Recently I've been helping my home students learn the past participles of some irregular verbs, in a "new" way. Basically, I show that sometimes the suffix -(e)n is added to the PRESENT stem. For example:

Base (infinitive) Past Base + (e)n suffix
arise arose ARISEN [arise+n]
be was/were BEEN [be+en]
blow blew BLOWN
draw drew DRAWN
drive drove DRIVEN
eat ate EATEN
fall fell FALLEN
give gave GIVEN
grow grew GROWN
know knew KNOWN
mistake mistook MISTAKEN
rise rose RISEN
see saw SEEN
shake shook SHAKEN
take took TAKEN
throw threw THROWN

In group 2, the -n suffix is added to the PAST stem, when it has the single "o" (Yes, I realise it's not an infallible rule, if it were, the past participles of arise, rise would be similar to CHOSEN, i.e. arosen and rosen. And the past participle of drive, should be droven.)

Base Past Past + (e)n
break broke BROKE [broke+n]
choose chose CHOSEN
freeze froze FROZEN
speak spoke SPOKEN
steal stole STOLEN
wake woke WOKEN

In the third group, the letters t and the d are doubled before the -(e)n suffix. This "rule" seems to work well.

Base Past Base/Past + (e)n
bite bit BITTEN [bit+t+en]
forget forgot FORGOTTEN [forgot+t+en]
hide hid HIDDEN [hid+d+en]
ride rode RIDDEN
tread trod TRODDEN
write wrote WRITTEN

In group 4, the suffix changes to -ne if the verb in the PRESENT stem ends in "o"

Base Past Base + ne
do did DONE [do+ne]
go went GONE

I read that the past participle inflection, the -n suffix, goes back to Germanic. Is it, therefore, safe to say that all the verbs listed above are of Germanic origin?

The frequency of the -en suffix seems to suggest that there was no such thing as irregular verbs in Old English and today the "regular" verbs that end in -(e)d is a relatively recent development of the language.


  1. Did more past participles use to end with -n?
    For example, (a) come––>came––>comen, (b) drink––>drank––>drinken
    (c) hold––>held––>holden. Why did it change?
  2. In group 4, were the past participles of do and go ever doen and goen respectively?
  3. Did been and seen use to have two syllables? Were they ever pronounced "be•en" and "se•en"? Etymonline has nothing about their historical pronunciations.
  • 1
    "-(e)n" is the historical "strong verb" past participle suffix. One online resource that discusses the verbs that took this suffix, and their patterns of vowel alternations, is the Wikipedia page Germanic strong verb; I think there would also be information about this class of verbs in a grammar of Old English. – herisson Feb 5 '18 at 23:37
  • 1
    It goes back even further than Germanic—it’s the descendant of PIE *-no-, a suffix used to form adjectives from verbs. How precisely it differed from *-to- in PIE is not entirely clear, but in Germanic, *-no- became associated with strong verbs (those who inherited the PIE ablaut in their inflection), while *-to- became associated with weak verbs (those who did not ablaut, inflecting instead with an added dental suffix). – Janus Bahs Jacquet Feb 6 '18 at 0:05
  • Incidentally, if you want to improve your group 2, say that you add -en to the past tense form if the past tense has the vowel /oʊ/ (or /ɔ/ if there’s an /r/ right after, such as ‘tear/tore/torn’), unless the base/present form has the vowel /aɪ/. – Janus Bahs Jacquet Feb 6 '18 at 0:15
  • Very interesting question, and very promising. (six upvotes in one hour.) – Centaurus Feb 6 '18 at 0:44
  • 1
    @ChrisH yes, I'm fully aware that group 3 is imperfect, but showing that the letters "d" and "t" are doubled in the PP is a visual reminder. Most Italian kids rattle off the irregular verbs in alphabetical order as if they were lines of poetry, I want to bring more "awareness", I hope it makes it more meaningful for some of the home students–they seem to appreciate it! – Mari-Lou A Feb 6 '18 at 9:12

Once upon a time, there were six regular classes of "strong" Germanic verb that formed their four principal parts by a — mostly predictable — vowel change in the stem. This at least partially survives today:

drink, drank, drunk; forbid, forbade, forbidden

In contrast, "weak" verbs had to be propped up by a final dental stop:

snow, snowed, snowed; build, built, built

This vowel change in strong verbs is called either vowel gradation or by its originally German name: ablaut. While weak verbs were by no means new to proto-Germanic or even Proto-Indo-European, they are still the only way to make new verbs, since a strong verb can only be added to the language by prefixing.

