Normalisation 2: additional eME forms

eME forms vs additional eME forms

This page deals with additional eME forms. How do they differ from the other eME forms? And how are they determined? Those questions will be answered below.

Principle 1 - eME forms

Principle 1 establishes that an eME form is a standard OE form with eME sound and spelling changes applied. Those changes are listed in alphabetical order in the table of sound & spelling changes (aka Table X).

Principle 2 - additional eME forms

Additional eME forms sit alongside eME forms, and can enter eME in two ways, as outlined by Principles 2a & 2b:

2. additional eME forms can enter eME, provided they conform to eME spelling, and are found in one of the following combinations:
   1. either PC2 or Orm and at least one other eME source (PC2, Orm, Lmn, AW, Owl, SO, Ch);
   2. ModE and ME with origin in an OE variant, ON, MDu or MLG, provided there isn't a similar eME form with the same meaning, supported by PC2 or Orm;

The next section discusses additional eME forms in more detail. The list in the final section, provides examples of eME forms based on standard OE forms, and examples of additional eME forms, drawn mainly from eME sources, but also from ModE through OE variants and a trio of medieval Germanic languages.

A separate page - Normalisation of grammar, deals with the application of principles 3 & 4, levelling within paradigms and the normalisation of forms within the grammar of the Early Middle English used in this site.

About additional eME forms

It is possible to write (and speak) eME using only words drawn from standard OE forms. There are however cases where two forms of an eME word co-exist - one drawn from the standard OE form, and another based on a variant OE form. There are also some eME words borrowed from other Germanic languages, chiefly ON.

In the great majority of cases, there is no additional eME form. In other words, the standard OE form, once it has undergone eME changes, coincides with the form found in PC2 or Orm, when eME spelling is applied to the latter. In the comparatively few cases where a variant form does occur in either PC2 or Orm, there is often no support for it in other eME sources. The full list of additional eME forms has 120 items, a small proportion of the total eME vocabulary which would comprise several thousand words. There are around 1600 root words used in the texts in this site.

The search for additional eME forms begins in either PC2/Orm or ModE. Those are the two starting points. PC2/Orm is an obvious starting point because the eME used in this site is based on the East Midland dialect circa 1150, and both PC2 and Orm are written in that dialect. It's important to keep in mind that neither text is free of anomalies. That's why Principle 2a requires that a variant form found in PC2 or Orm must be supported by at least one other eME source in order to qualify as an additional eME form.

PC2 and Orm are my primary focus. However, I've cast the net a little wider to include words of Germanic (rather than French) origin that appear in the Middle English period and have survived to ModE. Some of these words may be reflexes of alternate OE forms but others may not have a recorded OE antecedent. Most of the new words come from ON but a handful appear to be borrowings from MDu and MLG. And a few words have an indeterminate origin, which the MED describes as probable OE. It is possible that the MDu and MLG words presumed to be the source of additional eME words, may have had cognates in OE that were part of the spoken language but not the literary language, or were written down, but not in surviving texts.

Forms which emerge via Principle 2b (ModE words with a medieval Germanic origin), should conform to eME spelling norms and reflect where possible, the stage in the development of sounds in the East Midland dialect which had been reached around 1150. This applies particularly to words of OE origin. Without this corollary we risk introducing sound changes that don't belong to eME².

The difference between an eME form and an additional eME form may be as slight as a single vowel. As a rule, where there are two similar eME variants, I prefer the variant sourced from an additional eME form to the variant derived from a standard OE form. That's reflected throughout englesaxe. Of course, if you prefer eME forms based on standard OE forms, feel free to use those.

In theory there can be be two or more additional eME forms derived from variants of the same word via Principle 2a ("PC2/Orm plus one"). However, thus far I have only uncovered two such cases¹ of competing additional eME forms - als and alse as well as hwam and hwa_m. Similarly, for two or more additional eME forms derived from variants of the same word via Principle 2b ("ModE & medieval Germanic"), scri__ken and skri__cen come to mind (see below). It's a different story for similar additional eME words where one is derived via Principle 2a and the other via Principle 2b. Never the twain shall meet. The former rules out the latter. In such cases, priority is given to the additional eME form supported by PC2 and Orm, e.g. suster over sister (see below). In fact an additional eME form through ModE and a medieval Germanic origin, is blocked wherever either Orm or PC1 supports the eME form derived from the standard OE form, e.g. sex over six (see below).

