0
Views
0
CrossRef citations to date
0
Altmetric
Research Article

A New Interpretation of the Crux aþolwarum in the Old English Maxims I 198b

Received 16 Feb 2024, Accepted 26 May 2024, Published online: 17 Jul 2024
 

ABSTRACT

The word aþolwarum occurring in Maxims I 198b is a crux, and although various readings have been suggested, none of them are satisfactory: some are philologically implausible, while others do not make good sense in the context. Since it is spelled as one word in the manuscript, and since -warum has generally been interpreted as a suffix meaning “inhabitants” that cannot be used as a simplex, it has almost unanimously been believed that it is one word. However, in order to attain a more plausible reading, we should divide it into two words, namely, aþol, an Anglian variant of adl “disease, pestilence”, and warum, an Anglian variant of (or a simple mistake for) a dative plural form of wer “man”, or a dative plural form of wara “inhabitant”. According to either reading, the passage compares the strife and affliction caused and thereby spread ever since the primeval fratricide committed by Cain with a pestilence. Strife among people and a pestilence are two of the major causes of human sufferings in the Old English poetic tradition and are quite comparable, and they are both conceived as something that spread from one place to another, which also makes the comparison quite appropriate in the context of Maxims I.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 For details of this manuscript, see Ker, Catalogue of Manuscripts, 153; Gneuss and Lapidge, Anglo-Saxon Manuscripts, 201–3. The cluster includes many works that are often labelled as wisdom poems such as The Wanderer, The Gifts of Men, Precepts, The Seafarer, Vainglory, The Fortunes of Men, and The Order of the World. Maxims I is recorded on folios 88v-92v, between The Fortunes of Men and The Order of the World. For more information about wisdom poems, see Hansen, The Solomon Complex, 3–11; Shippey, Poems of Wisdom and Learning, 1; Hill, “Wise Words: Old English Sapiential Poetry”; Niles, God’s Exiles and English Verse, 93–110.

2 The lack of structural or thematic unity in Maxims I has repeatedly been pointed out by many scholars. For example, see Williams, Gnomic Poetry in Anglo-Saxon, 54; Krapp and Dobbie, The Exeter Book, xlvi–xlvii; Howe, The Old English Catalogue Poems, 133; Fulk and Cain, A History of Old English Literature, 249; Amodio, The Anglo-Saxon Literature Handbook, 312. As concisely summarised by Cavill in his Maxims in Old English Poetry, p. 157, various attempts have been made to explain the nature of the Old English Maxims, but none of them have been convincing enough to be accepted widely. Cavill himself writes in the same book that “[t]he overwhelming impression of the poems [Maxims I and II] is of free compilation and expansion of materials from various sources” (p. 168; square brackets are mine).

3 The following are some examples of gnomes in Maxims I that contain cruces: Hond sceal heofod inwyrcan (l. 67a); deop deada wæg dyrne bið lengest (l. 78); holen sceal inæled (l. 79a); and mere hafað mundum mægðegsan wyn (l. 106). For details, see, for example, Krapp and Dobbie, The Exeter Book, 306–7; Muir, The Exeter Anthology, 554–65.

4 As regards line numbers of Maxims I, I follow Muir, The Exeter Anthology, 248–57.

5 See Cameron et al. eds., The Dictionary of Old English (henceforth DOE), s.v. “aþol-ware”.

6 Pope, “A Supposed Crux”. For details, see below.

7 In Muir’s edition, from which I quote, MS nerede is emended to serede here, but the manuscript reading makes sense, as seen in Bjork’s translation, and is retained here.

8 Quotations of Maxims I are based on Muir, The Exeter Anthology, 248–57, unless otherwise stated.

9 This translation is taken from Bjork, Old English Shorter Poems, 81.

10 According to Wright, the source text of this passage is Aldhelm’s Carmen de uirginitate ll. 2721–29, while the same idea is also reflected in Genesis A ll. 982b-95a, which he argues is also based on the same work. Wright also points out that the notion that the fratricide by Cain is the origin of human malice and strife is restricted to these three works. See Wright, “The Blood of Abel and the Branches of Sin”.

11 Earlier, Hickes includes Maxims I in his Thesaurus, but he prints only lines 71–143. See Hickes, Linguarum veterum Septentrionalium Thesaurus, 221. For Hickes’s treatment of Maxims I in his Thesaurus, see O’Camb, “George Hickes and the Invention of Old English Maxims Poems”.

