Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site pucc-i
Path: utzoo!watmath!clyde!floyd!harpo!ihnp4!inuxc!pur-ee!CS-Mordred!Pucc-H:Pucc-I:ags
From: ags@pucc-i (Seaman)
Newsgroups: net.philosophy
Subject: Re: Why do mirrors reverse left & right, not up & down?
Message-ID: <184@pucc-i>
Date: Mon, 6-Feb-84 15:13:02 EST
Article-I.D.: pucc-i.184
Posted: Mon Feb  6 15:13:02 1984
Date-Received: Thu, 9-Feb-84 13:47:20 EST
References: <541@bbncca.ARPA> <543@bbncca.ARPA>
Organization: Purdue University Computing Center
Lines: 49

> RIZZO: I appreciate your point.  I do that agree your "front-to-back
> mapping" accounts for why points on the reflected object correspond 
> to points in the image that appear to be on the "wrong side".  For-
> mally restated, your "front-to-back mapping" is an automorphic (maps
> an object onto itself) transformation involving a 180 degree rotation
							       ^^^^^^^^
> about a top-to-bottom axis.  In effect, yours is a proposal to trans-
> fer the explanatory burden from one distinction, left/right, to an-
> other, front/back.

Wrong.  The front-to-back mapping is not a rotation at all, but a reflection.
There is no top-to-bottom axis, or left-to-right axis.  There is only the
plane of the mirror, which represents the plane of reflection.

It is exactly this tendency to replace a reflection with a rotation which
confuses people into thinking there is a left-right reversal.

> RIZZO: ... Why do mirrors use a vertical axis, & not, say, a horizontal one?
> 
> (A brief pause in the conversation ensues. Then:)
> 
> PSEUDO-TAYLOR: They don't.  We do.
> 
> RIZZO: Intriguing!  Why do we?
> 
> PSEUDO-TAYLOR: Because our vision is binocular.  Binocularity creates
> handedness.

PSEUDO-TAYLOR was on the right track, but he came up with the wrong
explanation.  The real reason that WE (not mirrors) use a vertical
axis rather than a horizontal one is that:

	(1) It is more natural.  We turn around much more often
	    than we do cartwheels.

	(2) Humans (and many objects) are approximately symmetric
	    left-to-right, but not even close to symmetric top-to-bottom.
	    If you lie down horizontally and look into a mirror, it will
	    seem more natural to use a rotation about a horizontal axis,
	    since this produces an image more nearly in agreement with
	    what you see in the mirror.

-- 

Dave Seaman
..!pur-ee!pucc-i:ags

"Against people who give vent to their loquacity 
by extraneous bombastic circumlocution."