The seven eME sources listed at Definition 1 are found in most ME anthologies. Four texts are written in the East Midland dialect - PC2, Orm, SO, Ch. Five were written prior to 1250 - PC2, Orm, Lmn, Aw, Owl. Each source shares either the time period or the dialect area of early East Midland. Only PC2 and Orm share both.

Beyond Principles 2a & 2b

Some factors in the determination of additional eME forms are difficult to distill into a pithy principle. The following common-sense and practical criteria proved useful with certain cases.

The Peterborough Chronicle - First continuation (1122-1131) (PC1) can stand in for PC2 in certain circumstances. PC1 can act as an eME source where a reflex of a particular OE word doesn't exist in either PC2 or Orm.

Occasionally I refer to the languages listed in Principle 2b - OE variant, ON, MDu and MLG, as "Medieval Germanic". It should be assumed that unattested OE (also - posited or hypothetical OE) and unknown Medieval Germanic source, belong there too. I have done so, as you'll see in a couple of items below.

Additional eME forms can arise from the combination of an active affix (prefix or suffix) with an eME form, even if that particular combination is not recorded in the OE corpus, provided that the affix existed in OE and was used for similar effect.

Additional eME forms can also arise from the duplication of an eME form, i.e. two instances of the same root in combination or side by side, even if that particular combination is not recorded in the OE corpus.

Dialectal and obsolete words are not discounted when looking for ModE reflexes of Middle English words with Germanic origins. Thus the northern dialectal word laik (play), together with Orm ley%y%kenn, support the elevation of ley%ken to the status of additional eME form (see below). The same applies to ModE (obs.) cratch (scratch) and SO crached (see below).

Examples of eME forms and additional eME forms

You've read the definitions of terms and the four basic principles which underlie the normalisation of word forms in eME. So how are these applied in practice? Here are some examples:

1. swester and suster

   Swister, swyster and swioster i.a., can all be found in the OE corpus, but they are all variants of the standard OE form - sweostor. That is the headword to which the variants point in Clark Hall ¹. So sweostor alone passes from OE to eME. After undergoing the eME sound changes /eo/ > /e/, and unaccented /o/ > /@`/, OE sweostor becomes the eME form - swester.

   However, since suster can be found in both PC2 and AW (as well as Ch), it becomes an additional eME form, courtesy of Principle 2a.

   Note that two (or more) variants of a word can co-exist in eME. Thus, swester and suster are both valid eME words. Note also that no other variant is attested in either PC2 or Orm.

   To determine an additional eME form, we can start from either PC2/Orm or from ModE. We've just seen that PC2/Orm provide the additional eME form suster. So, what if we start from the ModE form - sister and work back? Clearly this form has its origin in a medieval Germanic source, probably ON systir. On that basis, does sister also qualify as an additional eME form via Principle 2b? Close, but no cigar. A caveat in that principle stipulates that "a similar eME form with the same meaning, supported by PC2 or Orm" rules out the fresh aspirant. So, sister bows out and leaves the floor to suster.

2. selfer and silfer

   In Clark Hall ¹, we find the standard OE form seolfor. The same eME changes applied to sweostor above, give us the eME form selfer.

   The additional eME form silfer demonstrates the need to apply eME spelling not only to words derived from standard OE forms, but also to additional eME forms. The Orm form is sillferr which would have been pronounced /silv@`r/, as would PC2 syluer and SO siluer. In the eME spelling scheme, this pronunciation is represented by silfer. Since this form (with /i/) exists in both Orm and PC2 (as well as two other eME sources), silfer qualifies as an additional eME form through Principle 2a and coexists in eME alongside selfer.

3. clister and cluster

   clyster is the most common OE form according to Clark Hall ¹, however a variant cluster is also recorded in Clark Hall ¹. Neither PC2 nor Orm have any instances of this word or its variants, but cluster is the headword of the MED entry (in which none of the twenty four quotations display a spelling with <i> or <y>). Since cluster is the ModE form, with the same meaning, cluster³ qualifies as an additional eME form alongside clister, according to Principle 2b.

4. t`aeh and t`oh

   t`e_ah, meaning though, is the standard OE form (with variant t`e_h). That gives the eME form t`aeh. However PC2 and Orm have t`oh and t`ohh respectively, derived from ON t`o_/*t`auh. So t`oh qualifies as an additional eME form per Principle 2a, alongside t`aeh.

