Re: Why is high high? (Harry Erwin )

Subject: Re: Why is high high?
From:    Harry Erwin  <herwin(at)OSF1.GMU.EDU>
Date:    Wed, 26 Aug 1998 08:43:40 -0400

At 11:10 AM 8/26/1998, Pawel Kusmierek wrote: >Dear list-members, > >In several Indo-European languages (e.g. English, German, Italian, >Polish) words 'high' and 'low' are used to describe sounds of big >and small frequency, respectively. Do any of you know if this >relation appears in other (especially, non-Indo-European) >languages? > >Moreover, what may be the source of the relation? What has a >vertical linear distance (high/low) to do with sound frequency? >When you look at people, the relation of size and frequency >appears to be inverse: usually tall ('high')people (men) talk and sing >at lower frequencies than short ('low') people (women, children). Big >things sound lower than small things: a piccolo is smaller than >a tuba. > >I read in a review that as frequency of a sound increases, the >perceived location rises in elevation (I have not the original papers >yet). Could this be the cause? > >But what are the physiological bases of this perceptual >phenomenon? >Is it caused by some selective attenuation/amplification by pinnae? >Or is it a property of auditory centers in brain? Is it inherited or >learned? > >If it is inherited, it should have an evolutionary cause: did high- >frequency sounds come to an australopithecus from high elevation >(birds)? and low frequency sounds from low elevation (sounds of >buffalo's steps transmitted via ground)? > >If the perceptual phenomenon is learned, then again: do high >frequency sounds come to an infant from high elevation and low >sounds from low elevation? > >Can anyone comment my questions? > Multipath reflections in the pinnae seem to affect the intensity spectra of wideband sounds. The human and bat literature have identified a notch that shifts up for high elevations and down for low elevations. I've also seen evidence for alternative explanations. Harry Erwin, Internet: herwin(at), Web Page: Senior Software Analyst supporting the FAA, PhD candidate in computational neuroscience--modeling how bats echolocate--and lecturer for CS 211 (data structures and advanced C++). McGill is running a new version of LISTSERV (1.8c on Windows NT). Information is available on the WEB at

This message came from the mail archive
maintained by:
DAn Ellis <>
Electrical Engineering Dept., Columbia University