Anne Curzan starts by describing a Scrabble game:

[…] I then played riven. There was no question about its wordiness, but as I played it, I suddenly wondered (aloud), “What is riven the past participle of?” (Yes, this is what it is like to play Scrabble with me.) I paused. “It is clearly a past participle,” I added. “It has to be. Rive? Does anybody rive?”

“Rift?” Peter threw out there. “But no, can’t be. That’s a noun.” (But he is right that it is etymologically related.)

Luckily, a dictionary was at hand, and there was the verb [“To rend or tear apart”] […]

It’s an old Scandinavian borrowing into English, with a regular past tense rived and a past participle riven. But as I said, Does anybody rive? Or anything for that matter? I had to know.

Her investigations, unsurprisingly, turn up exceedingly few citations for the base verb rive, which, as she says, “doesn’t look healthy.” An interesting situation I had never thought about.


  1. J.W. Brewer says:

    Is there a broader name for this sort of phenomenon (root becomes obsolete but derived word does not) that would also cover well-known instances like “uncouth” remaining in common use although “couth” is not? Participles are a bit ambiguous syntactically, and as to “riven” I suspect that most recent uses are adjectival rather than in verbal constructions like “has/had riven,” which would be consistent with the verb-as-such having fallen into desuetude.

  2. One a table saw, there is an upright metal piece that stands behind the circular blade, to keep the halves of the board from bending in on each other after the blade has cut then apart. It is called a riving knife.

  3. marie-lucie says:

    I suspect that the verb to rive is related to to rip as well as to the noun rift.

  4. Doc Rock says:

    Is one’s “rival” riven from one’s self?

  5. Is there a broader name for this sort of phenomenon (root becomes obsolete but derived word does not) that would also cover well-known instances like “uncouth” remaining in common use although “couth” is not? Participles are a bit ambiguous syntactically, and as to “riven” I suspect that most recent uses are adjectival rather than in verbal constructions like “has/had riven,” which would be consistent with the verb-as-such having fallen into desuetude.

    “Fossil word”, “orphan word” and “cranberry morpheme” all seem to hint at the concept although the first two may refer to words preserved in collocations and the third may also include cases of false analysis or loaned vocabulary.

  6. marie-lucie says:

    Crom daba: cranberry morpheme

    This refers to the apparently unknown form cran needed to interpret the word as parallel to other words for specific berries (strawberry, raspberry, etc, even crowberry for an inedible species). But it seems pretty well established that this cran must be a shortened form of crane, especially since cognates of this word appear in other languages for the same berry species. So this example is different from that of uncouth and a few others where the stem word only occurs together with affixes.

  7. The word riven has a very peculiar place in my vocabulary. In running prose, it functions fine, but I have a very difficult time parsing it when it occurs in proper names or otherwise not in full grammatical sentences. I suspect the fact that riven is the participle of an obsolete verb plays some role in my difficulty. I was well in adulthood before it occurred to me that Rivendell was a perfectly ordinary compound word, not a made up fantasy name like Imladris. I did not get the symbolism of Riven: the Sequel to Myst for a long time either.

  8. A similar example is fraught, originally the pp of the verb freight (which just barely survives) – but fraught today exists only in its metaphorical sense and no longer has any connection with loading up a cart – to the point that most people don’t say “fraught with” anymore (except perhaps in the cliche fraught with peril) and usually just say fraught to mean tense or emotionally difficult.

  9. m-l: We’ve worked out the story before, but in short: The American cranberry, Vaccinium macrocarpon, is a New World species and doesn’t have a native English name. When Europeans settled here, they saw it was closely related to V. oxycoccos, whose native English name is marsh-whortleberry. However, Low German speakers called oxycoccos by the name of kraanbere, lit. ‘crane-berry’, which was partly etymologically nativized into cranberry. This name was then adopted in Britain for both native oxycoccos and imported macrocarpon.