   Note that it is t`oh which gives us the Ch and ModE form though.

5. dey%en

   Orm has a verb dey%enn, meaning to die, which has no known cognate in OE. It is derived from ON deyja. Similar forms are found in Lmn, AW, SO and Ch - deiy`e(n), deien, dye and deye(n)/dyen respectively. The spellings ey`, eiy`, ei, ey all represent the same sound - /ei/ which is normalised as ey% in eME. On the basis of its presence in Orm and four other eME sources, dey%en qualifies as an additional eME form via Principle 2a, alongside swelten from OE swelten.

   Note that dey%en is the forerunner of ModE (to) die.

6. ruke

   ModE has the noun (and verb) ruck, which means a large number or mass (of people), and has more specific applications in rugby and Australian Football. It has no known cognate in OE. It is derived from ON hruga or hruka. Similar forms are found in ME, including AW - ruken, rukelin for example, but not in either PC2 or Orm. The combination of ME presence, medieval Germanic origin and ModE survival, means that some form of ruke qualifies as an additional eME form through Principle 2b. The question is - what precisely is that form? In these cases, albeit rare, in which there is neither an OE form, nor an Orm or PC2 form, the final form is decided by the following method. The original Germanic form, ME form(s) and ModE form are compared. Generally speaking, a feature which is found in only one of the three is discarded in favour of a feature found in the other two. So for example, the initial 'h' of the ON forms is dropped, as is the 'g' of one of the ON forms, since neither the AW form nor ModE have these. And ModE's lack of a final vowel is ignored since the others both have a final vowel. Given that OE final 'a' passes to 'e' in eME, the same rule is applied to the ON form. That gives a final vowel of 'e'. Thus, the additional eME form emerges as ruke.

7. y%ugelen

   ModE yowl probably has an OE ancestor. The MED lists yoy`elinge as a headword with the same meaning, and gives both a putative OE origin in *g`eogelian and a couple of citations from pre-1250 texts, including AW y`uhelunge and Owl y`oy`elinge. Unfortunately, there are no citations from PC2 or Orm. Here we have to work with the ModE and ME forms. The OE form is hypothetical. So, we can't apply the method used with ruke above. Instead, ModE dictates the final form. In such cases, we recreate the eME spelling on the basis of the ModE spelling and pronunciation, by reversing the development observed in similar words, from ME to ModE. Fortunately we do have such a pattern. Just as fowl is derived from fugel, the ancestor of yowl would be y%ugelen, which thus qualifies in this case as the additional eME form, via Principle 2b - ModE and ME with origin in an unknown medieval Germanic language (possibly OE). This is one of only a handful of examples thus far of an additional eME form arising without a recorded OE ancestor. Two others are haccen and hippen. And another is:

8. pot`eren

   The MED indicates that there was probably an OE antecedent of ModE pother (uproar, fuss, cloud of smoke), akin to MDu poderen. The MED headword is patheren, with meaning to stir or poke (ashes). The sole entry is AW pead`ered` pr sg3 but there are several variations recorded, including pad`ered` and pot`eret`. In the absence of a recorded OE form, how do we determine the vowel? Fortunately, the ModE vowel - <o>, matches not only the related MDu word, but more importantly - one of the AW variants. Hence, the additional eME form is pot`eren courtesy of Principle 2b - ModE and ME with origin in MDu (or possibly an unattested OE word).

9. mingen

   mindgian, meaning to remember or remind, is the standard OE form according to Clark Hall ¹, which gives us the eME form mingen, after undergoing eME sound changes. However a variant myneg`ian is also recorded in Clark Hall <sup>1</sup>. Reflexes of this variant appear in both AW and Lmn as munien /myni@`n/. Neither PC2 nor Orm have any instances of this word or its variants. So, does *miney%en qualify as an additional eME form? The answer is no. And the reason is that the starting point for additional eME forms must be either PC2/Orm (see Principle 2a) or ModE (Principle 2b). *mineien/*minney%enn and *minny are conspicuous by their absence.

