Nominative ‘thee’

I thought I would shift directions slightly this time and look at a change in Quaker language that has very little to do with shifts in religious thought — and which also requires a little more real linguistics to make sense of — the replacement of thou with thee as the nominative case form of the second person singular pronoun.

Probably no word is more closely associated with Quakers in the popular imagination than thee, although very few Quakers actually still use it. The fact that Quakers continued to use thee long after most of the rest of the English-speaking world abandoned it is somewhat remarkable, but what I want to focus on more particularly here is the fact that even though Quakers retained thee, they eventually dropped thou (at least in colloquial style), using thee in its place.  A related change is the loss of the traditional -est second person singular verb ending, and its replacement with what looks like the third person singular form.  The result is that in Quaker speech, one says Thee goes to meeting instead of Thou goest to meeting.

It should be recognized that nominative thee is not exclusive to Quakers, but is used in a number of regional English dialects, especially in western England.  The Oxford English Dictionary gives attestations of thee in subject position as far back as 1375, but adds the caution “the or tha unemphatic often represents both thou and thee.” The point is that in unaccented positions, the vowels in thou and thee were both reduced (probably to “schwa”) so that the pronunciation was the same — an important clue to later developments.

In fact, the only reasonably detailed proposal for explaining the Quaker shift from thou to thee that I’m aware of, by Ezra Kempton Maxfield, depends crucially on just this vowel reduction.  Maxfield first developed his analysis in an appendix to his 1920 Harvard dissertation Quakerism and English Literature, 1650–1750, later reporting it in an oral presentation to the Linguistic Society of America in 1925, and his published article ‘Quaker Thee and its History’, in American Speech 1.12.638–644 (1926).

Summarizing briefly, before the Great Vowel Shift, thou was pronounced [ðu:] (like “thoo,” with a long oo) when stressed.   Unlike our modern pronunciation [ðaʊ] with a diphthong, the older form naturally reduces to [ðə] in unstressed positions (so that thou was pronounced just like the definite article the).  This is no different from what we do nowadays with you, which reduces to [yə] (sometimes spelled “ya”) when unstressed; except in formal style or unusually careful pronunciation, a sentence like Can you go? is pronounced “Can ya go?” unless one puts special stress on you. At the time Quakers began shifting from thou to thee, Maxfield suggests, the shift from [u:] to [aʊ] may not yet have been complete, so that thou was pronounced with some intermediate diphthong which still reduced to [ə] in unstressed positions.  (A caution: Scholars differ in their estimation of when the [u:] to [aʊ] shift was completed, and most estimates put it somewhat before the use of nominative thee became popular among Quakers in the late 18th and early 19th centuries.  But Maxfield’s assumption that the shift was still underway is not really essential to his analysis; one can easily imagine that the pronunciation of thou as [ðə] in unstressed positions continued well after thou became fully diphthongized under stress.)

Of course [u:] is not the only vowel which reduces to [ə], and pronouns are very frequently unstressed; the result would have been considerable obscurity as to the underlying vowel in thou.  (I would add: especially to children learning to talk.)  Because thee and the were both also pronounced [ðə] when unstressed, and became [ði:] (with a long “ee” sound) under stress, it was natural to analogize and extend this pattern to other occurrences of [ðə], specifically its use as a nominative pronoun.  The result is thee in subject position.

That is the core of Maxfield’s proposal, though he himself suggests that it is too simple, and that there were probably additional factors involved, including the possible confusion of ethical datives with postverbal subjects, and a deliberate adoption by some Friends of “countrified” speech patterns.  But as an explanation of the main original impulse of the change, it seems completely plausible, and in fact not much different from the usual account of how ye was replaced by you in standard English (somewhat earlier than the Quaker replacement of thou with thee).

Maxfield’s explanation of how verbs with thee subjects came to show the same agreement marking as third person singulars is similar: It is well-known that syllable-final t following an s is easily lost, especially if a consonant follows — just listen to people saying phrases like best friend or last call, and you will hear that the t is normally not pronounced, unless the speaker is engaging in unusually careful pronunciation or placing special emphasis on the first word.  The vowel in the old second person singular ending -est, which is schwa even in fairly careful pronunciation, naturally also drops out in fast or casual speech.  The result is that -est is reduced to -s, making it homophonous with the third person singular in all but irregular forms (such as art vs. is, etc.)  The irregular second person forms are then lost through paradigm leveling, with the result that a single set of forms serves as both second person and third person agreement marking.