Humanist Discussion Group

Humanist Archives: March 7, 2021, 7:48 a.m. Humanist 34.260 - psychology of quantification

              Humanist Discussion Group, Vol. 34, No. 260.
        Department of Digital Humanities, University of Cologne
                      Hosted by DH-Cologne
                Submit to:

    [1]    From: Henry Schaffer <>
           Subject: Re: [Humanist] 34.259: psychology of quantification (128)

    [2]    From: Manfred Thaller <>
           Subject: Re: [Humanist] 34.259: psychology of quantification (88)

        Date: 2021-03-06 16:15:55+00:00
        From: Henry Schaffer <>
        Subject: Re: [Humanist] 34.259: psychology of quantification

As someone whose nature and training is pretty quantitative, when I see “We
are bamboozled by numbers,” my first reaction is to think of "innumeracy".
Yes, people get very impressed by numbers - to the point of too often
ignoring context and units. I constantly see people comparing two
area/countries/groups/eras/... and ignoring the difference between the
count and the per-capita count.

Rather than feeling that numbers should be avoided, I'd like to see better
education about numbers.

--henry schaffer

On Sat, Mar 6, 2021 at 3:34 AM Humanist <> wrote:

>                   Humanist Discussion Group, Vol. 34, No. 259.
>         Department of Digital Humanities, University of Cologne
>                                 Hosted by DH-Cologne
>                 Submit to:
>     [1]    From: Willard McCarty <>
>            Subject: Material and embodied practices of people doing
> mathematics, symbolic logic, and calculation (66)
>     [2]    From: Inna Kizhner <>
>            Subject: Re: [Humanist] 34.258: psychology of quantification
> (87)
> --[1]------------------------------------------------------------------------
>         Date: 2021-03-06 07:17:49+00:00
>         From: Willard McCarty <>
>         Subject: Material and embodied practices of people doing
> mathematics, symbolic logic, and calculation
> [The following I've plucked from SIGCIS, with thanks. It points to a
> recent book
> on the subject. After a day devoted to searching around, I've unearthed a
> rather
> large list of items, from Woolf, Quantification: A history of the meaning
> of
> measurement (1961), containing a brilliant paper by Kuhn, to Kiss and
> Zétényi, Linguistic and cognitive aspects of quantification (2018).
> There's
> likely other possible, more discipline-specific lists, such as for history
> in its
> cliometric phase. What strikes me from an all-too-quick scan of this
> material, is that much the same tendencies in quantification emerge over
> and over again. From those who see the threat: reductive determinism;
> dismissal of the human; and imitation of natural science, whose abstract
> general laws remove individual experience in the process of making
> knowledge 'scientific'.
> “We are bamboozled by numbers,” Tim Ingold has
> written, “many of a magnitude that defy comprehension. But to add things
> up, they have first to be broken off from the processes that gave rise to
> them, from the ebbs and flows of life.” (2019: 667) Things have to be
> individuated, or returned to their individuality. Without that, however
> valuable quantitative methods may be, only an abstraction remains.
> Historian Jacques Barzun, in Clio and the Doctors (1974), makes the
> valuable
> observation about the difference between narrative and quantitative
> history, between which at that time swords were drawn:
> > In verbal history the critical sense acts upon the pattern
> > displayed—is it too neat? And on the motives alleged—are they
> > improbable? Whereas in quantitative history, despite the preliminary
> > account of how things were classified and computed and adjustments
> > made, the skeptical eye scans the method more than the results. (26)
> The key point, it seems to me, comes from modelling: not only (again
> to quote Barzun) to maintain "a sharp sense of the gap between the
> concrete world and the abstraction of quantity" but also between that
> world and story-telling that conforms to our current purposes.
> Comments?
> Yours,
> WM]
> -------- Forwarded Message --------
> Date:   Fri, 5 Mar 2021 16:21:01 -0600
> From:   Julie Cohn <>
> For reasons unrelated to the inquiry at the start of this email chain, I
> listened to a podcast called “Drinks with the Deal” that featured Will
> Deringer and a discussion of his book, /Calculated Values/,
> which “traces how numbers first gained widespread authority” in 17th
> century Great Britain (from MIT Press website). I have not read any of
> Will Deringer’s work, but it may be of interest.
> -Julie
> Julie Cohn, Ph.D.
> Non-Resident Scholar, Center for Energy Studies
> Baker Institute for Public Policy, Rice University, and
> Research Historian, Center for Public History
> University of Houston
> email: <>
> cell: 713.516.0849
> Author: The Grid: Biography of an American Technology (MIT Press,
> 2017)
> <>
> --[2]------------------------------------------------------------------------
>         Date: 2021-03-05 15:37:30+00:00
>         From: Inna Kizhner <>
>         Subject: Re: [Humanist] 34.258: psychology of quantification
> Dear Willard,
> Abraham Fet's *Pythagoras and a monkey: how mathematics caused a cultural
> decline* was written in the 1980s, this book is very much in line with the
> quotes and it is about the motivation of quantification. It is in Russian,
> written by a mathematician for the general public, and posted here
> Best wishes
> Inna Kizhner