10. tuwe and twiy%es

    tuwa, meaning twice, is the standard OE form according to Clark Hall ¹, which gives us the eME form tuwe, after undergoing eME sound changes. However a variant twig`ea is also recorded in Clark Hall ¹. Reflexes of this variant appear in both AW and Lmn as twie and in Ch as twye. However, to cloud the picture somewhat, reflexes of this variant also appear in Orm as twiy%ess, in AW as twies and in Ch as twyes. So, which qualifies as an additional eME form - *twiy%e or twiy%es or both? The answer is the second. The reason is the same as in the previous case - for *twiy%e to be considered as an additional eME form, it must be found in PC2/Orm (see Principle 2a) or its reflex must be found in ModE (Principle 2b). *twiy%e (*twi_e) and *twy are found in neither. twiy%es on the other hand, qualifies via Principle 2a - Orm twiy%ess plus AW twies and Ch twyes.

11. t`anen & t`enen, hanen & henen, and hwanen & hwenen

    The trio t`enen, henen, hwenen are closely related, and spawned the ModE series thence, hence, whence.

    t`anon (from there, thence), gives us the eME form t`anen, after undergoing eME sound changes. However a variant t`eonen is also recorded in Clark Hall <sup>1</sup>. Neither PC2 nor Orm have any instances of this word or its variants. But PC1 has t`enen and PC1 can act as proxy for PC2 where neither PC2 nor Orm have any instances of the word in question. There is support for this form with an <e> in the initial syllable, in at least one other eME source - Lmn, which has t`enene and t`ennen. So t`enen qualifies as an additional eME form via Principle 2a - PC1 with support from Lmn.

    The MED has thenne as a headword, with the same meaning. t`enne has wide support in eME sources - Lmn, AW, Owl and Ch, for example. But that particular form is not found in either PC/Orm or ModE.

    Turning to ModE, we find the reflex thence. That form derives from variants with the genitive ending -es used as an adverbial suffix, which first appeared in ME texts around 1250. The text Poema Morale (PM) has thannes. Nevertheless, t`ennes<sup>4</sup> doesn't make the cut as an additional eME form via Principle 2b, since a similar eME form with the same meaning exists, supported by PC1 - t`enen.

    Similar to t`enen is henen (from here, hence) which qualifies as an additional eME form beside hanen via PC1 heonon and Lmn heonnen. On the other hand, there are no reflexes of OE hwanon (from where, whence) or variants thereof, in PC1, PC2 or Orm, nor are there any instances of an OE variant *hwe(o)nen. Nevertheless, via Principle 4, hwenen can be admitted as an additional eME form, to match its siblings t`enen and henen. (For more on Principles 3 & 4, see Normalisation of grammar.)

12. fort`aem and fort`an

    fort`ae_m, meaning therefore, is the prime variant to which three others point in Clark Hall <sup>1</sup>, thus giving us the eME form fort`aem, after undergoing eME sound changes. However one of those other variants - fort`an appears to have eclipsed the standard OE form by the beginning of the eME period. fort`an is found in PC1, Lmn and Owl. Note that neither PC2 nor Orm has an instance of any of the variants. Nevertheless, with PC1 acting as proxy for PC2, and with the support of two other eME sources, fort`an qualifies as an additional eME form through Principle 2a.

13. writen and writelen

    writian, meaning to chirp, cheep, is recorded in Clark Hall ¹, thus giving us the eME form writen, after undergoing eME sound changes. However a form with the frequentative suffix -el - writelinge appears in Owl. Since the suffix -el/li is employed in the same way in OE, and -el is still active in ME, writelen should be considered an additional eME form.

14. scri__ken and skri__cen

    sc`ri_c, meaning shrike (a type of bird), is recorded in Clark Hall <sup>1</sup>, thus giving us the eME form scri__k, after eME spelling is applied. According to the entry on sc`ri_c in Wiktionary, its etymology is "ultimately from Proto-Germanic *skrīkijaną (to cry out), in reference to the shrike's call". The link to the ModE verbs shriek and screech is clear. Various related verb forms appear in the MED under skri_ken (shriek, scream, screech), i.a. in Owl - scrichest and bischrichet` and TH (1200) - shriked`. The forms recorded in the MED display an interesting interchange of sounds in the stem - skrike, skriche, skirken, shriken, shrichen, shirchen. ON skri__kja/skraekja is the most likely source for all variants⁵. Given the persistence of shriek and screech to ModE, the presence of matching forms in ME, and their origin in ON, both scri__ken and skri__cen⁶ qualify as additional eME forms via Principle 2b.

