Ian Batten igb at batten.eu.org
Wed Jan 18 02:16:11 EST 2012

On 18 Jan 2012, at 0421, Daniel R. Tobias wrote:


> And why are those legal issues more significant than the ones that

> will arise with respect to other laws, treaties, regulations,

> standards documents, etc., that specify GMT or some other form of

> solar time, once a redefined UTC (or whatever it's called then) no

> longer closely tracks it?

There are certain pieces of legislation that refer to GMT either by name or by implication (UT, mean solar time, etc). The Interpretation Act 1978 S(9), as has often been quoted as an example, says:

> Subject to section 3 of the M1Summer Time Act 1972 (construction of references to points of time during the period of summer time), whenever an expression of time occurs in an Act, the time referred to shall, unless it is otherwise specifically stated, be held to be Greenwich mean time.

However, it's somewhat disingenuous to claim that UTC as constituted meets this requirement, but UTC without leap seconds doesn't. S(9) doesn't say "GMT +/- 1s", it says "GMT". Why is one second's error bar axiomatically OK, while 1 minute, 1 hour, etc, not? The legislation specifies a timescale. The assumption that +/-1 1s is acceptable is a matter for case law, which as yet hasn't arisen. The assumption that +/- 1min isn't acceptable is not acceptable is also a matter for case law, which as yet hasn't arisen.

If people wish to argue that the '78 Act requires GMT (and, note, the act only relates to the interpretation of other legislation, not to civil contracts or "what your watch says") then that's fine, but the natural interpretation of GMT in 1978 would be the historic GMT with rubber seconds. No one seems to be arguing that the UK should, or indeed is legally bound to, abandon the SI second. After that, which approximation is compliant is a matter for courts, and they haven't as yet been asked. Arguing that UTC without leapseconds has an unbounded error bar on GMT in the future is unlikely to get far, as courts are reluctant to get involved in discussions about a century into the future. Over the course of a lifetime, the divergence between UT and "new" UTC would be around a minute. The claim that a second's error is within the terms of the act but a minute's isn't is not supported by case law.

I suspect the same applies, mutatis mutandis, to other treaties and legislation. GMT or UT is specified, but error bars are not.


More information about the LEAPSECS mailing list