12 Thorpe, Codex Exoniensis, 346.

13 See the DOE, s.v.v. “atol adj.”, “atol noun”, “atolian”, “atollice”, “atelic”, and “atollice, atelice”. The Germanic equivalent of the Old English noun atol would be reconstructed as *atulaz. See Orel, A Handbook of Germanic Etymology, 27.

14 Kirk, “A Critical Edition”, 122–23.

15 See Grein, Bibliothek der angelsächsischen Poesie, 345; Grein, Sprachschatz der angelsächsischen Dichter, s.v. “ađol-vare”. The same definition is kept in the latest edition of Sprachschatz published in 1974, s.v. “aðol-ware”.

16 Bosworth and Toller, eds., An Anglo-Saxon Dictionary, s.v. “aðol-ware”; and Clark Hall, ed., A Concise Anglo-Saxon Dictionary, s.v. “aðolware”. Bosworth and Toller also define it in Latin as cives. Earlier, Bosworth did not recognise the word but he followed Thorpe’s reading, listing the word aþol “dire” in his A Compendious Anglo-Saxon and English Dictionary, p. 34.

17 See, for example, Williams, Gnomic Poetry in Anglo-Saxon, 154; Dawson, “An Edition of Old English Gnomic Poetry”, 70 and 96; Muir, The Exeter Anthology, 564; Drout, How Tradition Works, 280; Orchard, Word-Hoard, s.v. “aþol-ware”. Gordon follows this reading in his translation of Maxims I, though adding a note saying that “[t]he text may be corrupt here”. See Gordon, Anglo-Saxon Poetry, 313.

18 See the DOE, s.v. “eþel”. The same can be said about eþel as a compound element, where both vowels are very steadily e.

19 Williams, Gnomic Poetry in Anglo-Saxon, 146.

20 The Old English text and its translation are both quoted from Mackie, The Exeter Book, 46–47. Italics are Mackie’s.

21 In his “Zum Spruchdichtung bei den Angelsachsen”, 61, Strobl suggests even more radical emendation, from aþolwarum to aðomswarian “son-in-law and father-in-law”. As Berkhout points out, however, his emendation “is quite extreme and depends for its sense on other (unnecessary) emendations in the passage”, and is untenable. See Berkhout, “A Critical Edition”, 117.

22 See Berkhout, “A Critical Edition”, 85 and 116–17; Shippey, Poems of Wisdom and Learning in Old English, 74–75. Bradley and Williamson also follow this reading in their translations of Maxims I, though Williamson, mentioning “plague”, may somewhat conflate Pope’s reading discussed below. See Bradley, Anglo-Saxon Poetry, 350; Williamson, The Complete Old English Poems, 495.

23 Pope, “A Supposed Crux”.

24 Venezky and Healey, A Microfiche Concordance to Old English.

25 For spelling variations of adl, see the DOE, s.v. “adl, adle”.

26 Pope, “A Supposed Crux”, 208–9. The three manuscripts of the Old English Bede are: Oxford, Bodleian Library, Tanner 10 (s. x in. or xi1); Cambridge, Corpus Christi College, 41 (s. xi1); and Oxford, Corpus Christi College 279 (s. xi in). The word aðol occurs in book 3, chapter 19, of the Old English Bede in these manuscripts. In the last manuscript, the original aðol is erased and changed to adl. See Miller, The Old English Version of Bede’s Ecclesiastical History, 240, and 262. Pope also points out that languores “diseases” is glossed as aðle in Psalm 102.3 in the Vespasian Psalter Gloss, while plagas “plagues” is glossed as aðlo in Mark 3.10 in the Lindisfarne Gospels Gloss. For the original texts, see Kuhn, The Vespasian Psalter, 98; Skeat, The Gospel According to Saint Mark, 23.

27 The spelling aiðulo is found in the Rushworth Gospels Gloss, at Mark 3.10. See Skeat, The Gospel According to Saint Mark, 23; Tamoto, The Macregol Gospels, 112.

28 Pope, “A Supposed Crux”, 209.

29 See the DOE, s.v. “aþol-ware”, where the passage is translated as “it was known widely afterwards that everlasting hate [or violence, hostility] was hurtful to men, as to dwellers in a plague”. Pope looked into the problem because editors of the DOE asked for his advice on the issue. See Pope, “A Supposed Crux”, 204.