A seventh, sort of catch-all class was added to include verbs which linguists determined used not only a vowel change to indicate tense, but whose stems were originally repeated. Hold was originally in this class: cf. Gothic hald, haihald/hehald (I hold, I/he held).

There was also a hybrid class where a vowel change occurred in the present stem only, the preterite and past participle reverting to the original stem with weak endings. A goodly number of these are still kicking around:

think, thought, thought; bring, brought, brought

While these 6+1 classes were still extant in the oldest Germanic languages — Old English, Old Norse, Old High German, and Gothic — change was already in the air. To preserve the 6+1 scheme in OE, for instance, class III had to be divided into five sub-classes to account for vowel changes caused by various consonantal environments. In all these languages, however, the past participle of strong verbs ended in -[v]n, in OE and OHG, an -an, while those two West Germanic languages also added the perfective ge- prefix. This became a y- in ME and eventually disappeared, but is still present in modern Dutch and German.

These old -en participles survive today mostly as adjectives:

a shrunken head, a freshly mown field, a drunken sailor, sunken treasure, a sodden drunk, molten lead, beholden to no one, grief-stricken, riven by ideological differences

Why did the -en suffix otherwise disappear? Likely for the same reason as the ge-/y- prefix: the double marker to signal "past participle" was no longer felt necessary.

Did more past participles end in -en? The adjectives show that to be true, but there are also quite common verbs that have changed from strong to weak or are still in the process of doing so, like strive. Help was originally a strong verb, even in ME:

helpen, holp, hulpen, (y)holpen

Now the third in this list of principle parts is the primary reason the original 6+1 order became modern chaos: old Germanic languages had two preterite stems, not just one. When the second preterite, used for the second person singular familiar (thou) and plural forms, began to disappear, the vowel in a now merged past tense could follow the first preterite, the second, or the participle.

OE: findan, fand, funden, gefunden

ME: finden, fond, funden, (y)funden

This is the same pattern as sing, sang, sung, but the merged preterite of find followed the second rather than the first preterite, yielding

ModEng: find, found, found; bind, bound, bound

ModGer: finden, fand, gefunden; binden, band, gebunden

This unpredictable paradigm levelling has forced English teachers simply to call strong verbs irregular and be done with it.

To the specific verbs you've asked about: OE gebeon > ME (y)been, one can assume intermediate stages on the way to a monophthong. The same could be said for OE (ge)sewan, (ge)segen > (y)seen, but by the time the two are written with ee, they're probably monosyllables. Done and gone (rhymed with "groan") were by this point most likely also monosyllables, regardless of orthography, which in this time period can be quite arbitrary.

  • 2
    +1 Nice to see some English linguistics answers here on EL&U. I don't suppose you know of a list of adjectives such as shrunken, beholden etc? (do you want sewn in that list, btw, as for many speakers it's still the PP?) – Araucaria - Not here any more. Feb 6 '18 at 16:26
  • 1
    And for many speakers the PP is mown, too. – Rosie F Feb 6 '18 at 16:47
  • I added a few other participles, but I've never seen a list. Of course for BrE I could add ill-gotten gains. I associate mown and sewn with speakers born 1900–1920. – KarlG Feb 6 '18 at 18:19

Did more past participles use to end with -n?

Yes. In Old English, strong verbs took the "-en" suffix in order to form the past participle:

The past participle was formed using a dental suffix for class 1 and 3 weak verbs ("-ed", "-t", or "-d", depending on the verb), and "-od" for class 2 weak verbs. Strong verbs took the suffix "-en" and the appropriate stem vowel change for their strong verb class, for example "singan" - "sungen".
Wikibooks: Old English/Participles

In Middle English, "-en" and "-e" were both used to make the past participle of strong verbs (in other words, "-en" was becoming less common):

The past participle of both strong and weak verbs frequently has a y prefix; the weak past participle ends in -d or -t, the strong in -e or -en. Example: (y)gonne and (y)herd.
Chaucer's Grammar

(Note that the "y" prefix comes from the Old English prefix "ge-".)