        Date: 2021-03-06 14:36:03+00:00
        From: Manfred Thaller <>
        Subject: Re: [Humanist] 34.259: psychology of quantification

Dear Willard,

as far as you talk about historical research, I wonder whether you do
not overly emphasize the "counting" in quantification as opposed to the
need to have some formal "theory" to prove or disprove something by the
quantitative act. Well, you probably don't, as your pointer to modeling
at the end of your post indicates.

If you have time, I'd recommend RobertW. Fogel and  Geoffrey Elton:
/Which Road to the////Past?/, New Haven etc. 1983. That volume is a
dedicated discussion by Robert Fogel, one of the two most prominent
American Cliometricists - the '93 Nobel Prizer - and Geoffrey Elton, a
fortress of stubborn methodological conservatism in British History
writing. You may find yourself particularly interested in their implicit
as well as explicit handling of modeling. That Fogel supports it - well
his background is in economic theory. What is much more surprising is,
that Elton has no problems with models as such. Op.cit. 119-120: "Models
do dictate the terms of reference, define the parameters, direct the
research, and thus are very liable to pervert the search for empirical
evidence by making it selective. ... One would feel happier if those
models were derived from a study of the evidence and not borrowed from
supposedly scientific work in the social sciences --- if, that is,
historical method were allowed to control the borrowing.''

Whether you read this as a generalized "not invented here" remark or
something deeper, is a shallow question. What do me seems more important
is, that here and through my reading of the whole text Elton quite
consistently has much more of a problem with "foreign" concepts invading
the turf of the historian; that these are quantitative is a secondary
issue, even if the ways of quantification open up certain lines of attack.

As far as I can see - I am currently looking at that a bit more in
detail - it can most easily be understood, if you start with an
observation ofEdward Hallet Carr: /What Is History?/, originally 1961,
quoted hereafter from the Penguin edition of 1990. He observes the
discussion, particularly virulent at the time, how far history has to be
understood as a series of chance events and how far as a sequence of the
results of various processes. In that context he notices, that
historians seeing there society in a successful phase have always had
the tendency, to describe it as the result of structures and processes,
while historians living in a society in crisis had a tendency to
emphasize that everything at the end is a matter of chance. P. 78:
“Germans today [sc. 1961] welcome the denunciation of Hitler's
individual wickedness as a satisfactory alternative to the moral
judgement of the historian on the society which produced him.”

At least for the wave of quantification in the sixties thru eighties,
particularly in Germany, that is quite easy to prove: The more
historians considered it necessary to overcome the tradition, the
friendlier they looked at social sciences as a provider of broader
models, which lead to an interest in empirical methods, which lead to
quantification. The more they wanted to write an apologia of history,
the more they abhorred models and quantification. As Germany had
particularly much to apologize about her recent history at the time, it
is not surprising, that that conflict was specially obvious there.

"Counting" as such was not the issue.

This "if you are progressive, quantify" may be not so obvious in
literary studies. Let's face it: Concordances were not the chosen weapon
of the brilliant young theorists of literary studies in those decades.
(What is frequently overlooked is, that Busa did NOT produce the first
major computer generated concordance. That was John W. Ellison in 1956.
Busa's everlasting merit is, that he connected the venerable 18th and
19th century tool of the concordance to a linguistic model.) So while
early adepts of computers and humanities [ == literary studies in those
circles ] saw themselves as innovators, methodologically the majority of
them was rather conservative. Cf. David Hoover: “The End of the
Irrelevant Text: Electronic Texts, Linguistics, and Literary Theory”,
in: Digital Humanities Quarterly 1.2 (2007),
And quantitative stylistics was a very small niche then, which scraped
along with such small samples of text, that it was very easy not to take
it serious.

So quantification as such may never have been a bone of contention; the
admissibility of the methodologies needed to employ it meaningfully
quite definitely was. But nagging about statistics not understood was of
course always easier than engaging with the theories.

Best regards,

Prof. em. Dr. Manfred Thaller
Zuletzt Universität zu Köln /
Formerly University at Cologne

Unsubscribe at:
List posts to:
List info and archives at at:
Listmember interface at:
Subscribe at: