-------- Original Message --------
Subject: boundary length and naturality
Date: Sun, 6 Nov 2011 22:34:20 +0000
From: Brendan Whyte <[log in to unmask]>
To: mapsL <[log in to unmask]>



Chris,

International boundaries? My speciality!

To clarify the terminology a little: European countries often sought
'natural boundaries' menaing, not simply those following physical
features (mountains, riversS), but the 'natural limits' of the culture
of the central state. The 'natural boundaries' of France in early
political literature were not necessarily physical features, but the
extent of the French peoples and its subsidiary cultures (Breton,
Basque, Alsatian) as opposed to Teutonic or Iberian peoples and cultures.
It's a slippery word easily misinterpreted by the layman.


Some boundaries I have studied in detail:

Much of the Burma-China boundary follows moutain ranges, but in places
it is also defined as straight lines between pillars. Even when these
are atop a mountain range, it is no longer' natural' in the sense of
following the watershed or crest exactly. Then from a mountain peak the
line follows a straight line to the source of a tiny rivulet several
tens or hundreds of metres away. Then it follows the rivulet until it
becomes a stream. If the stream has been canalised, is it natural any more?

The Malaysian-Thai boundary was originally defined in treaties as
following a watershed atop a mountain range. But it has since been
redefined as straight line segments between pillars (even if these are
only metres apart). This may deviate from the curvilinear watershed line
by only centimetres, but is it any longer a natural boundary?

The northern India-Bangladesh border follows the courses of rivers... as
they were in 1713. Every year since those rivers have changed course,
but the boundary stayed fixed. Is the arbitrary line snaking through
rice fields nowdays still a 'natural' border? It doesn't look like one
to the average peasant (or foreign researcher!).


conversely, but similarly, if you were asked for a list of all artifical
boundaries, particularly those following meridians or parallels, would
you include the US-Canada boundary along the 48th parallel? The parallel
is the general course of the boundary, but the boundary is legally
defined as the straight-line between the pillars that have been erected.
Some are on the parallel, some are not, and the connecitng segments may
deviate from it by a mile or an inch. Thus you may be north of the 48th
but still in the US in places, and in other you may be south of it and
still in Canada.

And is not the Equator a 'natural line' around the planet, and if it is,
is not every parallel (eg the Tropics or the Arctic circle?).

So even where a boundary is defined by politicians in a treaty text as
following a natural feature (river, mountain range), the boundary
commission will have erected markers that may deviate from the course of
that feature for practical reasons, and the treaty protocol resulting
from the commission's work will define the boundary as being straight
lines between the pillars erected. So is it a natural boundary any more,
or simply a series of line segments approximating the feature?

Where a river is chosen for a boundary, the boundary may follow one bank
(has it been stabilised by an artificial bund?), or the thalweg (which
may move, or may be maintained by dredging), or the median line, or a
series of line segments connecting points detemrined by riverbank pillars.

And as for coastlines, there is Mandelbrot's famous article "How long is
the coastline of Britain?", which gets into fractal mathematics.

So determining a 'natural' boundary as opposed to an artificial one is
very much a matter of definition, and especially of scale.

And detemrining its length is also a matter of scale (and for rivers,
will change as the river meanders by accretion).

That said, you might want to read Ewan W Anderson's "International
boundaires, a geopolitical atlas", Routledge, 2003, which, for each pair
of bordering countries, gives a statistic of "approximate boundary
accordance with topographical features" in percentage terms.

I think Gideon Biger's "Encyclopedia of International Boundaries" did
something similar, but can't ifnd our copy at present.


Dr Brendan Whyte
Assistant Map Curator
Map Section
National Library of Australia
Parkes
ACT 2600
AUSTRALIA