30 See, for example, Muir, The Exeter Anthology, 564; Bjork, Old English Shorter Poems, 80–81. Cocco, though emending to aþolware, basically follows Pope regarding the meaning of the word, glossing it as “chi vive in tempo di pestilenza” (who lives in times of a plague). See Cocco, Maxims I e Maxims II, 86–97 and 147.

31 Although bealu is also used as a noun, it is usually the adjective bealu that is employed as the first element of a compound whose second element is a noun, as in bealubenn “grievous wound”, bealuclam “pernicious bond”, bealucræft “evil art”, bealucwealm “violent slaughter”, and bealudæd “evil deed”. Thus, it is reasonable to suppose that bealuware means “evil people” rather than “dwellers in evil”. It is true that the latter meaning has generally been preferred as, for example, is reflected in the DOE definition “dweller in iniquity” (s.v. “bealu-waru”). This is primarily because of the element -ware “inhabitants”, but as will be seen, when it is preceded by an adjective, -ware seems to carry a more general sense “people”.

32 The DOE does not recognise the word haligware but takes it as an Anglian variant of haligwer, which consists of halig “holy” and wer “man”. See the DOE, s.v. “halig-wer”. Based on the same reasoning, bealuware, which is attested only once in the form bealuwara in The Dream of the Rood l. 79a, might be a genitive plural form of bealuwer “evil man”, although it is usually taken as bealuware, as in the DOE, s.v. “bealu-waru”. Whether or not this is the case, the existence of the word Lædenware (consisting of the adjective Læden “Latin” and -ware) is beyond doubt, since it is attested in various forms in different works several times.

33 According to Wright, the source text of this passage is Aldhelm’s Carmen de uirginitate ll. 2721–29, where a pestilence is not associated with the fratricide. See Wright, “The Blood of Abel and the Branches of Sin”, 14–15.

34 The same can be said about the translation provided in the DOE, s.v. “aþol-warer”, which is quoted in footnote 29 above.

35 Dwellers in a plague can suffer from human strife, but such an interpretation does not seem to be intended by Pope, and is implausible as well. It does not make sense and is also redundant to mention specifically to “dwellers in a pestilence” after mentioning people (ældum in l. 198a) in general as sufferers of human strife.

36 Folio and line numbers in the Exeter Book are indicated in parentheses. For digitised images of the Exeter Book, see “The Exeter Book”.

37 In either case, the half-line would be a type-C verse. The medial vowel in aþol is a parasite vowel and should be disregarded metrically, but if it is counted in, the half-line would be a type-B verse.

38 The Exeter scribe often confuses a and e. For example, see MS habbenne for hebbanne (The Gifts of Men 76a); dom eadigra for domeadigre (Juliana 288a); ge witað for gewiteð (The Whale 58a); æþelinge for æþelinga (Maxims I 89a). The scribe occasionally notices the confusion and makes corrections as in grymetade altered to grymatade (Juliana 598a), lædaþ altered to lædeþ (The Order of the World 55a), monne altered to monna (Maxims I 137a), and streama altered to streame (The Whale 18b). In relation to our discussion here, it is noteworthy that the scribe repeatedly confuses a and e in the compound element -wara “inhabitant”. See MS helwerena for helwarena (Juliana 544b); and hell werena changed to hell warena (Juliana 322a, 437a).

39 The spelling waras is attested three times in the Lindisfarne Gospels Gloss (in Matthew 14.35, Mark 1.5, and Luke 22.63), and twice in the Durham Ritual Gloss. See Skeat, The Gospel According to Saint Matthew, 125; The Gospel According to Saint Mark, 9; The Gospel according to Saint Luke, 217; Lindelöf, Rituale Ecclesiae Dunelmensis, 44 and 61.

40 Hogg, A Grammar of Old English, 103 (§5.44(2)); Sievers, Angelsächsische Grammatik, 69 (§150.3).

41 Hogg, A Grammar of Old English, 103 (§5.110); Sievers, Angelsächsische Grammatik, 73 (§156.3).

42 These glosses are found in different versions of glosses for Aldhelm’s De laude virginitatis. See Napier, Old English Glosses, 103 (no. 3903) and 125 (no. 4884); Goossens, The Old English Glosses of MS. Brussels, Royal Library, 1650, 396 (no. 3794) and 453 (no. 4766).

