Etymology: In senses 1 and 2 (‘action of riding’), ultimately < the same Germanic base as ride v. In other senses (including the now dominant senses at branch III.), perhaps a semantic development of the same word, although other explanations are also possible, and it is not certain that all of the material contained in this entry shows a single word history. It is also possible that two or more originally distinct words may have merged. (1) In senses 1 and 2 (‘action of riding’), cognate with Old Frisian -rēd (in rāfrēd ride on a stolen horse; West Frisian reed ride), Middle Dutch rēde ride, riding (Dutch regional ree , reed , rede ), Middle Low German rēt , rēit journey on horseback, military expedition on horseback, cavalry unit, Old Icelandic reið riding, raid, Old Swedish redh ride, showing the reflex of a Germanic noun from an ablaut variant of the same base as ride v.; compare also, with the same ablaut grade, Old High German reita carriage, chariot, vehicle (also Old High German reitī carriage, vehicle), Old Icelandic reið carriage, vehicle, and also Old High German reit- in compounds (as reitros : see road horse n.). (2) In sense 3 (‘place where ships ride’) apparently cognate with Middle Dutch reede , rēde (Dutch ree , rede ), Middle Low German rēde , reide ( > German Reede , and also > Old Swedish redh (Swedish redd ), Danish red ), all in sense ‘roadstead, place where vessels may safely lie at anchor’, which may all ultimately be from the same Germanic base as the forms cited above (compare to ride at anchor at ride v. Phrases 1a and the corresponding specific use of the West Frisian, Dutch, Middle Low German, German, Swedish, Old Danish, and Danish cognates of ride v.), although it is also possible that they may instead be from the same Germanic base as rede v.² and ready adj. (3) Sense 4 (‘path or way’), and hence the other senses at branch III., may show a semantic development from the earlier senses, and this has long been the usual assumption by most scholars. Although this sense is of very late first occurrence in English, compare West Frisian reed dirt road and early modern Dutch reed route, way (1642 in an isolated attestation), Dutch regional ree , reed , rede track or trace (of a wheel, an animal, etc.), track for wagons or carts (also ‘drainage ditch’), which suggest either a parallel semantic development or a shared sense which happens not to be attested earlier. Compare also the occurrence of -rād in Old English as the second part of compounds in the sense ‘place where a person or thing rides (in various senses of ride v.)’ (see below). However, it is also possible that branch III. may show a different origin, as the reflex of Old English rodu clearing, probably also linear clearing, track, way, road (attested only in charters and in place names: see below), probably ultimately from the same base as rid v. Old English rodu may also be the origin of Scots rod n.², the recorded spellings of which indicate the existence of variants with both a short and a long vowel (by Middle English lengthening in open syllables); such forms cannot be direct developments of Old English rād , since the expected developments of this in Scots would be *raid , *rade (see raid n.), and it seems unlikely that all such examples are attributable to southern English influence. Additionally, in modern regional varieties from the north of England the word typically (but not invariably) shows forms that are expected developments from Middle English open ō , not from ā (unless via southern forms with vowel rounding). It is possible that a Middle English form rode as reflex (with lengthening in open syllables) of Old English rodu may also underlie branch III. See further on this hypothesis T. Turville-Petre ‘The Etymology of Road’ in Notes & Queries 55 (2008) 405–6. Compare also place-name forms cited at cock-road n. However, this explanation leaves various difficulties, such as why apparently unlengthened forms as well as lengthened forms are found at rod n.² (although failure of lengthening in open syllables is sometimes found in Older Scots), and the occurrence of occasional forms suggesting development from ā in regional varieties from the north of England. It is possible that two words of distinct origins may have merged, or at least that one may have reinforced the other. See also further note below on the difficulties presented by Older Scots forms found in branches I. and II. in this entry, where forms developed uniformly from ā should be expected. A further complication is presented by words of similar form and meaning in Celtic languages. Early Irish rót main local road cannot be related (on formal grounds) to Old English rād. There is no reason to see Scottish Gaelic ròd way, road, path, track as anything other than the continuation of this word, although its meaning ‘anchorage, roadstead’ is probably due to influence from the English word of similar form and meaning. The possibility cannot be ruled out entirely that the Scottish Gaelic word also exercised some influence on the complex developments found in Older Scots, and perhaps, via Scots, on the wider use of road. See further on the evidence from Celtic languages T. O. Clancy ‘Spaghetti Junction’ in S. Arbuthnot & K. Hollo Fil súil nglais (2007) 17–28.In Old