Within the Vedic verbal system there is a series of second singular forms made by suffixing -si to the full-grade root, which differ from the present type ási ‘thou art’ in that they function primarily as imperatives; e.g. yákṣi ‘sacrifice’. This archaic formation is found frequently in the Rigveda, where approximately 150 forms occur from 23 roots (§2); in other Vedic texts -si imperatives occur only for roots which have such forms in the Rigveda and predominantly in Rigvedic mantras; e.g. maghónām ‘bring safely across the gift-giving of the patrons’ (RV 8.103.7, 9.1.3); vakṣi yákṣi ca ‘bring the gods and sacrifice’ (.5.21.6, 6.16.2, 8.102.16); see Bloomfield, Concordance sv. párṣi, .

In recent years there have been a number of interesting attempts to redefine the prehistoric stages of Germanic. The need for such a redefinition has long been evident, as van Coetsem, Makaev, and Lehmann have all pointed out. One of the reasons for this need is undoubtedly to be found in the prevailing view of a protolanguage as an entity which is ‘timeless, non-dialectal, and non-phonetic’, and therefore unreal. Pulgram, for example, considers ‘Reconstructed Proto-Indo-European’ (in contrast to ‘Real Proto-Indo-European’) to be a ‘timeless structure which is diachronically heterogeneous, though synchronically uniform’ (425). This characterization of our reconstruction of Proto-Indo-European may be accurate, but it is questionable whether all protolanguages are equally ‘timeless’, at least in the relative sense. The younger protolanguages which we can reconstruct within the Indo-European family, such as Proto-Germanic and Proto-Romance, cannot be placed in an identical frame of reference with Proto-Indo-European. The latter is re constructible only on the basis of internal evidence. The younger protolanguages, however, have the added advantage of disposing over comparative evidence from related languages outside their limited groups to provide them with additional depth. This is perhaps what Bloomfield had in mind when he made an exception of Proto-Germanic and Proto-Romance to his general rule that the comparative method ‘cannot claim to picture the historical process’ (318).

The question ‘What is syntax?‘ has been asked many times. In the framework of traditional grammar it cannot be answered by a definition. Traditional syntax, the kind of syntax upon which all of us have been brought up, is not a theory but a craft. We were taught by practical example how to find (say) the subject or the predicate of a sentence and how to recognize a given form as a 3rd-person singular verb indicative present active. These technical terms were never properly defined; but with a sufficient number of examples for practice plus some rules of thumb, we soon acquired an intuitive grasp of what they meant, and were thus able to work with them satisfactorily. Sometimes classical grammar did offer definitions, but they were invariably poor and clumsy.

In this article I propose a new explanation of the forms of the verb sum. This explanation is based on synchronic facts of Classical Latin and is expressed in terms of a set of ordered rules applying to basic forms. I do not propose that forms appearing in the derivations of the Classical Latin forms necessarily appeared phonetically in any stage of pre-Latin or Indo-European. Nevertheless, I feel that my analysis is not wholly devoid of historical and comparative interest. In this I follow Leonard Bloomfield, ‘Menomini morphophonemics’, Travaux du Cercle linguistique de Prague 8.105–15 (1939):

Four fundamental insights of tagmemics are the correlativity of function and set, the search for constructions of maximum relevance, the emphasis on explicit, systemic hierarchy, and the concept of relatedness in logical space (with transformation only one of the possible parameters which relate constructions). Developing these points gives an indirect answer to certain recent criticisms.