Arterburn, Otterburn, Atterburn, ____?____

   The pronunciation of the "Otterburn" surname with its anomalous

   variations of spelling found in 19th-century English census records 

   are remarkably similar to the variations in spelling of Peter's name

   as recorded in 18th-century public records of Maryland and Virginia.

   In 19th-century England, "ar" was often pronounced with broad A 

   (as in "father")  while the  "r"  sound was dropped or diminished

   (non-rhotic)  but retained in the spelling of many common words 

   of ordinary speech (e.g., "card," "arse," etc.). This speech sound of

   broad Ä could be confused with the sound of short O, and vice versa,

   when names were spelled only from vocalization,  as illustrated 

   by anomalous renderings of "Arterburn" for "Otterburn" names in

   England (see Some Research Notes (7th ed.), pp. 29-46; pp. 61-70).

   The "Great Vowel Shift" that resulted in the emergence of broad Ä 

   (/a:/)  and of non-rhotic pronunciation from Middle English and 

   Early Modern English had begun in the 15th century,  and also 

   impacted evolving American English in 18th-century Maryland

   and Virginia.

   The earliest spelling (May, 1773) we find in Shenandoah County is 

   of "Peter Otterburn" in Henry's record of baptism, which agrees with  

   the broad A sound heard in "Arterburn."    Some transcriptions of

   "Atterburn" (e.g., when not in an alphabetical list, originally) could

   arguably have been transcribed as "Otterburn,"  since examples

   of upper-case cursive "A" and "O" (without an upper loop) can be

   found that are virtually indistinguishable.  (For an example of Peter's

   name transcribed as "Otterburn" in Prince William County, see "Prince

   William County People," then compare this to the actual facsimile

   image of the "Tithables List of 1765," in Appendix #5 of Supplemental

   Notes.)  

   The broad Ä sound heard in "Arter-" and the similar sound of short O

   in "Otter-" suggest that broad Ä (not short A as in "cat") was probably

   represented in renderings of  "Atterburn,"  in those instances where

   cursive "A" was originally rendered, although this appears to us a

   deviation in spelling since a short vowel typically precedes double

   consonants in English, today.  "Atterburn" (as transcribed) is found

   exclusively for Peter in Maryland, and occurs predominantly for both

   Peter and William in Prince William County, which suggests that an

   "r" sound was not heard in the first syllable of Peter's pronunciation,

   originally (as suggested by "Peter Otterburn," also).

       While Peter was likely born in America and learned to

       speak English from childhood, his repeated use of a

       personal mark is evidence that he was almost certainly

       illiterate.  Thus, Peter's surname as spelled in public

       records was almost certainly the result of the phonetic

       renderings of others based on Peter's pronunciation,

       not Peter's spelling.  We cannot safely assume that

       "Atterburn"  or  "Otterburn"  simply and literally

       rendered in English whatever speech sounds Peter

       may have initially vocalized as a surname.  However,

       we can reasonably infer that Peter's vocalization of 

       his name apparently remained largely consistent from

       1736 on—assuming that broad Ä and not short A was

       originally intended in the renderings of  "Atterburn,"

       which appears likely.  Peter may have early conformed

       or adapted his own pronunciation to more closely

       match that heard from other English speakers reading

       or repeating his name, as well.

   The evidence is clear that Peter's patrilineal ancestors could not have

   been English (or Scottish) "Otterburns."    Could Peter's mother have

   been an "Otterburn?"    There is no history of "Otterburns" in colonial

   Maryland or Virginia.  Both Native Americans and free African Americans

   in colonial Maryland and Virginia apparently adopted non-inherited

   surnames.    We must seriously consider that Peter may have simply

   adopted his surname.   Peter's Asian ancestry also makes this prospect

   more likely, since he would have been less likely to have had an inherited

   English surname on his father's side.  The additional prospect, explored

   below, that Peter's mother may have been Native American increases

   the probability that Peter adopted rather than inherited his surname.

   In his monumental work of research, Paul Heinegg's summary statement

   that   "East Indians apparently blended into the free African American

   population"  clearly does not describe the ARTERBURNS.  In his foreword

   to Heinegg's book, Maryland historian Ira Berlin notes that  "Heinegg's

   genealogical excavations reveal that many free people of color passed

   as whites—sometimes by choosing ever lighter spouses over succeeding

   generations."  Although Berlin primarily had free African Americans in

   mind, his observation also applies to East Indians and Native Americans. 