15. gre__diy` and graeden

    grae_dig`, meaning greedy, is the prime variant to which one other - gre_dig` points, in Clark Hall ¹, thus giving us the eME form graediy`, after eME spelling is applied.

    An additional eME form gre__diy` qualifies on two grounds, firstly - Principle 2a through Orm grediy` with the support of AW gredi, and secondly - gre__diy` belongs to a group of words which underwent a sound change peculiar to the Anglian dialect (the predecessor of East Midland) - OE WS ae_ appeared as e_ in OE A where the ancestral vowel was WGmc a_. For more detail, see sound changes before 1100 and Normalisation issues.

    By the same token, *gre__den, meaning to shout, does not qualify as an additional eME form, to accompany graeden from standard OE grae_dan. That's because there is no reflex of that verb in either Orm, PC2 or ModE.

16. galegale

    galan, meaning to sing, call, squawk, cry, is recorded in Clark Hall ¹, thus giving us the eME form galen, after an eME sound change is applied.

    galegale meaning chatterbox is found in Owl. This can also be considered an additional eME form, despite the fact that it has no recorded antecedent in OE nor a descendant in ModE. That's because it is simply a duplication of the root gale (< Proto-West-Germanic *gala_ singer) that we find in OE nihtegale.

17. rubben

    rubben meaning to rub is found in the MED. Two of the sources cited for this entry date from 1325 - robbe and robby. Because the origin of these forms is either ON or MLG, and the word has survived to ModE, rubben qualifies as an additional eME form via Principle 2b.

18. craccen

    cracchen meaning to scratch is found in the MED. The earliest entry dates from 1330 approx - SO crached. In a side note, ModE scratch is a blend of cracchen with another ME form with the same meaning - scratten, probably borrowed from OF esgrat(t)er. Given that the origin of cracchen is either ON *kratsa or MLG kratsen, and that the word survived to ModE cratch (obs.), craccen qualifies as an additional eME form through Principle 2b.

19. sex and six

    OE siex meaning six, is the prime variant to which two others point in Clark Hall ¹, giving us the eME form sex, after undergoing eME sound changes. That form closely matches Orm sexe. On the other hand, Lmn sixe and AW six support the ModE form. So does six qualify as an additional eME form? The answer is no. And the reason is that additional eME forms via ModE and ME reflexes of OE variants (Principle 2b) don't come into play where either Orm or PC1 supports the eME form derived from the standard OE form.

20. nit`er and net`er

    nid`er, meaning beneath, (be)low, down(ward), is the standard OE form according to Clark Hall <sup>1</sup>, which gives us the eME form nit`er. However a variant neod`er is also recorded in Clark Hall <sup>1</sup>. Reflexes of this variant appear in both AW and Lmn as neod`er and in Ch as nether. On the other hand, Orm has the related verb nit`t`renn. So, does net`er (with e rather than i), qualify as an additional eME form? The answer is yes. The neod`er variant may not have reflexes in PC2 or Orm (needed for Principle 2a) but on the basis of ModE nether and its origin in an alternate OE form, nether gets a guernsey, thanks to Principle 2b.

21. la__ken and ley%ken

    OE has la_can (to play; jump, leap, move up and down), and g`ela_cian (to present, bestow), which give us the eME forms la__ken and ila__ken. Orm has ley%y%kenn (to play, take pleasure in), which is derived from cognate ON leika and corresponds to the first sense, and lakenn (to present with, make an offering to), which corresponds to the second. Since none of the other six eME sources has a form similar to ley`ken, that variant doesn't immediately present itself as an additional eME form. However, ModE has the dialectal words laik and lake, which both mean 'to play' and are both pronounced /leik/. Note that a direct ModE reflex of la_can would be *loke or *loak. According to Principle 2b a variant qualifies as an additional eME form if it is found in ModE and has a medieval Germanic origin. ley%ken, with its origin in ON, meets those conditions.

More on normalisation

Looking for more background on the normalisation of the East Midland dialect c 1150, in englesaxe? For an outline of the basic principles applied to the normalisation of eME word forms, together with an introduction and a list of the abbreviations and definitions used, see normalisation 1: principles. For a discussion of levelling within eME paradigms, see normalisation 3: grammar. For a discussion of the principles governing the eME spelling scheme used in englesaxe, see normalisation 4: spelling. For a discussion of some of the finer points, see normalisation 5: issues.