Taking, then, first the problem of the reflexive voice, we will formulate it in the following way. Can the group "verb + self-pronoun" (i. e. myself, himself, ourselves, etc.) be the reflexive voice of a verb, that is, can the self-pronouns ever be auxiliary words serving to derive a voice form of the verb? This is putting the problem in purely morphological terms. But it also has a syntactical side to it. From the syntactical viewpoint it can be formulated in another way: does a self-pronoun coming after a verb always perform the function of a separate part of the sentence (the direct object), or can it (in some cases at least) be within the same part of the sentence as the verb preceding it (in the vast majority of cases this would be the predicate)? If we approach this question from the point of view of meaning, we shall see that different cases may be found here. For instance, in the sentence He hurt himself badly we might argue that himself denotes the object of the action and stands in the same relation to the verb as any other noun or pronoun: he hurt himself badly would then be parallel to a sentence like he hurl me badly. On the other hand, in a sentence like He found himself in a dark room things are different: we could not say that he found himself is analogous to he found me. We could not, indeed, say that he performed an action, that of finding, and the object of that action was himself. Here, therefore, doubt is at least possible as to whether himself is a separate part of the sentence, namely, a direct object, or whether it is part of the predicate. We might possibly have to class he hurt himself and he found himself (in a dark room) under different headings and this would influence our general conclusions on the category of voice. Considerations of this kind cannot, however, bring about a solution that would be binding and could not be countered by a different solution which might also be confirmed by more or less valid reasons. If we are to achieve some objective solution, we have to rely on objective data in this case, as in so many other cases. Objective investigation requires that we should find various syntactic contexts or patterns in which the group "verb + selfpronoun" can appear. For instance, we ought to look for examples of the pattern "verb + self-pronoun + and + noun or pronoun". If such examples can be found, they will argue in favour of the view that the self-pronouns standing after a verb are actually treated as standing in the same relation to the verb as any other noun or pronoun denoting the object of the action. If, on the other hand, no such example could be found, this would go some way towards proving that a self-pronoun is not apprehended as standing in the same relation to the verb as any other noun or pronoun following it, and this would be an argument in favour of acknowledging a reflexive voice in the Modern English verb. Other considerations of a syntactical character might also influence our judgement on this question. The problem has been treated by O. Ovchinnikova,1 who has collected some examples of the pattern "verb + self-pronoun + and + noun or pronoun", for instance, / see this man Meek doing everything that is natural to a complete man: carpentering, painting, digging, pulling and hauling, fetching and carrying, helping himself and everybody else ... (SHAW) and also examples of a noun functioning as apposition to the self-pronoun which comes after a verb, e. g. / am defending myself — an accused communist. (FOX) These cases, few as they are, show that a self-pronoun following a verb can at least be apprehended as a separate member of the sentence. If it were only part of the predicate it obviously could not have an apposition attached to it. So we may take it as proved that in some cases at least the self-pronoun following a verb is not an auxiliary word serving to express a voice category of the verb. But the question remains, what we are to make of cases such as the following: It was done, and Catherine found herself alone in the Gallery before the clocks had ceased to strike. (J. AUSTEN) Here the self-pronoun cannot either be joined by and to a noun (pronoun), or have a noun in apposition attached to it. Without going into many details concerning these cases, we can merely say that two ways are here open to us. One way is to say that, since in a number of cases the self-pronoun is not an auxiliary word used to form a verbal voice, it is never an auxiliary. Then we should have to treat such cases as he found himself . . . etc. as phraseological units and refer their peculiarities to the sphere of lexicology rather than of grammar. The other way would be to say that in some cases a self-pronoun does become an auxiliary of voice. Then to find oneself would be treated as a form of the reflexive voice of the verb find and the group (and, of course, other groups of a similar kind) would remain in the sphere of grammar and we should recognise a reflexive voice in English. There seems at present no binding argument in favour of one or the other solution. We shall have to leave the question open until such a solution can be found. The treatment of the problem would be incomplete if we did not mention the cases when a verb is used without a self-pronoun to denote an action which the "doer performs on himself. Examples of this kind are not numerous. We can mention the verb dress, which may be used to mean 'dress oneself, and the verb wash, which may be used to mean 'wash oneself'. This is seen, for example, in sentences like the following: At daybreak the next morning Hame got up and dressed. (E. CALDWELL) As we see, these verbs denote habitual everyday actions and this appears to be essential for the possibility of such a usage. It would not, for instance, be possible to use the verb hurt in the sense of 'hurt oneself', or the verb accuse in the sense of 'accuse oneself', etc. Since in the sentence he dressed quickly there is no self-pronoun and no other special sign to indicate that the doer is performing the action on himself, we cannot include such cases under the category of the reflexive voice even if we were to recognise the existence of such a voice, which, as we have seen, cannot be objectively established. Cases of this kind will best be considered together with the problem of the middle voice, which see