in sentences of the type, 'What shall it profit a man though he gain the whole world?' Old English almost invariably employs deah in such sentences, even when translating a Latin si. Examples: M. 16.26 Hwæt fremað ænegum menn þeah he ealne middaneard gestryne...? L. 9. 25; Bl. H. 55. 4; Inst. 480. 24; HL. 149. 130.

Deah may also be an interrogative particle meaning whether. The reason for including this usage among concessive constructions is that there may be some concessive force in these questions. It is a singular fact that the interrogative use of deah occurs only after nytan, uncuỡ, and possibly weald. Two of these words ar enegative. Examples of nytan deah: Bo. 64.9 Ic nat þeah du wene pæt hi on hiora agenre cyððe ealne weg mægen; 107. 25; 110. 32; 131. 23: Sol. 20. 1; 47. 1; CP. 411. 22; HL. 134. 614: Jos. 9. 7: Cart. 2. 179. 5. It is to be observed that most of these examples are from works ascribed to Alfred, and that the final one is from his will. Of uncuð deah I have found only one instance: LS. 1. 390. 119 Clypiao git hluddor; uncuð þeah þe he slæpe. The meaning is evidently 'perhaps'; one may compare 'Who knows whether?' The singular phrase weald Jeah also means 'perhaps.' Of weald I have been unable to find any satisfactory explanation. When used alone it seems to mean 'lest.' Whether it is to be explained as originally noun or verb cannot be determined. The phrase weald deah is evidently an old formula which can hardly be analyzed. Examples: ÆH. 2. 340. 9 Weald þeah him beo alyfed gyt behreowsung; 466. 10: Jos. 9. 7: Nic. 488. 27, 33.

CHAPTER III.

DISJUNCTIVE CONCESSIVE CLAUSES WITH CONJUNCTION.

The alternative concession, which in Modern English is expressed by an indirect question with whether ... or, was by preference expressed in Old English (at least in the later prose) by parallel inverted clauses without connective. There are, however, two welldefined modes of expressing this relation in clauses with conjunctions.

THE SAM-CLAUSE.

Old English possesses a special particle for alternative concessions, the somewhat rare word sam. This connective is singular in having no exact analogue in other languages. Its conjunctive use suggests a connection with Old Norse sem; but as the New English Dictionary and Wright's Dialect Dictionary, in discussing the English representative of sem (for which see articles on howsomever) do not mention sam, this connection is doubtful. The use of sam, moreover, does not correspond to the use of sem. We should expect sam, like sem, to have relative uses, if sam were the source of the Middle English relative som. But if sam ever was a relative corresponding to sem, that use is lost in the literature we possess. Its meaning, however, may be akin to that of swa, which is also used in disjunctive concessions. As in this case... so in that case' can easily be read into most concessions of this type. It should be noted at the same time

that sam seems related to the series som (noun), samod, samen, gesamnian, and same (in the combination swa same).

If this view is correct, the original use of the correlative probably is, 'the same in this case ... the same in that.' One may perhaps compare the modern colloquial phrases, all the same,' 'just the same,' often used as correlatives to concessions. But this notion probably faded early into the more general one of connection, and the particle has no more specific meaning than correlative swa... swa in its concessive use, or than our comparable whether... or.

The distribution of the word is also curious. It is not included in Grein's Sprachschatz. By far the greater number of occurrences are in the works attributed to Alfred the Great. Yet the word persists, not only in the Winteney version of the Benedictine Rule, which copied it from the older version, but also in a passage of Middle English prose: OE. Hom. 2. 107.8 on þis wis cumeð ech iuel ponc and speche and dede neden uppard, sam it haue angun of þe mannes lichames wille sam it haue þe biginning of the deules for-tuhting.