43 Bosworth and Toller, An Anglo-Saxon Dictionary, s.v. “wara”.

44 Cf. the following gnome in Maxims I: Meotud ana wat / hwær se cwealm cymeþ, þe heonan of cyþþe gewiteþ (29b-30) “the Creator alone knows where the pestilence goes that departs from the native land”. Here the Maxims poet seems to deal with the unknowable, and therefore, unavoidable nature of diseases and deaths caused thereby. As shown in the DOE, s.v. “cwealm 1 and 4.a.”, the word cwealm generally means “death” but it can also mean “pestilence, epidemic, plague”. Cwealm in the quoted gnome has often been taken as meaning “pestilence”, since it is said that it spreads from one place to another. See, for example, Berkhout, “A Critical Edition,” 92; Shippey, Poems of Wisdom and Learning in Old English, 65; Williamson, The Complete Old English Poems, 488.

45 For adl and yldo in Old English poetry, see Cavell, Weaving Words and Binding Bodies, 218.

46 Unlike aþol/adl “disease, pestilence”, old age or yldo, another major cause of physical decline and death of a human being in the Old English poetic tradition, is not appropriate in the context of Maxims I to be compared with nið “enmity, strife”, because it is not contagious and does not spread from one person to another.

47 Kirk, “A Critical Edition”, 117. Maxims I 117b reads adl gesigan, but it does not make sense. The half-line also lacks alliteration, and Krapp and Dobbie, following Kock, emend adl to hadl, regarding it as an unattested, metathesised variant of heald in Anglian. See Krapp and Dobbie, The Exeter Book, 160 and 307. For Kock’s argument, see Kock, Jubilee Jaunts and Jottings, 39. The metathesised form hadl is unattested, but forms like hald, halde, and haldi are attested. See the DOE, s.v. “heald adj.”

48 See Drout, How Tradition Works, 219–92; and O’Camb, “Bishop Æthelwold and the Shaping of the Old English Exeter Maxims”.

49 I cannot go into details here, but to give an example, O’Camb claims that Maxims I was composed under the influence of the Benedictine revival, because four “Æthelwoldian” words, cildgeong “youthful, childlike”, acyþan “to show oneself”, styran “to steer”, and þristhycgende “brave-minded”, cooccur in lines 45-50a. Yet these words are used not only in Æthelwoldian texts but also in others including those predating the English Benedictine revival. As Neidorf contends in his “On the Dating and Authorship of Maxims I”, pp. 138–39, there is no reason to regard them as especially Æthelwoldian.

50 See Cronan, “Poetic Words”; Neidorf, “On the Dating and Authorship of Maxims I”; Wenisch, Spezifisch anglisches Wortgut, 328. See also footnote 73 below.

51 It has been debated whether Maxims I is a unified work or consists of three separate ones, because it is divided into three sections in the manuscript. The Exeter scribe marks the beginnings of works and sections by means of capitals, but it is indecisive whether those in Maxims I are intended to mark the beginnings of distinct works or those of sections of a unified work. Yet all three parts equally contain Anglianisms, and it does not affect my argument here whether or not Maxims I is a unified work. For the sectional divisions in the Exeter Book, see Muir, The Exeter Anthology, 17–20.

52 With the exception of The Menologium, the preposition on is clearly preferred to in in West Saxon texts, and unstressed in is generally rare in late southern poems of West Saxon origin. See Fulk, A History of Old English Meter, 331–32 (§362).

53 These conjunctions do not appear in late southern poems of West Saxon origin. See Fulk, A History of Old English Meter, 330–31 (§361). In Anglian texts, their synonym butan is also used, whereas according to Jordan, only butan is used in the major West-Saxon prose texts. See Jordan, Eigentümlichkeiten des anglischen Wortschatzes, 46–48.

54 In Anglian texts, both gen(a) “yet” and giet(a) “yet” are used, whereas only the latter occurs in pure West-Saxon texts. See Fulk, A History of Old English Meter, 329–30 (§360); Wenisch, Spezifisch anglisches Wortgut, 161–65; Jordan, Eigentümlichkeiten des anglischen Wortschatzes, 48–50.

