Survey of English Place-Names

Historical Forms

• [æt] Tureces iege 873 ASC c900

• [æt] Turkes ege 874 ASC l10

• [in] Turcesige e11 Æthelweard

• [æt] Turces ige 873 ASC c1050–1100 ASC 873 1121

• [æt] Turesigge 874 ASC c1050

• Turc c979–85 Coins

• Tor 985–8,c1017–23 Coins

• Tore c1009–16 Coins

• Dorchesyg 1086 DB

• Torchesyg, Torchesig, Torchesiy, Torchesey 1086 DB

• Torchesi 1110–14 RAi

• Torcheseia 1135–47,c1151(1325) RAi 1155–8 1329 P 1166–7 RPD Hy2

• Torcheseige m12 HenHunt

• Torcheseiam 1153 RRAN

• Torchesia 1154–76 BC

• Thorkesye 1146 Dugdi

• Thorkeseye 1230 P 1237 Cl c1240–5 RAx 1255 Pat 1258,1283 Ipm 1288 Ass 1296 RSu

• Thorkeseia 1230 P

• Thorkese 1237,1265 Pat

• Thorkesie 1237 Cl

• Thorkeseya 1263,1274 RRGr

• Thorkesei 1277 RAii

• Thorkesay 1275 Ipm 1292 RSu

• Thorkesey 1428 FA

• Torkeseie 1147 Bard p1269

• Torkeseia 1173–4,1176–7 P 1214 P 1313 Cl 1318 Pat

• Torkesey Hy2(1407),c1200(1407) Gilb 1201 Ch 1237 RAii 1216–18 1233 FF 1234 RAii 1237 Dugdiv 1237 RAii 1331 Ass 1241 RH 1271–2 RAii 1274–5 Pap 1305 Pat 1310,1323,1347 NI 1322,1327,1339 Ipm Ed3 Gough 1355 Peace c1360 Pat 1374 Pat 1377,1381 1559

• Torkesay 1182 Gilb 1417 Misc 1325 SR 1327 Par 1335 Ipm 1356 Peace 1373 RRep 1412 Fine 1421

• Torkeseia 1197 P c1200–20 RAiv 1201,1202,1203 P 1202 Ass 1214 P 1204,1230 ChancR

• Torkesie 1197 P 1202 1206,1210,1211,1212 P 1218 Ass

• Torkeseye John RPD 1209–35 LAHW 1227 Pat 1230 Cur 1242,1260 Cl 1274–5 RH 1377 Fine 1403,1405 Cl 1425,1431 Pat

• Torkeseya 1200 1216–18 RAii 1227 Pat

• Torkesei 1210–18 RAii 1283 Dugdvi 1331 Ch

• Torkesheya 1247 RRG 1249 Cl

• Torkesheye 1275–6 RH ?1280–90

• Torkeshey 1291 Tar 1463 Fine

• Torkiseia 1173–4 ChancR

• Torkseya 1189 Ch 1382 Kirkst 1193 13 Pat 1201 Pap 1227 1333

• Torksey 1227 Ch 1499 Cl 1517 ECB 1526 Sub 1531 Wills 1551 Pat 1576 LER 1577 Pat 1576 Saxton

• Torksaye 1559 Pat

• Torquesei 1195 Ch 1335

• Torkeisie 12 HC

• Torkes' 1226 ClR 1230 Memo 1232,1246 Cl 1254 ValNor

• Torkese 1227 Ch

• Torksey 1227 Ch 1499 Cl 1517 ECB 1526 Sub 1531 Wills 1551 Pat 1576 LER 1577 Pat 1576 Saxton

• Torksaye 1559 Pat

• Thorkes' 1237 Cl 1295 RSu

• Torkes 1239 Ch

• torkesh' 1245–6 RRG

• Thork' 1246 RRG

• Thorkesseye 1263 FF

• Thorkessey 1536 LP

• Torgeseye 1274 Pat

• Torqueseye 1305 Ipm

• Torkisey 1309 ChancW

• Torkeszay 1427 Pat

• Torkessay 1475,1475–6 AD

• Torcksey 1533 VisitN

• Torkysay 1535 VE

• New Torkesey 1536–9 Leland

• Torxsey 1553 Pat

Etymology

This is a difficult name. The second el. is ēg 'an island; dry ground in fen, raised land in a wet area', but the first, though apparently a personal name, is hard to interpret. In the first edition of DEPN, which was published in 1936, Eilert Ekwall interpreted the name as 'Turec's island' and remarked that the pers.n. *Turec (or rather *Turoc ) is not attested independently and may be derived from the root contained in Gothic gatarhjan 'to distinguish'. This etymology was carried over unaltered in all subsequent editions of Ekwall's dictionary. Ekwall (DEPN) also suggested *Turec /*Turoc as the first el. of Torkington Ch and was followed in this interpretation by PN Ch 1 299. DLPN 128 proposed an unrecorded OE  pers.n. *Turc , an OE loan from British belonging to Brit  *torco - 'a boar' and comparable with the Breton form Turch , as the first el. of Torksey. CDEPN 622 is somewhat ambiguous, commenting on Torksey: “OE  pers.n. *Turoc , genitive sing. *Turoces , + ēġ . The earliest spelling suggests *Turoc ; the rest would be compatible with pers.n. *Turc < Brit  *torco - 'a boar', cf. Breton pers.n. Turch. ” For an (early Middle Breton) example of Turch from the mid-11th-century cartulary of the Breton abbey of Landévennec and a discussion of the Breton reflexes of British *torco -, see K. H. Jackson, A Historical Phonology of Breton (Dublin, 1976), 118, §186. However, we should note that Jackson's form occurs in the place-name spelling et Penn Guern in plebe Turch (Cartulaire de Landévennec , ed. Le Men & Ernault (Paris, 1886), no. 22), the modern Tourc'h (Département Finistére). This name is examined by B. Tanguy, Dictionnaire des noms de communes , trèves et paroisses de Finistère (Douarnenez, 1990), 218. Tanguy suggests that Tourc'h most probably derives its name from the family name Torc 'h (attested as Turch in 1384). Oliver J. Padel (pers. comm., 20 Feb 2009) wonders whether Turch would have existed as a surname in western Brittany as early as the middle of the 11th century (the date of the cartulary of Landévennec). Instead, he has suggested that we should regard the Latin in plebe Turch of the cartulary as a translation of a Breton p.n. *Ploudurch in which Turch would have been a pers.n., not a surname. He goes on to point out that Breton plou is normally followed by a pers.n., often that of the patron saint of the church, and suggests that we are concerned here with an elliptical formation in which the generic Plou - had been dropped. It should however be noted that the patron saint of the church of Tourc'h is St Cornelius (Tanguy 218). Given the difficulties surrounding Breton Turch , it would seem unlikely that an etymologically identical Brittonic pers.n. could form the first el. of Torksey.

We must return to Ekwall's interpretation as the most plausible of those so far suggested, but it is necessary to explain the etymology of the pers.n. *Turoc . As noted above, Ekwall suggested a link to Gothic gatarhjan 'to distinguish, to note', which is connected etymologically to Primitive Germanic *torga - 'view' and OE  torht 'bright, clear' (see W. P. Lehmann, A Gothic Etymological Dictionary (Leiden, 1986) 150, G65). However, the presence of the fricative /x/ in torht makes a link to *Turoc difficult. We should rather link *Turoc to an Old English reflex of the Germanic weak verb *þurēn -, the base of ON  þora 'to dare' (cf. ásgeir Blöndal Magnússon, íslensk Orðsifjabók (Reykjavík, 1989), 1186–7; V. Orel, A Handbook of Germanic Etymology (Leiden-Boston, 2003), 429). Orel takes the verb to belong to a substantive *þuraz and compares ON  þori m. 'the greater part' < *þuran - (not *þurōn -, despite Orel's indication to the contrary). The IE  base would seem to be *turo-s 'energetic, powerful, strong' (cf. Sanskrit turá -ḥ ) and we can also cite the ancient Illyrian pers.ns. Turus and Turelius as members of this word- group (see Pokorny 1083; H. Krahe, Die Sprache der Illyrier , I : Die Quellen (Wiesbaden, 1955), 70). Other members of this word-group include the Celtic tribal name Turones and the ethnonym of the Thuringians, PrGerm  *þuringōz (OE  þyringas ) 'the daring ones' (W. Haubrichs, 'Der “Name” der Thüringer', in Die Frühzeit der Thüringer : Archäologie , Sprache , Geschichte , ed. H. Castritius, D. Geuenich and M. Werner in collaboration with T. Fischer, Ergänzungsbände zum Reallexikon der Germanischen Altertumskunde 63 (Berlin & New York, 2009), 83–102, at 94–5). Cf. also OE  þyrs m., ON  þurs m. 'giant' < PrGerm  *þur -is-az and such latinized East Germanic personal names as Torisa < *þurisa and Thorisarius < *þuris -harjaz (Haubrichs, 95). We also have Thuruarus (with variants in Thar -), the name of a third-century Gothic leader recorded by Jordanes, which would appear to stand for a Gothic *þur -wars , cf. M. Schönfeld, Wörterbuch der altgermanischen Personen - und Völkernamen (Heidelberg, 1911), 200, 239.