   Peter was always listed as  "White"  in the tax lists and censuses of

   Shenandoah County, apparently a clue that he was accepted without

   racial distinction among his peers.  Prior to Shenandoah County, extant

   records in Prince William County and in Maryland make no explicit mention

   of Peter's status, with the reasonable inference that Peter was regarded

   as a "free person." 

       Art and Jan Arterburn knew almost nothing of the ARTERBURN story

       before Shenandoah County when they published The Arterburn Cousins. 

       We owe them an immense debt of gratitude for their groundbreaking

       research, but they missed or discounted important clues and evidence

       even in Shenandoah County public records.   They also made some

       mistakes, but then made determined efforts to publicly correct those

       mistakes when discovered (e.g., "William Smith Arterburn" and Mary

       "Polly" Arterburn as William's daughter).  Regrettably, Jan and Art had

       both left us before new evidence was discovered which corrects their

       assumption that Peter and William were brothers.    We may never

       have "smoking gun" evidence to positively prove that William was the

       son of Peter.  But the weight of the evidence that we do have clearly

       does not support the assumption that Peter and William were brothers.

       The evidence does reasonably support the conclusion that Peter was

       much older than William, and that an unnamed William was apparently

       present and counted in Peter's household — before the approximate 

       time of William's marriage and birth of his first child.  These and other

       clues make it far more likely that Peter was William's father.

       The Arterburn Cousins does not present any evidence to support the

       "brothers" assumption because there was none, and none has since 

       been found.  This assumption was largely based on the fact that both

       Peter and William had young children in Shenandoah County.  But we

       now have evidence of additional and unidentified members in Peter's

       household overlooked before who may have been children and older,

       of different and earlier birth years for some of the children of record,

       and clues that Sarah was younger than Peter.  Sarah does not appear

       by name in any public records until 1773, so we cannot be absolutely

       certain which children born before 1773 belonged to her.

       We know Peter's age and birth date (1711) from public records in

       Maryland.  Peter appears several times in public records of Charles

       County, Maryland from 1736, and in Prince William County (PWC),

       Virginia from 1759.    William appears for the first time by name in

       any public record much later,  in 1767, and only in Glassford's credit

       accounts, in PWC — about the time that William married and started

       his family, and long after Peter's credit had already been established.

       Peter appears in the Tithables List of 1765 in PWC as head of house-

       hold with another male, age 16 or older;   William is not otherwise

       listed.   Peter appears in court records of Prince William County, but

       never William.   Peter was a landowner in Virginia, but William did

       not become a landowner until much later, after moving to Kentucky.

       Peter requested and was granted a permanent exemption from the

       poll tax in Shenandoah County, in 1783,  "for age or infirmity,"  as

       provided by Virginia law.     Peter continued to appear in tax lists

       as exempt (from the poll tax) but with taxable personal property,

       and we know that Zachariah McKay charged Peter for rent in 1790,

       neither of which would likely have occurred if Peter had been "infirm"

       in 1783.   Thus, Peter's permanent exemption was almost certainly

       granted because of age.   William never at any time requested or

       received such an exemption. 

       Peter drafted his Will in 1796 after an absence from the tax lists for

       a few years — perhaps a sign of frailty or failing health, and died in

       1803.  In his Will, Peter left his entire estate, which consisted only of

       personal or chattel property,  to Sarah unless she should remarry—

       implying that she was considered a prospect for remarriage.  If Sarah

       should remarry, then Peter's property was to be divided among "each 

       of my sons and daughters," implying children other than Sarah's.  If

       Sarah had been comparable in age to Peter, and Peter's progeny were

       limited to his family with Sarah,  there would have been no need for

       such a proviso.  Finally, William made the arduous move of hundreds

       of miles to Kentucky with his family, about 1798, and apparently lived

       a vital life until his death there— almost 20 years later, in 1817.

       Peter and William apparently moved their families about the same

       time from Prince William, Fauquier, or Culpeper County to old Frederick

       or Dunmore County, Virginia, and were obviously closely related (as

       confirmed by Y-DNA testing).  We have no evidence of a "Peter Sr." and

       "Peter Jr."— all of the evidence we have clearly points to only one "Peter"

       who was old enough in Shenandoah County to have been born in 1711. 

       The scenario that best explains the evidence is that William was most

       likely the son of Peter, born to a previous wife in Maryland, probably

       within the time frame, 1732-49.  There may have been other children,

       too,  whether or not they all survived,  as suggested by unidentified 

       members in Peter's household in the censuses of 1775 and 1785 (see

       also Some Research Notes and Supplemental Notes). 