    So in fact cranberry is not < *crane-berry, which never existed, and cran- is truly not an allomorph of crane but a cranberry morph without meaning in English.

  10. Trond Engen says:

    Riven wouldn’t be the first adjective to have originated as an n-participle of a lost or synchronically unrelated verb. I’m too lazy to go looking for examples, but I know I have been campaigning for green and brown.

  11. I always just assumed it was from the verb “reave.” But I’m usually wrong about most things.

  12. @Trond Engen: There’s forlorn and shriven, at least. Molten and sodden are examples where the verb has become regular, but the irregular participle survives as an adjective (the verbs in question being melt and seethe).

    @GeoX: Reave and rive are an etymological doublet. The former was inherited directly through Old English, while the latter was borrowed latter from Norse.

  13. Bathrobe says:

    @ Bloix

    I would use ‘fraught with difficulty’ or ‘fraught with danger’. The use of bald ‘fraught’ sounds almost ‘postmodern’ to me. (‘Postmodern’ is one of my favourite diss-words.)

  14. marie-lucie says:

    JC: Thank you for the whole explanation. I had never encountered it before, but it makes sense.

  15. Coincidentally, I was just pointed to a Web site about Dutch recipes. (The context was a post about about a putanesca recipe from David Bowie from a food blogger I know.)

    One of the individual recipe pages, I found this horticultural and linguistic note.

    While cranberries are native to North America, many people don’t know that the plant grows wild in some parts of the Netherlands. As the story goes, in 1845 a vat of cranberries was found washed up on the beach of Terschelling, the West-Frisian island, after which the berry was grown there by the hardy islanders. The mystery fruit was officially identified as cranberries by botanist Franciscus Holkema in 1868.

    Because cranberries thrive in this ‘Wadden island’ environment, some 48 ha (119 acres) of cranberry fields can be found on the island today. Cranberries are commonly known as veenbessen in the Netherlands, but on Terschelling some still call them Pieter-Sipkesheide, after the man who found that fateful vat of cranberries.

  16. J.W. Brewer says:

    1. The cran morpheme isn’t really an example of what I was thinking of, because any X-berry is transparently a kind of berry. So the root hasn’t been lost.

    2. Copyright be damned. If the internet doesn’t have a complete bootleg pdf scan of the 1988 WMMR celebrity-recipe compilation from which that Bowie puttanesca recipe comes, what good is it? Admittedly (to cut and paste from wikipedia) many fans and critics “regard the mid-to-late 1980s as a low point of creativity and musical integrity for Bowie,” but maybe pasta sauce is an exception to that general negative evaluation of the era?

  17. It would be logical if Bowie’s creative energies at that time were lavished on pasta sauce and stinted from music.

    I say this so I can ask: can one use “stint” like that?

  18. Marsh-whortleberry, you say? I am speechless.

    OED says a whortleberry is a barberry, and is but a southwestern dialectal form of hurtleberry, which is “apparently a derivative of hurt n. (3)” ‘id.’ (SW whort), which may be related to the heraldic term hurt n. (2) ‘a roundel azure; usually held to represent a hurtleberry’. For this hurt there’s a 1578 quote: “heurtes, small Azure balls, tearmed (in Heraldry) hurts on men, and tongue-moles on women” (Cotgrave). French heurt does refer to a bruise, and is the source of the heraldic term, but is not used for any berry. So perhaps the heraldic term and the berry both ultimately mean ‘bruise blue’.

    I’m still not bringing the word whortleberry into my house, thank you very much.

  19. There is an Old English word known only in the plural hortan, genitive hortena and attested only in glosses (facinia [i.e. Latin vaccinia, bilberries, whortleberries] = hortan) that provides a more likely etymology for the English family of whortleberry. There is a brief discussion of the word here and of the diminutive formation (whortle-) here.

  20. I occasionally use the word “rive” meaning torn or pulled off (with force). It feels a bit “Yorkshire” rather than standard English.