English both rād and rodu are strong feminines of the ō -stem declension. For charter evidence for Old English (inflected) rod- (probably reflecting an Old English nominative rodu ) in the sense ‘linear clearing’, compare the following:eOE Charter: Bp. Swiðwulf to Beorhtwulf (Sawyer 1276) in W. de G. Birch Cartularium Saxonicum (1887) II. 202 West be ðy wioda andlanges ðare rode oð ðane pyt.OE Charter: Bp. Ealdred to Dodda (Sawyer 1405) in D. Hooke Worcs. Anglo-Saxon Charter-bounds (1990) 368 Of ðære dune andlang þære rode oð hit cymð beneoðan stan cnolle.lOE Bounds (Sawyer 699) in W. de G. Birch Cartularium Saxonicum (1893) III. 292 Of waddene andlang þære rode innon syx æceres. Evidence that Old English rodu also had the sense ‘road, way’ is provided by the following reference to a section of Roman road in a late copy of an early 11th-cent. charter:a1400 ( Bounds (Sawyer 943) in C. Hart Early Charters of Eastern Eng. (1966) 190 Of ottanmere þuyrs ouer bugenroda, of bugeroda into mærmer.The Roman road referred to, part of which is still called Buckle Street, ran from Bicester in Oxfordshire to Stow in Buckinghamshire; compare the names Buggilderode (1226), Buggerode (a1272) for other sections of this road, beside Bogildestret (c1220), Buggestret (11th cent.), Bugildewaye (c1250), etc. It is not possible to determine the length of the vowel in Old English (inflected) rod- with complete certainty (and so confirm the assumption that it ultimately derives from the same base as rid v.). Compare, however, the following place names which apparently show the same word: Rodo , Somerset (1086, now Rode), Rode , Northamptonshire (1086, now Roade); for further discussion of the place-name evidence for this word see M. Gelling & A. Cole Landscape of Place-names (2000) 243–4 and also M. T. Löfvenberg Stud. Middle Eng. Local Surnames (1942) II. 167–8. Aside from the question of possible connection with rod n.² noted above, the Scots forms present various points of difficulty. The expected Scots development of Old English rād is shown by raid n. The Older Scots forms listed in the present entry for uses in senses at branches I. and II. are difficult to explain (for examples compare quots. 1584-5 at sense 1a, 1544 at sense 1c, 1542, 1584, 1672, a1699, 1817 at sense 2, 1545, 1667 at sense 3a). It seems unlikely that all of them show southern English influence (especially in the case of spellings apparently reflecting a short vowel, since such forms are very rare outside Scots). Confusion or association with rod n.² is perhaps possible, but it is hard to see the basis for this within Scots; perhaps the existence of Scots rod n.² alongside southern English road in branch III. led also to analogous use of similar forms in branches I. and II. (The length of the vowel of Older Scots rod is uncertain. Examples have been assigned to the γ. forms, although it is possible that some may show a long vowel.) With sense 1a compare also radknight n. With sense 1c compare Old English ēored , ēorod troop, band, legion (of soldiers, demons, angels), cavalry ( < eoh horse (see hippo- comb. form) + rād ). The derivation of this very early compound had become obscured already in Old English as is shown by the phonology and semantics of the word, and also by the predominantly neuter inflection (probably after semantically similar weorod wered n.). With sense 3 compare ( < English) Anglo-Norman rode (compare quot. 1325-6 at sense 3a), Anglo-Norman and Middle French rade , radde (14th cent.; French rade ), post-classical Latin roda , rooda , rada (from 13th cent. in British sources). Earlier currency of sense 4 is sometimes taken to be shown by certain Old English compounds in which -rād occurs as a second element with apparent locative force in sense ‘place where a person or thing rides (in various senses of ride v.)’, compare hwēolrād mark left by a wheel, wheel-rut, strēamrād bed or channel of a stream (also in poetic use with reference to the sea, but here probably rather with dynamic force denoting the coursing of the current), and three poetic compounds characterizing the sea or a part of the sea as a place where the whale (or swan or sailing ship) rides: hronrād , swanrād , seglrād . Although the second element in the poetic compounds is frequently glossed as ‘road, way’, it is uncertain whether any of these compounds implies the element of direction that is a core feature of sense 4. (The compounds brimrād and wīgrād are also sometimes cited in this context; however, brimrād , often glossed ‘sea-road, sea’, probably refers rather to the coursing or surging of water, and wīgrād , included in Bosworth-Toller and later dictionaries of Old English with suggested meaning ‘war-road, road along which an army passes’, is a ghost word based on a disputed emendation of wigrode in Genesis A 2084.) For further discussion of the Old English compounds see C. Brady in PMLA 67 (1952) 538–71. It is uncertain whether the following, describing Jonah's perception of the whale's intestine (rope n.²), shows an isolated early example of sense 4 (rood n. 2 or rood n. 6 seem unsatisfactory alternative suggestions), and if so whether it shows a reflex of Old English rād or rodu:c1400 (?c1380) Patience l. 270 He glydes in by þe giles..Relande in by a rop, a rode þat hym þoȝt.