Moby Zipf
« previous post | next post »
A recent radio presentation: "The Law of Languages", Living on Earth 5/31/2019.
Other mass media: "Dolphins, aliens, and the search for intelligent life", Astrobiology 8/29/2011; "Dolphin Studies Could Reveal Secrets of Extraterrestrial Intelligence", 9/2/2011; To talk with aliens, learn to speak with dolphins", Wired 2/15/2011; "SETI Evolution: Searching for Aliens Using Whale Songs and Radios", Space.com 7/2/2013; "Dolphins Are Helping Us Hunt for Aliens", Nautilus 5/28/2018; "The Law all Languages Obey", Nova 9/26/2018; "Humpback Whale Communication and the Search for Alien Intelligence", SETI Institute 11/28/2018; "Humpback whales could hold the key to communicating with alien intelligence, say experts", Metro 12/6/2018; "What humpback whales can teach us about alien languages", Vox 12/6/2018.
Technical publications: Arik Kershenbaum et al., "Acoustic sequences in non‐human animals: a tutorial review and prospectus", Biological Reviews 2014.
See also Google Scholar searches for {Zipf dolphins}, {Zipf whales}, {Zipf birdsong}, {Zipf animal communication}, etc.
No time today for commentary — maybe later.
Victor Mair said,
June 1, 2019 @ 8:56 am
George Kingsley Zipf seems to have been an incredibly brilliant person. In addition to being Chairman of the German Department at Harvard, he was University Lecturer, a rare honor which meant that he could teach any subject he wanted. He died on September 25, 1950 at the age of 48 after a three-month illness. Yet, within that short life, not only did he discover Zipf's law, which has such important implications for linguistics, he applied similar models to human behavior (the principle of least resistance), frequency distribution of individual incomes and its implication for national unity and disunity, and other vital fields. It is said that his statistical insights can explain properties of the internet, even though he arrived at them before it was discovered.
I'm especially intrigued to learn that he worked with Chinese and wonder what he focused on in that regard.
All in all, a fascinating person. If there's not a biography of Zipf, he's ripe for one.
https://en.wikipedia.org/wiki/George_Kingsley_Zipf
Peter S. Shenkin said,
June 1, 2019 @ 9:47 am
I've read Zipf (many years ago) and of course the kind of rank-frequency relationships he popularized proved to also characterize a number of non-linguistic phenomena; stock-market fluctuations, fluctuations in voltage of domestic power lines, etc. I have to say that I found his explanations for why this relationship occurs rather opaque.
I do get that if there were fewer words that could each mean more things, it would be easier for the writer (or speaker) and that if there were far more words with more precise meaning it would be easier for the reader. Zipf suggested that the kind of relationship he observed is due to a compromise between these competing needs, which makes sense.
But it never struck me that the relationship he observed was dictated by that reasonable hypothesis. For instance, a uniform distribution of word use with some number of words would also satisfy that hypothesis, by striking a balance between a uniform distribution with more words with more precise meaning and one with fewer words with less precise meaning.
And some of the phenomena which fit the hypothesis don't strike me as amenable to explanations involving compromise between two agents or two processes (like speaking and listening, or writing and reading).
AntC said,
June 1, 2019 @ 8:33 pm
rank-frequency relationships [Zipf] popularized proved to also characterize a number of non-linguistic phenomena;
Yes, phenomena which are no sort of medium for representing information. So if dolphin squeaks/whale songs exhibit Zipf distribution, why is that any sort of evidence of intelligence or 'language'? (Same q wrt human squeaks, actually.)
~flow said,
June 4, 2019 @ 11:00 am
It would probably be prudent not to jump to conclusions when forms of expression are found to exhibit regularities as described by Zipf's law. It has long been known that a host of superficially completely unrelated phenomena display statistical properties that follow that distribution. From the Wikipedia (https://en.wikipedia.org/wiki/Zipf%27s_law) entry:
"""The same relationship occurs in many other rankings unrelated to language, such as the population ranks of cities in various countries, corporation sizes, income rankings, ranks of number of people watching the same TV channel,[5] and so on. The appearance of the distribution in rankings of cities by population was first noticed by Felix Auerbach in 1913.[4]"""
"""Wentian Li has shown that in a document in which each character has been chosen randomly from a uniform distribution of all letters (plus a space character), the "words" with different lengths follow the macro-trend of the Zipf's law (the more probable words are the shortest with equal probability).[11]"""
This is not to disparage Zipf's work; rather, what Pareto, Auerbach and Zipf found apears to be of such a pervasive and fundamental nature that it affects vocabularies, city populations, and river lengths alike. Or, to put it the other way around: It looks as though the same laws that govern statistical properties such as ranked river lengths are also effective in word frequencies; it would therefore seem that something other than a specifically linguistic explanation to account for these phenomena should be sought.
Also see https://mugrahum.com/2017/02/01/city-hierachy-testing-zipfs-law-in-suffolk/