  21. Lars (not the original one) says:

    Isn’t this simply a defective verb?

  22. Bloix, on fraught:
    no longer has any connection with loading up a cart – to the point that most people don’t say “fraught with” anymore

    Surely you don’t need a by (with or from) to accompany ‘burdened’ either. My expression appears burdened or fraught whenever I’m in the dock at the Old Bailey.

  23. David Marjanović says:

    Is one’s “rival” riven from one’s self?

    Nope, that’s your neighbor on the same river in Latin. -al- is a dead giveaway.

  24. Lars (the original one) says:

    Hmm, I usually look for a strong past when I see a participle in -n, but the dictionary says rived. Did it have a strong one earlier?

    (Danish has river, rev, reven/t, FWIW).

  25. Icelandic:

  26. Lars (the original one) says:

    reif — yes, of course, but I meant in English. Borrowing a strong participle but not the strong past is possible, just not intuitive to me.

  27. Well, MW3 has:
    rive I. \ˈrīv\ verb (rived \ˈrīvd\ ; also rove \ˈrōv\ ; riv·en \ˈrivən\ ; also rived \ˈrīvd\ ; riv·ing \ˈrīviŋ\ ; rives \ˈrīvz\) Etymology: Middle English riven, from Old Norse rīfa; akin to Old Frisian rīva to tear, rend, Latin ripa bank, shore, Greek ereipein to dash down, tear down, Old English rāw row — more at row

  28. Lars (the original one) says:

    Thanks, juha. I don’t quite get how reif is borrowed as rove, but it might just be an adaptation to existing strong verb ablaut patterns.

    According to etymoline, rive and reave are not a doublet, being cognate to G reifen and rauben (and PIE roots meaning ‘scratch’ and ‘tear’). Reave is weak so probably denominal (causative?).

  29. marie-lucie says:

    I can see the relationship between the Germanic words (Mid. English, Old Norse, Old Frisian), but what is Latin ripa doing there? And row ?

  30. January First-of-May says:

    Isn’t this simply a defective verb?

    Might be, though it probably wouldn’t count in some of those cases because the other forms are still hanging on (if barely).

  31. Lars (the original one) says:

    @m-l, it’s the noun row = F rang, so deriving it from a root meaning ‘scratch’ is not that far-fetched. And there are relatives in Germanic with the same sense.

  32. J.W. Brewer says:

    “Rive” is only one vowel-tweak away from the now-archaic verb to reave or reive (probably from a different Prot-Gmc predecessor …), which survives in the almost-but-not-quite-obsolete derived noun reaver/reiver and in the not-at-all-archaic derived participial adjectives bereaved and bereft.

  33. Rive was a class I strong verb in Old Norse and came into English as a class I strong verb, thus constituting an exception to the rule that borrowed verbs are weak. In Middle English, the OED shows the forms of the preterite classified into three groups, which would have surfaced in Modern English as rave, rove, reeve. With competing possibilities, there was an easy path to regularization, especially in points of dialect mixing like London, and eventually rived dominated all the rest. The fact that all three already meant something else probably didn’t help them. The participle riven had some competition from roven, presumably mostly among rove-users, but riven had no other meaning to help push it out. The similarity of rive to Romance words in riv- and Latin words in rip- is a matter of common origin from *rei- followed by parallel development.

  34. David Marjanović says:

    I don’t quite get how reif is borrowed as rove

    I’d guess etymological nativization; Norse ei ~ OE ā was a common correspondence.

    G reifen

    That would be the regular cognate, but its existence is news to me and to Wiktionary. There’s reiben “rub”… and I can’t see how rauben “rob” could be cognate, having a whole different diphthong.

    (Uppercase Reifen is a tire or several.)

    common origin from *rei-

    In Wiktionary, following LIV² and Mallory & Adams (2006), that’s *h₁reyp-; the *h₁ is needed to explain the Greek reflex and fits the observation that no PIE roots beginning with *r have been identified so far.