It should be noted that examples of a pers.n. Turo are attested in the Traditionen of Freising c770 and in the confraternity books of Salzburg (c800) and Reichenau, cf. N. Wagner, 'Zu einigen althochdeutschen Kurzformen und anderen einfachen Personennamen', Beiträge zur Namenforschung N.F. 41 (2006), 159–217, at 170–1. Wagner assumes that this pers.n. has a base with a long vowel, *Tūro , a name cognate with the Eddic dwarf-name Dúri , an original byname, 'he who slumbers', belonging to ON  dúra 'to slumber'. Wagner's OHG  *Tūro is, as he points out (170), clearly the first el. of Dauernheim (Kreis Büdingen) in Hesse, in villa Wetereibe Turenheim 750–802 (12), ad curtem Tûrenheim inferiorem 951 (orig.), but the situation seems less clear elsewhere. E. Schwarz, 'Baiern und Walchen', Zeitschrift für bayerische Landesgeschichte 33 (1970), 857–938, at 867–8, took it to be unclear whether the vowel of the Freising and Salzburg examples of Turo was long or short or whether initial /t/ goes back to /d/ or to 'old' (i.e. pre- Germanic) /t/; as he rightly indicates, an original Germanic name *Turo /*Tūro would have given rise to *Zuro /*Zūro in Oberdeutsch on account of the Second Sound Shift. At another point in the same article (865 and n. 34), Schwarz mentions the existence of Turus in Illyrian and Venetic. The MGH edition of the Confraternity Book of Reichenau lists Turo under the lemma thur - (Das Verbrüderungsbuch der Abtei Reichenau (Einleitung , Register , Faksimile ), ed. J. Autenrieth, D. Geuenich and K. Schmid, MGH Libri Memoriales et Necrologia Nova Series 1 (Hanover, 1979), 162, th 202). This implies a base belonging to Germanic *þur -, itself in turn belonging to the same IE  root as Illyrian Turus , Turelius , namely the above-mentioned *turo-s . If we wish to link the *Turoc of Torksey with this root, then we have to explain the replacement of initial Germanic [θ], which we would expect to remain, by [t]. We can plausibly do so by suggesting a process of hypocorism similar to that which gave rise in Danish to such originally hypocoristic forms as Tobbi (< Thorbiorn ) and Tōki (< Thorkil ) (cf. R. Hornby, 'Fornavne i Danmark i middelalderen', in Nordisk Kultur VII : Personnamn /Personnavne , ed. A. Janzén (Stockholm, Oslo & Copenhagen, 1947), 187–234, at 208–9). There are traces of a similar phonological process in Old English hypocoristic forms of names in þēod -. Examples are *Tēoduc , which forms the first el. of Tewkesbury Gl and of the lost Warwickshire names Teodeces leage (charter text), teodeces læge and teodeces broc (boundary clause) 963 (e11) S1307 (PN Gl 261–2), and *Teotta , which forms the first el of Teddington Gl (PN Gl 245–6). OE  *Turoc and *Tēoduc both contain the velar hypocoristic suffix *-oka -/*-uka -. The -ec - spellings in the oldest form for Torksey and in the Wa Teodeces leage reflect reduction of unstressed /o / and /u/ > /ə/ in the weakly stressed medial syllable. It should be noted that Germanic /u/ in stressed syllables normally developed to /o/ in North and West Germanic when /a/, /e/ or /o/ occur in the following syllable, cf. H. Krahe, Germanische Sprachwissenschaft , I : Einleitung und Lautlehre , 7. Aufl. bearbeitet von W. Meid (Berlin, 1969), 58–9, §36.2. However, there are numerous exceptions to this in Old English, where /u/ remains in words which occur with /o/ in Old High German, examples being OE  fugol 'bird', OE  wulf 'wolf', OE  ufan 'from above' corresponding to OHG  fogal , wolf and oban (a ), respectively (cf. Campbell 43, §115), so that *Turoc would be a perfectly acceptable Old English form.