After the publication of "Erowid Punctuation Policy: Data-safe Quotes" in Erowid Extracts No. 4 (May 2003), it was pointed out to us that the concept of "datasafe quotes" exists as part of the standard British punctuation convention. The British too appear to regard the practice of placing punctuation inside quotes, regardless of the sense of the original quoted material, as illogical. The arguments against the British/Erowid "logical" style are mainly aesthetic: U.S. detractors claim that periods and commas outside quotes look awkward.

The British policy is documented in Fowler's canonical Dictionary of Modern English Usage, first published in 1927 with updated versions that are still used as guides today.1 More recent rewrites of Fowler's dictionary describe both British and American styles, strongly favoring the British. Although we have been unable to find even the barest substantiation, one explanation for the American rule is that early American typesetters, using cast metal letters, found periods were too fragile when placed outside quotation marks and the use of ." instead of ". arose to avoid accidental breakage. This unreferenced story is given in Mark Israel's 1997 alt.usage.english FAQ, which also says that the British-style quote policy "has gained ground, and is especially popular among computer users, and others who wish to make clear exactly what is and what is not being quoted."2

In trying to nail down how this difference arose, we spent time driving to distant libraries and digging through books to track down such thrillers as Pause and Effect: An Introduction to the History of Punctuation in the West3 and You Have a Point There: A Guide to Punctuation and Its Allies.4 Partridge's book even has a chapter by Jonathan Clark on "American Usage" that acknowledges and describes the differences. Clark opines "Here there can be no doubt, to my mind, of the logical superiority of the British practice: it is more sensible to be guided, in this matter, entirely and consistently by logic, as British practice is and American is not."4 Unfortunately, despite a noble and tedious hunt, we found nothing describing when or where the cross-Atlantic difference came to be.

Perhaps the most interesting information about this issue comes from Parkes, who describes the development of quotation marks in some detail. Apparently, a variety of competing and inconsistent methods had been used to denote spoken passages and dialog including lines breaks, dashes, italics, parentheses, and diples, until:

"At the beginning of the eighteenth century English printers [began to use] a new punctuation symbol we may properly call 'quotation marks'. As a first stage in this new development they [inserted] inverted commas in the text immediately before the passage of direct speech or quotation to 'open' it. The second stage was to insert raised commas in the text at the end of the passage to 'close' it... The use of graphic devices to indicate dialogue became stabilized in the nineteenth century [and]... the convention that spoken discourse should be indicated by inverted commas had become... firmly established by the twentieth century."3

So, because the American convention for periods inside quotation marks was recognized by grammar books from the early twentieth century and it was in the late 1700s or early 1800s that "inverted commas" became common, there is only a hundred year or so span during which the two standards differentiated themselves. Perhaps punctuation fetishists who know of written documentation for the genesis of the "illogical" American policy can step forward and let us know how we got into this pickle.

We were pleased to learn that "data-safe quotes" are not unique to Erowid. Perhaps what is strangest is that we had no idea that this was normal use in Britain, and it has equally surprised just about every U.S. person we describe it to.