   We have no solid clues about the identity of William's mother, Peter's 

   previous wife.  John Elsea's description of James Arterburn's mother as 

   a  "fair skin woman"— presumably Sarah, suggests a decided preference

   for an Anglo-European identity, as do the intermarriages of their children

   and descendants (and of those of Peter's son, William) with Germans,

   English, Scots, and Scotch-Irish.  Peter's apparently enduring ties with

   William Davis and the Davis family— presumably European, also suggest

   as much.  

   The anomalous rendering of  "Peter Arturberner"  in Shenandoah

   County appears more likely a case of an identity attributed than

   of a surname revealed, given what we now know.  "Arter," "Bern,"

   "Berner," and "Burner" were Swiss/German surnames, and both

   "Berner" and "Burner" were represented in Shenandoah County.

   However, this compound rendering of  "Arturberner"  cobbled

   apparently from the sound of Peter's name has no history in either 

   Germany or Switzerland.  Germans of dark complexion (e.g., "Black-

   Dutch")  from the south of Germany and neighboring Switzerland

   were present in the Middle Colonies, and in northern Virginia.  Peter's

   physical appearance, informed perhaps by the knowledge that he

   had German relations, and the fact that he was in court supporting

   a German, "Henry Speelman,"  when this early spelling was rendered

   (1774)  might have suggested such an identity to the clerk who

   heard Peter and recorded his name.  If this clerk happened to have

   been a German literate in English,  he might have heard either 

   English broad Ä or German long A,  both of which are practically

   identical in sound.   Similarly, the consonant  "r"  at the end of a

   syllable (as in "Ar tur bern er") is often muted or slurred by German

   speakers, similarly to English non-rhotic  "r."   In either case, this

   spelling would probably have sounded like   "Äh tuh buhrn uh"

   when pronounced in this environment,  and essentially matches

   that heard in the renderings of Peter's vocalization spelled as 

   "Otterburn"  and "Atterburn"  (i.e., with broad Ä).

   The evidence is clear that Peter's patrilineal ancestors were not

   dark Germans or Swiss.  However, Peter's support of Speelman

   and apparent connections with other Germans, together with

   intermarriages and a family tradition all suggest some kind of 

   familial or cultural relationship, very likely derived from Sarah.

   William's wife, Nancy, who has now been confirmed through

   mtDNA testing to have been of European ancestry, possibly

   could have been German, too.

   "Arturberner" probably does not tell us anything about the origin

   of our surname, but does mark the emergence of "Arterburn" as the

   predominant spelling found in public records of Shenandoah County.

   "Arterburn" occurs only twice in Prince William County (PWC), in

   John Glassford's Dumfries Stores accounts— for Peter, compared to   

   ten entries of "Atterburn" (as transcribed) for Peter and William in

   these same accounts, not including extant PWC court records for

   "Peter Atterburn."  These two anomalous "Arterburn" spellings in

   Prince William County appear to be instances of interpolating "ar"

   in rendering Peter's pronunciation, interpreted as English broad A 

   with non-rhotic or diminished "r," similar to anomalous renderings

   of  "Arterburn"  (and  "Atterburn")  for  "Otterburn"  in England.

       John Walker used "ar" to indicate the sound of long (broad) Ä

       for pronouncing "aunt"— a word not spelled with "r"— in his

       London dictionary in 1775.  Walker's example seems remarkably

       revealing of the extent to which  "ar"  had become identified

       during Peter's time (e.g., Tidewater English) with the speech

       sound of English broad Ä paired with non-rhotic "r." 

      Whether the "Arterburn" spelling became the final form of our

      surname as a result of the intersection of German and English

      cultures and languages in 18th-century Shenandoah County, 

      aided perhaps by our family's embrace of a German-American

      identity the better to fit in, may be impossible to prove but seems

      likely.    The origin of the tradition of our German ancestry and

      the origin of our surname are probably not related, otherwise.

      As American English continued to evolve and as rhotic pronun-

      ciation became the norm, and as literacy improved, the spelling

      of "Arterburn" was established beyond Shenandoah County in

      family and public records.    However, spelling variations of our

      surname can be found in public records throughout the 19th

      century.   Even today, many of us have probably heard others

      pronounce "ar" in "Arterburn" as broad Ä with and without the

      sound of "r."