Rashi's Biblical and Talmudic commentaries contain 3157 laazim, of which 967 occur in the Biblical commentaries and 2190 in the Talmudic, forming in the two commentaries together a vocabulary of about two thousand different words. In the Biblical commentaries, concerned, as a rule, not so much with the explanation of the meaning of a word as with its grammatical form, the laazim reproduce the person, tense, or gender of the Hebrew word; in the Talmudic commentaries, where the difficulty resides in the very sense of the word, the laazim give a translation without regard to grammatical form. At the present time these laazim are of interest to us, not only as the expression of Rashi's ideas, but also as vehicles of information concerning the old French. As early an investigator as Zunz remarked that if one could restore them to their original form, they would serve as a lexicon of the French language at the time of the Crusades. But even Zunz did not realize the full value to be extracted from them. The rare specimens that we possess of the <I>langue d'oil</I>[55] of the eleventh century belong to the Norman dialect and to the language of poetry. Written, as they were, in Champagne, the laazim of Rashi represent almost the pure French (the language spoken in Champagne lay between the dialect of the Ile-de-France and that of Lorraine [56]), and, what is more, they were words in common use among the people, for they generally designated objects of daily use. These laazim, then, constitute a document of the highest importance for the reconstruction of old French, as much from a phonetic and morphologic point of view, as from the point of view of lexicography; for the Hebrew transcription fixes to a nicety the pronunciation of the word because of the richness of the Hebrew in vowels and because of the strict observance of the rules of transcription. Moreover, in the matter of lexicography the laazim offer useful material for the history of certain words, and bring to our knowledge popular words not to be found in literary and official texts. In the case of many of these terms, their appearance in Rashi is the earliest known; otherwise they occur only at a later date. And it is not difficult to put the laazim back into French, because of the well-defined system of transcription employed. Even the laws of declension (or what remained of declension in the old French) are observed.