One of the language myths that have been brought to light by the Rubber Bandits’ recent tweet on Cassidy’s eymology is the claim that the English colloquial ‘so long’ is really a corrupted form of Irish slán, a parting salutation.
There are several claims for the origin of this term. Some derive it from Arabic salaam, or from Hebrew shalom. Neither of these seems very convincing. The etymology websites (along with Cassidy’s book – unusually, he admits that he didn’t come up with the slán derivation) say that it first appears in 1860 in the works of Walt Whitman.
Most of them agree that it probably comes from the German expression Adieu so lange (something like ‘farewell until we meet again’) or from related Scandinavian phrases Norwegian Adjø så lenge, Farvel så lenge, Mor’n så lenge, literally “bye so long, farewell so long, morning so long;” and Swedish Hej så länge “good-bye for now,” with så länge “for now” attested since 1850 according to Swedish sources. The German expression Adieu so lange dates back to at least 1791.
In a recent OED blog post, Anatoly Liberman quotes a Mr Paul Nance who has found an earlier reference to so long from 1835.
However, I think I can do better than that. I Googled the phrase the other day and I have come across an earlier example of it. It’s in a book of humorous material called Salmagundi written in magazine form by Paulding in New York. The book is available on Google Books. It is dated 1835 but the internal headings show that the particular magazine containing the entry was first published in 1819.
The article takes the form of a humorous letter from a lady of means, who signs herself off at the end with the salutation: Adieu, so long, Aurelia.
Personally, I think that’s a smoking gun and gives a clear bridge between the German and the English expressions. But even if you choose to say that it’s just a coincidence that the earliest known use of so long in English has Adieu stuck in front of it, there are other reasons for dismissing the Irish origin. Why is this expression always written as two words? Why does it never occur as slawn, or slong? And why does it never have any other words attached? Why don’t we find it sometimes as so long go foyle (slán go fóill), or so long lath (slán leat) or so long a wallah (slán abhaile)? These are common expressions in Irish. And then again, why don’t we find it commonly in stage-Irish idiom? Begorrah, sor, it’s so long and farewell to yous …
So, let’s just forget the idea that the expression so long comes from Irish. There’s no evidence for it and there’s something so needy and desperate about these attempts to trace words to Irish. It’s as if our language and culture have no reality or value outside their relations with the English-speaking world. Anyone who thinks that should learn some real Irish – NOW!
Could you please give a fuller bibliographical reference (including the URL) for the attestation of “so long” dating to 1819 and republished in 1835? i could not find the passage you quote in Google’s version of Salmagundi online.
After seeing it, I may be able to say something.
For the time being: it has no Jewish connection of any kind.
Hi David, I’m a bit pressed for time at the minute so I’ll get some proper bibliographical details later, but for the time being here is the URL: https://books.google.co.uk/books?id=vy8hAAAAMAAJ&pg=PA82&lpg=PA82&dq=adieu,+so+long+salmagundi&source=bl&ots=ADbE9I3IU4&sig=LNZGWhPfSbPYedUypNvpw8JC2dE&hl=en&sa=X&ved=2ahUKEwiv-LXfsqPdAhWnCMAKHUgSCEsQ6AEwAXoECAcQAQ#v=onepage&q=adieu%2C%20so%20long%20salmagundi&f=false
You are absolutely right about the Jewish connection. Neither the Jewish nor the Irish claims have much value, imho. My money would be on German and I think the Salmagundi reference helps to confirm that.
Here is further evidence that the letter ending “Adieu, so long, / Aurelia” (reproduced at the top of this screen) indeed dates to 1819:
The imprint of Salmagundi: Second Series part of which is reproduced above is the reprint of 1835.
The Library of Congress holds two earlier imprints of the same title, one dated 1819 and the other, 1819-1820.
The publishers’ Advertisement (dated April 1835) at the beginning of the reprint of volume I of Salmagundi. Second Series dated 1835 says, “The Second Series of Salmagundi, it will be perceived by the dates of the papers, commenced many years after the publication of the first, and during the absence of Mr. Irving in Europe” (p. ).
The First Series appeared from 24 January 1807 to 15 January 1808.
The dates of the papers in volume 1 of the reprint of 1835 are all in 1819 (in volume II, they are presumably all in 1820).
1819 or 1819-1820 might, by a stretch, be considered “many years after” 1807-1808.
Washington Irving (1783-1859) left the United States for Europe in mid-1815 and returned on 21 May 1832.
Thus, 1819 but not 1835 falls within the period of his absence.
In the reprint of 1835, the letter appears in section headed “No. II.—June 29, 1819.”
Thus, all the other dates and other information noted above conspire to corroborate 29 June 1819 as the date of the earliest known use of “so long!”
It would be good to see the imprints of 1819 and 1819-1820 just to be sure of the date of the letter, the author of which was presumably James Kirke Paulding.
Thanks for that, David. It looks pretty good, doesn’t it? I think it’s pretty much a smoking gun.