55 Fulk, A History of Old English Meter, 336 (§367).

56 Unstressed in occurs in lines 7, 11, 24, 37, 41, 51, 66 (2x), 67, 83, 122 (2x), and 184, nefne in lines 105 and 184, and gen in line 11.

57 Sceððan is attested only in Anglian and Anglian-influenced texts, and it never appears in poetry of southern origin. See Wenisch, Spezifisch anglisches Wortgut, 211–15; Fulk, A History of Old English Meter, 332–33 (§363). Its West-Saxon equivalent is derian, which never occurs in pure Anglian prose texts.

58 According to Fulk, the accusative forms of personal pronouns, mec, þec, incit, uncit, usic, and eowic “are characteristic of Anglian and Anglian-influenced prose texts, but not of pure Southern prose texts”, and “[t]hey are not found in any poem of the Southern group”. See Fulk, A History of Old English Meter, 322–23 (§355(7)).

59 Fulk points out that verbs with this prefix are “effectively limited to Anglian and Anglian-influenced texts”, and “is never found in the Southern group” of poems. Fulk, A History of Old English Meter, 315–16 (§353(13)). The use of the verb prefix in- meaning “in” as in ingan “to go in”, and that indicating the reversal of an action as in inbindan “to unbind” are not restricted to Anglian and Anglian-influenced texts.

60 See Wenisch, Spezifisch anglisches Wortgut, 130, 142, 184–85, 189–205, 208–11, 283–84, 300, and 325–26. See also Jordan, Eigentümlichkeiten des anglischen Wortschatzes, 65–66; Menner, “The Vocabulary of Old English Poems on Judgment Day”, 585; Fulk, Bjork, and Niles, Klaeber’s Beowulf, cliii–cliv.

61 Apart from the one in question in Maxims I, there are as many as eleven compounds having morþor as their first element, but they never occur in poems of the Southern group: morþorbealu (Beowulf 1079a, 2742a) “wicked murder”, morþorbedd (Beowulf 2436b) “murder bed”, morþorcofa (Andreas 1004b) “wicked prison”, morþorcræft (Andreas 177a) “wicked skill”, morþorhete (Beowulf 1105a) “murder-hatred”, morþorhof (Elene 1303a) “murder-house”, morþorhus (Christ III 1624a) “murder-house”, morþorhycgende (The Wife’s Lament 20b) “plotting murder”, morþorlean (Christ III 1611b) “reward for murder”, morþorscyldig (Andreas 1599b) “wicked-guilty”, and morþorsliht (Elene 650b) “murder-slaughter”.

62 On comparison, Maxims II contains only one word, nægi (63b) “nobody”, from the list of Anglian words in Wenisch, Spezifisch anglisches Wortgut, 325–26. The Battle of Maldon also contains only one word, ricene (93b) “quickly”, whereas The Battle of Brunanburh includes none. Though it is exceptional among poems of late southern origin in that unstressed in is frequently used, The Menologium also contains otherwise only one word, symbel (200a) “feast”, from Wenisch’s word list.

63 See Megginson, “The Written Language of Old English Poetry,” 171–203; Muir, “Watching the Exeter Book Scribe Copy Old English and Latin Text”; Muir, The Exeter Anthology, 29.

64 It is beyond the scope of this article to discuss its details, but as Förster points out, the “standardised” spelling system in the Exeter Book shows a strong influence of Anglian. See Förster, “General Description of the Manuscript,” 66. For example, the past forms of cuman “to come” with w, such as cwom and cwomon, are consistently used instead of the West Saxon forms without w such as com and comon. Short o in stressed syllables before nasals (instead of a), as in noma “name” and long “long”, is another example that is consistently found throughout the codex. See Sisam, Studies in the History of Old English Literature, 101. Moreover, the verbal prefix bi- is predominant in the Exeter Book, used some 194 times, while be- being used 87 times, and this also seems to reflect an Anglian influence. The verbal prefix bi- is used substantially only in Anglian texts such as the Vespasian Psalter Gloss, the Lindisfarne Gospels Gloss, the portion of the Rushworth Gospels Gloss copied by Owun (Rushworth Two), and the Old English Bede. In Old English poetic manuscripts other than the Exeter Book, it is attested only six times, in Beowulf 2009a, 2035b, 2396b, Andreas 1591b, Elene 473b, and Leiden Riddle 3a, whereas be- occurs some 571 times. Thus, the substantial use of the verbal prefix bi- cannot be ascribed to the Old English poetic convention.