  35. Trond Engen says:

    David M.: I’d guess etymological nativization; Norse ei ~ OE ā was a common correspondence.

    Supported by correspondence between a number of verbs on the same ablaut pattern.

    Not particularlily to the point, but I’ll note for the record that there was continuous but uneven contact between England and Scandinavia from the first Germanic settlement in Britain to the Norman conquest. A word could have been borrowed to southern at one time and to northern at another.

  36. AJP Crown:
    Fraught didn’t originally have a negative connotation. Wagons were fraught with hay; inkwells were fraught with ink. It came to be used metaphorically but only in negative phrases – fraught with care, not fraught with joy. And now in a natural development (the same as the one that as you note has taken place with respect to burden) it’s become an adjective meaning tense and difficult even without a prepositional phrase attached. E.g., “A deal with Kim Jong Un could mark the beginning of the end to a fraught history of American military involvement on the Korean Peninsula.”

  37. J.W. Brewer says:

    On the one hand it seems like the current pejorative-only sense of “fraught” has probably developed among speakers with no knowledge of the etymology, so there’s not necessarily some “carrying freight” metaphor buried in there. On the other hand, it’s intriguing that the metaphorical/emotional extended sense of “having baggage” is solely negative.

  38. CuConnacht says:

    Somewhat similar to bloix’s “fraught” are “wrought”, originally a past participle of “work” as in “What hath God wrought,” now only adjectival and perhaps limited to “wrought iron”; and “clad”, originally a past participle of “clothe”. “Clad” has now become (for some) a present tense verb with forms like clads and cladding appearing fairly often.

  39. Wrought is now sometimes interpreted as the past of wreak.

  40. Could “Rip” be the word Anne Curzan is seeking? It is the modern word used in carpentry (and a few other similar places) for using a saw or other tool to cut a piece of wood in two. When done, the two planks have been riven. When a warrior cuts or rips off a rival’s arm it is riven.

  41. And the rival’s wife is bereaved.

  42. I have now consulted three etymologies and gotten three different answers as to the respective positions of reave and rive. They all agreed that the former is from West Germanic and the latter from North Germanic. However, one site derived them from the same Proto-Germanic root; another site said that reave was connected to rauben, while the ultimate origin of rive was unknown; and the third site said the reverse. Apparently, the picture is not so clear.

  43. J.W. Brewer says:

    Perhaps all the authorities will be able to agree on is that the movie starring Steve McQueen was titled The Reivers, not The Rivers.

  44. The latter would have starred Joan.

  45. I hate the fact that I have (more than once!) had to ask the question: “Which Steve McQueen?”

  46. Lars (the original one) says:

    its existence is news — yes, sorry, that’s a Da-G non-friend, so to speak, that I haven’t been able to extirpate. I don’t know if reißen is any help.

    My grandmother taught me this faux-German stanza:

    Der alte Greis saß im Schatten eines Ölbaums. Dann rief er seine Kinder zusammen und sprach seine letzten Worte

    Or rather this faux translation into Danish:

    Den gamle gris sad i sjatten af et øltræ. Så rev han sine kinder sammen og sprak sin sidste vorte

    In English that would be

    The dirty old man (‘the old pig/boar’) sat in the dregs of a beer tree. Then he tore his cheeks together and burst his last wart

    What fun we had in the old days.

  47. David Marjanović says:

    However, one site derived them from the same Proto-Germanic root; another site said that reave was connected to rauben, while the ultimate origin of rive was unknown; and the third site said the reverse. Apparently, the picture is not so clear.

    E ea – G au correspondences are regular, so I’m sure the first site is right. E rob, with b < *bb instead of v, would then come from a Kluge mess; and indeed Wiktionary cites Kroonen’s (2013) dictionary for “*ripp/bōn-”.

