William Allen

AUDITED 19 Sep 2026 - [] WILLIAM ALLEN, created this day from ONE record. He exists in this tree on the strength of a single document: the 1850 U.S. census of Washington borough, Washington Co., Pennsylvania, p. 350b (NARA M432 roll 834), dwelling 204 family 223, read on the image 19 Sep 2026 (register K746). It gives him as head, aged 63, born ENGLAND. THE HOUSEHOLD AS WRITTEN, with the birthplace column, which is the whole argument: William Allen 63 England - Lydia 52 do - Mary 35 do - Ellen 21 do - William 18 MARYLAND - Josephine 16 do (Maryland) - Margaret Radcliff 6 PENNSYLVANIA. An English-born couple, English-born elder children, then two Maryland-born children of 18 and 16, then a Pennsylvania-born child of 6: a family in Maryland c.1832-34 and in Pennsylvania by c.1844. WHY HE IS HERE: his daughter [111] Josephine married W. T. Crouch in this borough on 4 Nov 1852, and her 1909 obituary gives her as born Baltimore, Md., 21 Oct 1834, nee Allen. Name, age, birth state and place of marriage all agree. See [111]'s Audited note for the proof summary; the working paper is ancestry/sources/allen-111-1850-washington-borough-pa-2026-09-19/NOTES.md. WHAT IS NOT KNOWN: his own parents, his wife's maiden name, where in England he was born, when he emigrated, his occupation (the census cell is a ditto to an unread line), and his death. NOTHING here is cited to anything but the one census. NEXT, and cheap: an English-born head of household in 1850 Pennsylvania admits a NATURALISATION record - a document class never tried on this branch - and his own death or probate in Washington Co. would name him a father in his own right. CORROBORATION, 19 Sep 2026 (later), from three records read on the image that same day (K749, K750): [111]'s own lines in the 1880 Arkadelphia, 1900 Decatur and 1870 Arkadelphia censuses state that HER FATHER WAS ENGLAND-BORN - the 1870 schedule by its binary foreign-birth column, the other two by naming England outright. The only England-born adult male in the household she was living in at 16 is this man. That is independent of the 1850 census itself and it is why the father half stands. A CANDIDATE DEATH, PROPOSED AND THEN REFUTED THE SAME DAY (K751, then K756 and K757). The register of the First United Methodist Church, Washington, Washington Co., Pa. - the same congregation whose 1840-1866 member roll carries [111] Josephine Crouch and [110] Wm T Crouch, the latter with a dated entry of 21 Jan 1849 - lists, under the heading "Deaths during the two years": "William Allen - June 26, 1862". That entry was recorded here as a candidate and is now WITHDRAWN: it is not this man. THE RECORD THAT KILLED IT: the 1860 census of East Ward Washington Borough, Washington Co., Pa. (page 130, dwelling 927, family 914) enumerates Lydia Allen 63 England, Mary Allen 45 England and Maggie Radcliff 16 Pennsylvania in the household of George Murphy, a wheelwright - and NOT William. Those are the surviving females of his own 1850 household, identified by Maggie Radcliff, whose rare surname and exact ten-year age progression tie the two enumerations together where the name Allen could not. A widow lodging with her daughter and granddaughter in another man's house is not a wife enumerated apart. AND THE RESCUE WAS TESTED RATHER THAN ASSUMED AWAY (K757): a statewide sweep of the 1860 census for an England-born William Allen born 1782-1792 returns exactly one man, in Montgomery County, on the far side of Pennsylvania. => HE DIED BETWEEN AUGUST 1850 AND JUNE 1860. No death date is written as a fact here, because none has been read; the window is what the evidence supports. WHAT THIS CHANGES FOR THE NEXT DOCUMENT: a Washington Co., Pa. probate dated 1862 is the wrong decade and must not be re-proposed - K758 also swept that county's full-text probate series across five spellings of his name for 1858-1870 and found nothing, against a measured denominator that proves the 1860s are well covered there. The right document is the SAME church register volume (Ancestry coll. 2451, HSP reel 427, c.1840-1866, 559 images, never browsed): it keeps its death lists biennially, so a death in 1850x1860 should appear on an EARLIER page of it. Pennsylvania kept no civil death registration between 1855 and 1893, so for this family that register is not a second-best source, it is the only one.