Middle English was a period of drastic change, where many things in the language were simplified. This is but one of these changes. It's explained somewhat in this paper:

Middle English was a period of extensive syntactic change, and only a selection of the changes that have attracted a good deal of scholarly attention can be mentioned here. The role of language contact in shaping Middle English syntax is currently an issue of considerable debate. An influential paper here is McWhorter (2002), who argues for a large number of syntactic changes that he attributes to contact with Norse.

You mention comen, drinken, and holden. These were all past participles at one point, among other forms, according to the OED (emphasis theirs):

OE cymen, OE gecumen, OE (Northumbrian)–eME cumn- (inflected form, in prefixed forms), OE–ME cumen, OE (Northumbrian) ME–15 cummen, lOE ecymen, lOE gecuman, eME cumenn ( Ormulum), eME ȝecumen, eME hicomen, eME icumen, eME icumn, eME ikimen, eME ikumen, ME comene, ME comin, ME cominne, ME commene, ME commin, ME common, ME commun, ME commvn, ME comn, ME comun, ME comvn, ME comyn, ME coomen, ME coomyn, ME cumin, ME cummin, ME cummun, ME cumne, ME cumyn, ME cumyne, ME icomen, ME icomin, ME icommen, ME icommin, ME icomyn, ME komen, ME kumen, ME ycomen, ME ycomin, ME ycommen, ME–15 commen, ME–15 commyn, ME–16 comen, 15 coommne, 15 cummyn, 15–16 com'n, 15–16 comne; Eng. regional 18 cummen (Northumberland), 18 cummun (Cheshire), 18–19 comen (Yorks. and Shropshire); Welsh English 19– cummun (Flintshire); Sc. pre-17 commen, pre-17 commyn, pre-17 comyn, pre-17 cumene, pre-17 cumine, pre-17 cuming, pre-17 cummen, pre-17 cummin, pre-17 cumming, pre-17 cummyn, pre-17 cummyne, pre-17 cummyng, pre-17 cumyn, pre-17 cumyne, pre-17 cwmin, pre-17 cwming, pre-17 18 cumen, pre-17 18 cumin, pre-17 19 comen, 19– comin.

OE druncen, ME– drunken, (ME Orm. drunnkenn, ME–15 dronken, ME dronckyn; Sc. ME drukken, 15 drokin, 16–18 druken, drucken)

α. OE--ME ( ge)halden, ME ihalden, ME halden-yn ( alden, etc.), ME–15 haldin, (15 Sc. haldinehaulden, 18 Sc. haddennorth. hodden).

β. ME y-i-holden-yn, ME– holden, (ME -in-yn-unolden).

γ. OE ( ge)healden, ME ihealden, ME–15 helden.

(Note that none of these lists is complete; OED lists more forms for each.)

In group 4, were the past participles of do and go ever doen and goen respectively?

I wouldn't say that they were THE past participles. Again, spelling was not very standardized, so A past participle would be most correct. Note that both verbs are irregular (and have always been irregular in English).

Here's part of the list for past participles for these words from the OED (emphasis theirs):

OE gedoan (Mercian), OE gedoen (Anglian), OE gedonn- (inflected form), OE gedoon (rare), OE gidoen (Northumbrian), OE godon (probably transmission error), OE (chiefly Anglian) ME–15 doen, OE (rare) ME–16 don, OE–eME gedon, eME dom (transmission error), eME ȝedon, eME geydon, eME gie-don, ME dooun, ME doun, ME doune, ME doyen, ME doyn (chiefly north.), ME dun (north.), ME dune (north.), ME hi-don, ME idon, ME i-don, ME i don, ME idone, ME i-done, ME idoon, ME i-doon, ME i doon, ME idoone, ME ydon, ME y-don, ME y don, ME ydone, ME y-done, ME y done, ME y-donne, ME ydoon, ME y-doon, ME y doon, ME y-doyne, ME–15 down, ME–15 downe, ME (chiefly north.)–15 (Westmorland) doyne, ME–16 donn, ME–16 donne, ME–16 doone, ME–16 (17– regional and nonstandard) doon, ME– done, lME edoone, 15 dooen, 15 i-doen (arch.), 15–16 don't (with personal pronoun affixed); Eng. regional 16 deaun (Yorks.), 18 den (Lancs.), 18 din (Devon), 18– deean (north.), 18– deen (Northumberland), 18– deeun(Northumberland), 18– deughn (Northumberland), 18– deun (north.), 18– diun (Northumberland), 18– doin (Yorks.), 18– duin (north.), 18– dun (north. and north midl.), 18– dyeun (north.), 18– dyun (Northumberland), 19– a-done (south.); Sc. pre-17 deun, pre-17 doen, pre-17 doin, pre-17 donne, pre-17 doon, pre-17 doone, pre-17 doun, pre-17 doune, pre-17 dovnne, pre-17 dowin, pre-17 down, pre-17 downe, pre-17 downne, pre-17 doyn, pre-17 doyne, pre-17 dwin, pre-17 dwne, pre-17 dwyne, pre-17 17– done, pre-17 18– dune, pre-17 19– don, 17 dene, 17– deen, 18 deune, 18 dön (Shetland), 18– daen, 18– dane, 18– duin, 18– düne (chiefly Shetland), 19– dain, 19– din; also Irish English (north.) 19– daen, 19– din.