    Der alte Greis saß im Schatten eines Ölbaums. Dann rief er seine Kinder zusammen und sprach seine letzten Worte

    Oh, that’s perfect literary German (except that “old old man” is rather clumsy); it just seems to be packed with false friends with Danish. The trick is that rief is the ridiculously irregular past tense of rufen, “call”.

    “The old old man sat in the shade of an olive tree. Then he called his children together and spoke his last words”

  48. Nelson Goering says:

    ‘However, one site derived them from the same Proto-Germanic root’

    That site is just wrong. There really is no way to take _rive_ and _reave_ as being from the same Proto-Germanic form: they go back *rīfan- and *rauƀôn-, respectively. For its part, *rauƀôn- is a class II weak verb, which is derived from the class 2 strong verb _reuƀan-. The sound correspondances (and the morphological patterns) here are very well understood, and there’s not much possibility for confusion or interchange between the *rīfan- group on the one hand, and the *reuƀan- group on the other.

    One kind of funny thing thing is that the pattern of stranding the past participle was actually repeated twice in the history of English, once with each of these verbal roots. At some point in the prehistory of Old English, the strong verb *reuƀan- was mostly lost. If it had survived, it would have turned up as OE ˣrēofan, but this doesn’t actually occur. The only trace of it we find is the past participle _rofen_ (or, more commonly, the prefixed form _be-_ or _birofen_). Even this participle is relatively rare, and confined entirely to poetry.

    This seems pretty parallel to the situation with _riven_ and defunct _rive_, though this group wasn’t even borrowed into English until centuries after the fossilization of _rofen_.

  49. How about “bereft”? Is that related?

  50. What fun we had in the old days
    This reminds me of another Scandinavian false friend, a 30L ølfat; not thirty litres of Criscoesque oily fat but a thirty-litre beer barrel.

  51. David Marjanović says:

    How about “bereft”? Is that related?

    It belongs to bereave, so reave < *rauƀ-.

  52. Stephen Carlson says:

    ‘Shredded cheese’ in Sweden is ‘riven ost’; sounds so poetic.

  53. shredded cheese

    I thought repiä, as in revitty juusto was a loan, but, apparently, it isn’t.

  54. 30L ølfat

    As you say, not oil-fat, but rather ale-vat. Vat is one of the two native words (the other is vixen) that begins with /v/, and was therefore picked up by West Midlands dialect (the ancestor of the standard) from some southwestern variety of English.

  55. Stephen Downes says:

    About native English words beginning with /v/: how about vole and veer?

  56. “This is coming at a time when Prime Minister Theresa May’s government is deeply unpopular and rived by its own divisions.”

    Stephen Bush, “Why Won’t the Labour Party’s Anti-Semitism Scandal Go Away?” New York Times 10 August 2018, https://nyti.ms/2Orz1Ln

    Divided by its own divisions? W. C. Fields would say, “Pardon the redundancy,” but what would Mr. Casaubon say?

  57. David Marjanović says:

    About native English words beginning with /v/: how about vole and veer?

    No idea about veer, though I wonder about voire; vole, however, is from Norn

  58. reifen exists, but it’s a weak verb, derived from reif “ripe” and accordingly means “to ripen, mature”. It has nothing to do with “rive”.
    But is it possible that reiben, rieb, gerieben “to rub” is a cognate? It’s in the same class as treiben “drive” and looks like an exact parallel.

  59. Lars (the original one) says:

    FWIW, rive and drive are strong and of the same class in Danish as well. And rive isn’t just about tearing, also raking (leaves) and grating (cheese) — and don’t you have geriebener Käse?

  60. The full OED entry for rive is fascinatingly complicated. Along with entries for reave, it documents a confusing intertwined history for the several verbs. It does confirm the different Proto-Germanic origins for the West Germanic and North Germanic words, although it turns out that that is not quite the same as saying that rive and reave are unrelated.