65 For this word, the Anglian spelling with the initial æ is consistently used throughout the Exeter Book. See Megginson, “The Case against a ‘General Old English Poetic Dialect”, 127.

66 In West Saxon, the retraction did not occur and *ældi- became *ealdi- by breaking and then ield- by i-mutation. For the Anglian retraction of æ to a before l plus a consonant, see Campbell, Old English Grammar, 55–56 (§143); Hogg, A Grammar of Old English, 83–84 (§5.15); Fulk, A History of Old English Meter, 283 (§335 (1)); and Fulk, Bjork, and Niles, Klaeber’s Beowulf, cxxx.

67 The corresponding West Saxon forms would be medoræden, wicfriþu, edor, and unwitod, respectively, since in West Saxon, back mutation of /i/, /e/ and /æ/ is generally restricted to cases where these front vowels are followed by a labial (/p/, /f/, /m/), labiovelar (/w/), or liquid (/l/, /r/) and a back vowel. See Campbell, Old English Grammar, 85–93 (§§205-21); Hogg, A Grammar of Old English, 153–66 (§§5.103-5.112).

68 Fulk, A History of Old English Meter, 306 (§349).

69 See Lindelöf, Rituale Ecclesiae Dunelmensis, 6 and 106, where the word appears in the spelling vnwvted. The Durham Ritual Gloss is written in Northumbrian by Aldred, who also wrote the Lindisfarne Gospels Gloss.

70 The word more generally means “enclosure, fence”. See the DOE, s.v. “eodor”. For its archaic and poetic meaning “protector”, see Cronan, “Poetic Words”, 33.

71 For the use of waldend, see Stanley, “Spellings of the Waldend Group”; Lutz, “Spellings of the Waldend Group”.

72 See Krapp and Dobbie, The Exeter Book, 307, although the editors seem to be more willing to take it as a slightly corrupt dative plural form of leod “people”.

73 See Wenisch, Spezifisch anglisches Wortgut, 328; Neidorf, “On the Dating and Authorship of Maxims I”. On the other hand, in his “Bishop Æthelwold and the Shaping of the Old English Exeter Maxims”, O’Camb argues for the late West Saxon origin of Maxims I chiefly on the basis of lines 45b-50a and the use of words therein that he regards as “Æthelwoldian”: cildgeong (48a) “childlike”, acyþan (48b) “to show oneself”, þrysthycgende (49b) “brave-minded”, and styran (50a) “to steer”. Yet none of these words could reasonably be labelled as Æthelwoldian, since they are used in texts earlier than, and having no direct relation with, Æthelwold and his circle. O’Camb’s interpretation of the passage in question is also peculiar, not quite compatible with the context of the work, and is not readily acceptable. It is also unlikely from the linguistic point of view. As briefly demonstrated in footnote 62 above, West Saxon (or southern) poems usually do not contain so many Anglian traits as Maxims I does. Moreover, the passage in question contains the uncontracted, trisyllabic geþeon in line 49a, but the metrical use of this type of uncontracted forms is, according to Fulk, “missing altogether” in the tenth century. See Fulk, A History of Old English Meter, 103 (§105). Furthermore, the passage in question occurs in the middle of a (very solid) hypermetric cluster, which never occurs in works of late West Saxon origin. See Bredehoft, “The Date of Composition of Beowulf”, 106.

Additional information

Funding

This work was supported by Japan Society for the Promotion of Science: [grant number JP22K00389].

Log in via your institution

Log in to Taylor & Francis Online

PDF download + Online access

  • 48 hours access to article PDF & online version
  • Article PDF can be downloaded
  • Article PDF can be printed
USD 53.00 Add to cart

Issue Purchase

  • 30 days online access to complete issue
  • Article PDFs can be downloaded
  • Article PDFs can be printed
USD 363.00 Add to cart

* Local tax will be added as applicable

Related Research

People also read lists articles that other readers of this article have read.

Recommended articles lists articles that we recommend and is powered by our AI driven recommendation engine.

Cited by lists all citing articles based on Crossref citations.
Articles with the Crossref icon will open in a new tab.