OE gegan, OE gen (in prefixed forms (not ge-), rare), OE (rare)—eME gæn (in prefixed forms (not ge-)), OE—ME gan, eME ȝegan, eME gun, ME gain(north.), ME gane, ME geen (in representations of northern speech), ME gone, ME gonne, ME goone, ME goyn (north.), ME igan, ME igon, ME igone, ME igoon, ME jgon, ME ygan, ME ygoen, ME ygon, ME ygone, ME ygoon, ME—15 goen, ME—16 goon, ME—17 gon, lME gonn (in a late copy), 15 i goen (arch.), 16 gaene, 16 ygone (arch.), 19— gawn (regional and nonstandard), 19— gorn (regional and nonstandard); Eng. regional 16 18— gane (north.), 16 18— gean(north.), 17 guone (Somerset), 18 gaan (north.), 18 gain (Yorks.), 18 gayn (Westmorland), 18 geyan (Lancs.), 18 gwon (Northants. and Somerset), 18 gyan(Cumberland), 18— gaen (north.), 18— gaine (Yorks.), 18— geean (Yorks.), 18— geyen (Northumberland), 18— geyn (Northumberland), 18— gi'en (Durham), 18— goan, 18— gooan (Somerset), 18— goon, 18— gwone (west.), 18— gwun (west midl.), 18— gyen (Northumberland), 19— a-gone (Berks.), 19— gan (north.), 19— gæan (Lancs.), 19— geayne (Yorks.), 19— gian (Yorks.), 19— gon; U.S. regional 18 gown; Sc. pre-17 gaan, pre-17 gain, pre-17 gaine, pre-17 gayn, pre-17 gein (east.), pre-17 geine (east.), pre-17 goyn, pre-17 ygone, pre-17 17— gaen, pre-17 17— gane, pre-17 17— gon, pre-17 17— gone, pre-17 18 gayne, pre-17 18— gan, 18 geane (south.), 18— geen (north-east. and Orkney), 18— gin, 19— gaun, 19— gean, 19— gien; also Irish English (north.) 19— gane.

(Again, there's actually more forms than just these, but you get the point.)

Did been and seen use to have two syllables? Was it ever pronounced "be•en" and "se•en"?

I don't know the whole history. In the 1848 book A Critical Pronouncing Dictionary, and Expositor of the English Language, been and seen are both listed as having one syllable.

  • 1
    Nice! It looks to me like the past participle forms for "seen" and "been" aren't exactly in the same etymological boat. The spellings listed in OED for "been" show no indication of more than one syllable, but for "seen" we have "OE (ge)sewen, ME sæwon, sawen, sægon, se(o)gon, sewen, seȝen." Unfortunately I don't know enough about OE or ME to know what those spellings truly imply about syllables. – RaceYouAnytime Feb 6 '18 at 1:51
  • @RaceYouAnytime Certainly seen has a lot of OE and ME spellings that look like they may reflect two-syllable pronunciations: OE sehen ME seien, sewyn, seyen, seiȝen perhaps seine, seane. – tchrist Feb 6 '18 at 3:29
  • @tchrist That's essentially what I was trying to (hestitantly) suggest: That it looks like "seen" has a history of two-syllable pronunciations but "been" appears not to (at least in its earliest forms). – RaceYouAnytime Feb 6 '18 at 3:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.