NOTE. There is one passage in which sam has been interpreted as a simple, not a correlative conjunction: Bl. H. 53. 17 ne sylþ he hit us to pon þæt we hit hydon, oppe to gylpe syllan, sam hwylcum mannum þe naht swipe God ne lufiap ah we hit sceolan... þæm earmestum mannum dælon. Morris translates sam 'or,' a meaning which might arise from the correlative use. But may it not be possible that sam is prepositionally used? 'Not that we should hide it, or give it ostentatiously, like some men who do not truly love God.' This use of sam would arise more naturally from the meaning 'the same' than from a merely relative meaning. It may be, however, that we have here simply the pronoun samhwylc.

Clauses introduced by sam are short, and always, where the mode can be determined, with an optative verb. The correlative group may stand in any part of the sentence: Sol. 24. 1 Sam ic wylle, sam ic nelle, ic sceal secgan nide riht; O. 21. 17 hy gedoð þæt ægper bið oferfroren, sam hit sy sumor, sam winter; Herb. 166. 9 Eft wid wunda, som hy syn of iserne som hy syn of stence oððe fram nædran, genim þysse ilcan wyrte wos. The number of parallel clauses may be greater than two: Sol. 58. 13 þæt ælc man on pisse wurlde swa miclum lufað God swa he wisdom lufað, -sam he hine miclum lufige, sam he hine lytlum lufige, sam he hine mydlinga lufige? The adversative deah follows the sam-clause in this sentence: Bo. 60. 6 sam hi þyrfen, sam hi ne þurfon, hi willað þeah.

Most occurrences of this construction are in original passages, though in translated works. The following sentence, however, is directly from the Latin: BR. 66. 14 Sam hy fæsten, sam hy ne fæsten... arisen hy sona swa hy heora mete hæbben (Et ideo omni tempore sive jejunii sive prandii, ... mox ut...). The same construction is retained in WV. 89. 4. The interlinear version here has, apparently, the alternative clause without conjunction: IG. 74. 8 sit fæstenes sit gereordung; which, as Logeman's foot-note points out, may be a scribal error for sig... sig.

If this understanding of the gloss be correct, the two translations of the Latin phrase illustrate how entirely equivalent is the sam-clause to the correlative clause without conjunction, to be discussed in another chapter. This is further illustrated by the use of sam in the formula sam ic wylle, sam ic nelle (Sol. 24. 1); etc. The equivalent formula without conjunction -as wylle we nelle we (EH. 1. 532. 7)-is a familiar idiom.

CONSTRUCTIONS WITH SWA... SWA...SWÆÐER. The correlative construction formed with swa and the interrogative pronoun hwader is rather frequently employed for disjunctive concessions. These can best be understood by comparison with the general use of the same idiom. It may simply define alternatives: Bo. 91. 19 Nim donne swa wuda swa wyrt, swa hwæðer swa pu wille-'either tree or plant, whichever thou wilt'; Laws 204. 8 Gyf he þonne ne mæge, gewylde man hine swaðor man mæge, swa cwicne swa deadne (Latin version: exsuperetur quibus modis poterit sic uiuus sic mortuus). The alternatives are balanced by means of swa; and are also covered by the indefiniteinterrogative clause with swa hwæder swa or its contraction swæder. The verb of the swa hwæder swa clause, in this more general use, is often, but not always, optative. Swader may become adverbial, the construction otherwise remaining the same, as in the example just cited from the Laws. On this point, cf. Wülfing 1. 449, § 351.

Precisely the same form of clause is used when the sense is concessive, and it is only by the logical relation of clause and context that we can make sure of concessive import. The mode of the verb furnishes no certain test, for though these clauses, when concessive, usually have the optative, the indicative is also found. It is impossible to decide by its form whether the clause in the following passage is concessive or merely modal: Bo. 101. 7 Forðæm du ne þearft nauht swipe wundrian deah we spyrien æfter dæm de we ongunnon, swa mid læs worda swa mid ma, swæðer we hit gereccan magon. The original affords no clue, since the clause was added by the translator. As, however, no other idea is contrasted with the swa... swa clause, I have regarded the latter as modal.