    Under the rive entry, there are nine different streams of development for the past participle, seven strong and two weak. (The second weak one is the double-inflected “riven’d.”) Really, the development of all the principal parts shows a very high degree of variability, and there are attested spellings that indicate, when compared with contemporary attestations of reave forms, there is a very long history of people conflating the two verbs. For some speakers, the antecedents of rive and reave appear not to have represented two different lexemes. The merger was probably made easier by the fact (mentioned above by Nelson Goering) that the present form reofan must have been rare (or entirely extinct), since it is not attested in Old English; and the borrowed rive forms were available to supplete the defective native verb.

    The OED records the final result of the conflation as the existence of an alternative verb reave v.2 that is a chimera. It is not a direct etymological derivative of reave v.1, but instead is descended from the borrowed Old Norse rive lineage. The meanings of reave v.2, involving splitting, follow naturally from those of rive. However, its past and past participle forms (although only the participle is in current usage) are taken from reave v.1: reft.

    This is all complicated even further by the existence of a separate native English word rive (which has both noun and verb forms, although the noun is attested earlier in English). The meaning of this rive is “rake” (as in the garden tool), which may be similar enough to the meanings of the other lexemes to have added further confusion. [The comment by Lars (the original one), written while I was first composing my own comment, seems to indicate that the two unrelated rive verbs also coexist in Danish.] The OED comments dryly on the confusing history of similarity among all these forms, as well as wrive, which is cognate to (and shares a meaning with) German reiben mentioned by Hans. The difference in spelling is due to wrive having reached English via Middle Dutch.

  61. Trond Engen says:

    Lars: And rive isn’t just about tearing, also raking (leaves) and grating (cheese)

    And demolishing buildings, at least in Norwegian, e.g. riveentreprenørdemolition contractor“.

  62. Good lord, what a mess! Thanks for the summary.

  63. English veer < French virer < Latin gyrare < Greek.

  64. @John Cowan: The OED lists two more native words that begin with v.

    Vew is a dialectical variant of yew from the North of England.

    There is also vinny (meaning “to become mouldy or musty”; cognate to fen), which survived in dialectal usage through at least the late nineteenth century, but which is now listed as obsolete.

  65. David Marjanović says:

    and don’t you have geriebener Käse?

    Yes, but grating is considered a form of rubbing, not of ripping. Now I wonder if that’s a reinterpretation, though.

  66. Lars (the original one) says:

    Now I looked in the ODS — and there is some conflation of different roots in Danish as well. ON rífa gives the ‘scratch,’ ‘tear,’ and ‘grate’ senses, and claims MLG riven = ‘rub’ for a cognate. However, the ‘grate’ sense is “influenced” by G reiben, and earlier (but not in ON) it could also be ‘rub’ in various technical senses, like machinery running without the intended lubrication.

    The garden tool and its derived verb on the other hand seem to come from ON hrífa (verb = ‘reach for, scratch’) — but even though it may be a denominal verb from a deverbal noun, it has completely merged with the strong verb. On the other hand, Cleasby / Vigfusson claims that hrífa is the same verb, just spelled with an h- in those senses, and AFAICT Zoëga totally omits rífa in favour of hrífa — and I don’t know what PGmc root it would come from if it was regular development.

    Then the whole thing also conflates with ON rjúfa, the cognate of reave/rauben which also gives Da røbe = ‘reveal’ by regular sound change. (This verb specializing as ‘open’ > ‘open up’ > ‘reveal’ while other kinds of tearing merge with rive. Cf. cognate L rumpō, rūpi).

    And then we have the LG loan røve = ‘rob’ (weak denominal verb, but ultimately from the ‘reave’ root). A right mess.

  67. David Marjanović says:

    various technical senses

    I should mention Reibung “friction”.

  68. J.W. Brewer says:

    So I guess we’re concluding that “vauxhall” is insufficiently “native” as an English word? I like to think of it as a variant spelling of “foxhole,” trying to indicate some regional-dialect pronunciation.

