This page is part of © FOTW Flags Of The World website

Flag cantons


Last modified: 2012-01-14 by rob raeside
Keywords: cantons |
Links: FOTW homepage | search | disclaimer and copyright | write us | mirrors

See also:

My hypothesis is that by 1777, when the US flag was adopted, some two centuries of English-British practice had led English-speaking mariners to assume that an ensign necessarily consisted of a field and a canton, and that the flag flown on the jack staff was of the same design as the canton of the ensign. As far as I can tell--and subject to correction from others, such as David Prothero and Chris Southworth who know this history better than I--this was the most usual practice in England even before the adoption of the Union Jack design in 1606--a cross of St. George at the jack staff, and an ensign of either striped or solid field with a canton of St. George at the stern, then later (for warships) the Union Jack at the jack staff and an ensign with a UNION JACK canton at the stern. (It would be interesting to know if Scottish ships pre-1707 wore a St. Andrew jack along with the Scottish red ensign.

Over time, with the full introduction of the Union Jack as the device in the upper hoist of British ensigns, the term "union" became synonymous with what we would call the "canton." Thus the term for an inverted ensign, "union down."

The language of the resolution adopting the US flag on 14 June 1777 is, I think, indicative of these assumptions and this terminology regarding the design of an ensign. Note that the resolution says the flag consists of 13 stripes, alternating red and white, and that "the union" will be 13 white stars on a blue field. There is no indication of where "the union" is to be located (or of the direction of the stripes), which many modern vexillologists have taken to be a case of unclear legal drafting. I doubt that it was unclear at all to the people who had to implement the law--I believe they knew perfectly well that "the union" was the part of the flag in the upper hoist. And my contention is that they also knew perfectly well that the proper design of a jack was "the union," because for over 200 years that had been how things were done. And that's why, apparently without any further legislative or regulatory enactment, US ships began using a jack consisting of a blue field with white stars; after all, didn't the law say "the union" was thirteen white stars on a blue field? What further authority was needed? Such jacks are to be seen in contemporary ship paintings as early as the late 1780s.

As further evidence of the assumption that a ship's jack should match the "union" of the ensign, I'd cite the example of the U.S. Revenue Marine. As shown in a painting of the revenue cutter Alexander Hamilton in 1840, reprinted in Bruno Cianci's new book on U.S. Coast Guard Flags, "Due secoli sui mari," the U.S. Revenue Marine used as a jack the same design--the US coat of arms in blue on a white field--as that of the canton of the revenue ensign prescribed in 1799. This jack eventually transmogrified itself into the modern U.S. Coast Guard color.

By the way, I think the same process worked for the design of warship pennants. A British warship pennant had (and has) the emblem of the old St. George canton in the hoist, with the fly matching the color of the field of the ensign flown with it. A US Navy warship pennant had the 13 stars on blue in the hoist and a horizontally striped red and white fly. The old Revenue Marine pennant of 1799 (shown in Bruno's book on p. 17) has the blue coat-of-arms on white in the hoist and vertically striped red and white fly, just like the canton and field of the revenue ensign.

The original Continental Congress resolution defining the US flag referred to the blue area with white stars as "the union." Furthermore, as it did not say where the union was to be located on the flag, it seems clear to me that by 14 June 1777 it was commonly understood already that a "union" was the upper hoist area of a flag.

I found my original source at, citing Ray Morton. Re-reading Mr. Morton's explanation on this page, I'm not entirely sure I accept his analysis.

Now, the ensigns of sailing warships were indeed very large by modern standards. But I've seen no evidence that US ensigns were appreciably larger than those used by the UK. And we know that the largest UK ensign in 1687 was 24 x 42 feet (7.3 x 12.8 meters) for a 1st rate ship of the line. A surviving white ensign flown at the battle of the Glorious First of June (1794) measures 20 x 40 feet, and per the 1822 Royal Navy specifications, the largest ensign flown by a 1st rate ship of the line would have been 19.5 x 37 feet. (All from Timothy Wilson's Flags at Sea [wil86] and [wil99]).

The most famous Royal Navy ship of the line from this period is, of course, HMS Victory. Victory's moulded beam is 50 feet 6 inches. If Morton's analysis were correct, one would expect Victory to have flown a battle ensign with hoist and fly measurements approximately 20% larger than those prescribed in 1687 and 35% bigger than those prescribed in 1822. Not likely.

Moreover, USS Constitution was not a ship of the line. By the British rating system, it would have been classed as a 5th rate, or perhaps 4th at most. The 1822 specifications would have given it an ensign with a maximum hoist of 16.5 feet (4th rate) or 15 feet (5th rate). And, indeed, this is roughly consistent with the US Navy's 1818 circular on the design of the 20-star flag, which provided for dimensions of 14 x 26 feet. the 1822 dimensions reflected a reduction in the width of the bunting used to make Royal Navy flags, so ten years earlier the flags would have been somewhat wider, but I believe the length remained the same. In any case, the difference would have been insufficient to validate the Morton hypothesis.

By the way, the largest historic ensign in the collection of the British National Maritime Museum is that of the Spanish ship of the line San Ildefonso, captured at Trafalgar. It measures 9.8 x 14.4 meters. I don't know the dimensions of the ship, but this one might conceivably be about as long as the San Ildefonso was wide; San Ildefonso was a 74 gun, which in Royal Navy terms would have been about a 3rd rate.
Joe McMillan, 12-16 September 2003

It should be perhaps remembered that the table of sizes drawn up by Pepys in 1687 (referred to by Joe and quoted by Tim Wilson) remain recommendations only - although certainly indicative of naval practice in the late-Stuart period - because the author lost his office before they could be implemented. The exact size of the ensign concerned was, by the way, given as 28 breadths x 14 yards, but I have been unable to find any research as to how much fabric would be taken up by hemming at this period? The same source also gives a list of flags supplied from Chatham Dockyard in 1691, which includes an ensign for the flagship of 32 breadths (nearly 30 feet) along the hoist, by 49 feet long (9.1 x 14.6 metres).
Christopher Southworth, 16 September 2003