‘a true (correct) judgment is a reflection of objective reality’ is logically equivalent to “ ‘p’ is true if and only if p” .
While recognising that the reflection relation is a metaphor, Schaff considered that the metaphor holds a strategically important position with respect to two problems: to the classic poser of true negative propositions and to the antinomies which arise from the use of self-referring sentences. There is apparently no adequate solution of these problems outside the materialist theory of truth.
A satisfactory theory of truth should be able to give an account of the opposite of truth, that is, of error. This condition has not always been fulfilled in the past and H. H. Joachim in The Nature of Truth, published in 1906, made use of this omission for the justification of the coherence theory of truth. Taking Joachim as his witness, Schaff assumed that a non-materialist theory of truth is incapable of accounting for false beliefs unless it postulates the existence of the non-existent facts, that is, unless it is self-contradictory. This conclusion is wide of the mark.
Truth and falsehood are correlative properties of beliefs; if there were no false beliefs, there would be no true beliefs either. To have both true and false beliefs, we need sentences and propositions on the one hand, events, states of affairs or facts on the other. This does not mean, however, that sentences or propositions name events, state of affairs or facts in a similar manner as proper names or predicates name something that is there. If a proposition were true in virtue of naming a fact, what would there be for a false or a true negative proposition to name? Sentences or propositions do not name but refer to or state, if true, a fact. The problem of what false sentences and true negative sentences are about largely arose from the confusion of the semantical relation of naming and referring.
G. E. Moore and Bertrand Russell realised a long time ago that if a theory of truth is to account for falsehood it must abandon the assumption of beliefs being a two term relation of the mind to a single object. The analysis of belief which reduces it to the relation between the believing person on the one hand, and what is believed (the fact or the proposition) on the other is incorrect. Consequently, Russell suggested that judging and believing should be conceived as a multiple relation which unites several terms into a complex whole. Sentences or propositions are constructed by the act of judgment. They are true, if to their relational complex whole there correspond events, states of affairs or facts which are also a relational complex whole. If the correspondence relation does not hold, the sentences or propositions are false and our beliefs or judgments are erroneous.
Russell’s theory has been widely accepted, and it underlies Kotarbińki’s and Ajdukiewicz’s formulation of the correspondence theory of truth. Schaffs solution of the classic philosophical poser is in essence that of Moore, Russell, and the Polish philosophers, though their clarity and freedom from ambiguity is obscured in Schaff’s exposition by the persistent idea that judgment is a mental reflection, in some sense of this term, of a single fact.
The idea that a judgment is a mental reflection of objective reality is claimed to provide an epistemological solution of the antinomies arising from the use of the term ‘true’ in colloquial language. Schaff argued that if the theory of reflection is accepted, the antinomies disappear, for statements like ‘I am lying’ or ‘c is not a true sentence’ in Tarski’s example turn out to be no sentences at all but disturbances of air waves or patches of ink. On the other hand, if the theory of reflection is ignored, there is no valid reason simply to forbid the use of such sentences. Thus, we would be prevented from distinguishing truth from falsehood and the concept of truth in colloquial language would lose all significance. It has been already explained that this conclusion is fallacious and misinterprets the consequences of Tarski’s investigations.
There is a basic difference between Schaff’s and Tarski’s approach to the problems of truth. For Schaff assumes that we can ignore the questions of language when we investigate ‘objective reality’ and that what reality is can be fixed in abstraction from and independently of the language adopted for its description. On the other hand, Tarski followed the path now familiar to contemporary thinkers: the answer to any question concerning reality depends upon reality itself, but also upon the language chosen for its description. Consequently, every question of fact has its linguistic aspect. The investigations of the structure of language, in which facts are stated, is a part of the investigations concerned with the facts themselves. This difference of approach was apparent in the whole Marxist-Leninist criticism of the Warsaw school, and it turns up again in the examination of the concept of truth.
Tarski’s essay is largely concerned with the question of the bearing that the logical structure of language has upon the definition of the concept of truth. If all the results of his examinations had to be fully accepted, the concept of truth in colloquial language would still be not devoid of significance as much as that of a square circle. Apart from the knowledge provided by linguistic rules and definitions, there are other means by which true and false sentences can be distinguished. We have an intuitive knowledge of the concept of truth and in this sense we can be perfectly acquainted with it in spite of having failed to analyse it and to provide a satisfactory definition of a true sentence. Tarski never denied that we are familiar with the concept of truth and he drew upon our understanding of it in proposing his semantical definition of truth.
If by a proposition is meant anything that is believed or disbelieved, a sentence is a verbal expression of a proposition. Since Aristotle we assume that not every sentence expresses a proposition but only that which is either true or false. Sentences of this kind are usually called ‘logical sentences’. Schaff’s assertion is that not every logical sentence is a genuine logical sentence, and only the latter can be either true or false. A logical sentence is a genuine sentence if it expresses either directly or indirectly a mental reflection of objective reality. Some selfreferring sentences, like ‘I am lying’ or ‘c’ in Tarski’s example of ‘c is not a true sentence’, are not verbal expressions of a cognitive experience, and they do not express a judgment that reflects objective reality. Consequently, neither ‘true’ nor ‘false’ can legitimately be predicated of them. No antinomies arise if the use of ‘true’ and ‘false’ is restricted to sentences which express either directly or indirectly a judgment. For primarily only a judgment which does or does not faithfully reflect objective reality is true or false.
What Schaff has in mind is a hierarchy of mental reflections extending indefinitely upwards and downwards based solidly on reflections of what is there. The idea is exactly the same as that of a hierarchy of languages, formulated by Bertrand Russell. A hierarchy of languages presupposes a language of the lowest type, called by Russell the ‘object language’ or the ‘primary language’. What are the implications of this idea with respect to the truth and falsehood of sentences in the object language is best explained in Russell’s own words:
“It is clear from Tarski’s argument, that the words ‘true’ and ‘false’ cannot occur in the primary language; for these words, as applied to the sentences in the nth language, belong to the (n+ 1)th language. This does not mean that sentences in the primary language are neither true nor false, but that, if ‘p’ is a sentence in this language, the two sentences ‘p is true’ and ‘p is false’ belong to the secondary language. This is, indeed, obvious apart from Tarski’s argument. For, if there is a primary language, its words must not be such as presuppose the existence of a language. Now ‘true’ and ‘false’ are words applicable to sentences, and thus presuppose the existence of language. . . . . In the primary language, therefore, though we can make assertions, we cannot say that our own assertions or those of others are either true or false.” 
If we had to express the same ideas in terms of a hierarchy of mental reflections, as Schaff suggested, we would at once be lost in confusion. We cannot imagine going indefinitely upwards from a reflection of a lower to that of a higher type, from a cognitive experience of the nth to that of the (n+ 1)th level. Confusion would creep in almost at the start and we would not know what we were talking about. This is not the case with a hierarchy of languages, for we have a symbolism ready at hand and the various languages can be neatly differentiated from each other. While no advantages are gained,clarity of thought is lost, if a hierarchy of mental reflections is substituted for Russell’s hierarchy of languages.
The merits of the theory of reflection are in this case not only illusory, but in fact non-existent. We do not need to assume some kind of linguistic metaphysics, that is, to reduce thought and knowledge to the use of language, in order to consider the analysis of language and linguistic expressions as a fruitful approach to the examination of thought and knowledge. Schaff probably felt that the latter cannot do without the former and came to the conclusion that Russell’s hierarchy of languages must be expressed in terms of the theory of reflection to become acceptable to a Marxist-Leninist.