  69. Did you mean to put that in the other thread? In any case, I don’t see who’s concluding any such thing; I explicitly suggested using “vauxhall” in a translation.

  70. January First-of-May says:

    I’m reminded of a statement by Mark Rosenfelder, in one of his early articles, that the native English cognate of loot (which is borrowed from Hindi) is leaf – and of my further comment (forgot where, but likely in a Wikipedia discussion) that, as far as I could figure out from Wiktionary etymologies, this is not actually correct, and the true native English cognate of loot is rip.

    More recently, I’ve read somewhere that this isn’t quite true either, and the actual cognate is… um, one or more of the verbs already explored in this thread (which is why I posted this here). I don’t recall which one specifically it was, however.

  71. J.W. Brewer’s comment is on the right thread. It’s part of the conversation about how few native English words start with v.

  72. Oh! Sorry, I didn’t even register the v- thing.

  73. Nelson Goering says:

    ‘It does confirm the different Proto-Germanic origins for the West Germanic and North Germanic words, although it turns out that that is not quite the same as saying that rive and reave are unrelated.’

    The OED’s entries are a little confused, in my opinion. They claim that their ‘reave 2’ is ‘Apparently a variant of rive v.1 arising as a result of association with reave v.1’, but looking at the forms they cite, this seems entirely backwards. The _forms_ of ‘reave 2’ are largely just those of ‘reave 1’, and are normal dialectal developments of OE _rēafian_. It’s just that the semantics of _reave_ eventually were influenced by the later, borrowed _rive_, which in turn became rarer and more obsolete. You can regard this as a merger, but it’s not a balanced one: _reave_ absorbed _rive_.

    Only insofar as some speakers might take _riven_ as belonging to _reave_ (not something the OED mentions!) is there any formal conflation.

    In any case, it’s really not that complicated of a situation. There are two verbs, totally unrelated in origin and with distinct formal reflexes and semantics. There is some complexity _within_ each group, but most of this has nothing to do with contact or conflation. For instance, _rive_ does indeed show many past tense stems, but the majority fall into four main groups. Two are based on the regular class 1 strong preterite from Old English, _rāf_: the northern developments that kept this as such (spelt variously), and the southern ones that rounded this to _rōf_ (also spelt variously). A third group is based on the Norse strong preterite _reif_. A fourth group shows a weak past tense based on the present stem. Forms of _reave_ have little or no impact on any of this.

    On its side, _reave_ largely shows weak past tense forms. The few possibly strong past tense forms could just possibly represent a conflation with _rive_: namely late ME _raff_ and _raufe_. But the former might just be a scribal error for *raft, and _raufe_ is not exactly a simple or obvious conflation with _rāf_.

    Basically, if we keep form and semantics distinct, the picture isn’t quite as messy as all that. We have two formally-distinct verbs, which even under the most generous reading only show some occasional and slight contamination here and there (the only place this might be more than minor is in the recent use of _rive_’s past participle for _reave_). On the semantic side, the two verbs sounded similar, and both had violent meanings. This allowed the reave-form group to semantically absorb the rive-meanings. It looks like this semantic raid took place earliest in the North, so the variety of forms associated with ‘reave 2’ is indeed a bit smaller and northern, but it’s still all based on forms of _reave_ — there aren’t any forms listed under ‘reave 2’ in the OED which would derived from _rive_.

    So it’s complicated, but maybe not _that_ complicated. And all the complications date only from later Middle English.

  74. John McConnaughy says:

    To answer Ms. Curzan’s question, “does anybody rive?”, some woodworkers do. It’s a term for splitting wood with a froe. For example, from Wikipedia’s entry for froe: “Another use [of the froe] is to rive pieces of wood to be used for steam bending, as this wood follows the grain and is less prone to breakage.”

  75. Interesting, thanks!

Speak Your Mind