LINGUIST List 3.177

Tue 25 Feb 1992

Disc: Replicability, Phonics

Editor for this issue: <>


  1. , Replicability in Linguistics
  2. "Bruce E. Nevin", replicability
  3. (Bruce E. Nevin", hooked on phonics

Message 1: Replicability in Linguistics

Date: Wed, 19 Feb 1992 23:40 PSTReplicability in Linguistics
Subject: Replicability in Linguistics

Re: Alexis Manaster-Ramer's comments about the infrequent use of
replicability as a criterion in evaluating linguistic hypotheses
(Linguist 3.146).

Another of the rare instances of the failure to replicate reported
data being used in a linguistic debate may be found in a recent paper
by E. D. Cook, "Linguistic Divergence in Fort Chipewyan" (Language in
Society 20: 423-440, Sept. 1991). In 1969 Ronald & Suzanne Scollon
claimed that that Chipewyan (Athabaskan) had "converged" with Cree
(Algonquian) at Ft. Chipewyan, Alberta, presenting a seemingly con-
vincing abundance of data in several papers and a book. Cook has
recently carried out fieldwork in the same community and dismisses
much of the Scollons' data as "spurious" (i.e., unreplicable, hence
unsupportive of their theory). I have to admit that when I first
read Cook's paper I was a bit shocked; such criticism seemed almost a
breach of scholarly etiquette. On reflection, it is clear that Cook
was merely behaving like a normal scientist, and the fact that
such behavior stands out like a sore thumb in linguistic discourse is
what should really be shocking.

Victor Golla
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue

Message 2: replicability

Date: Thu, 20 Feb 92 08:25:56 ESreplicability
From: "Bruce E. Nevin" <>
Subject: replicability


Replicability is especially important in sciences that deal
statistically with large populations (of molecules, photons, mice and
men), and which consequently are abominable at predicting behavior of
individuals. Most linguistics aims at modelling something that results
in the behavioral outputs of individuals. The idealizing and
abstracting away from raw performance data that characterize most
linguistics are not due to statistical averaging. Rather they
constitute some kind of claim about internalized norms or targets that
the individual language user's performance approximates.

Replicability also has to do with claims to objectivity based on
intersubjective agreement. There are well known epistemological
problems with this.

Surprisingly few studies in other sciences are actually replicated.
This has been a cause for some concern in recent years, with disclosures
of flagrant fabrications of data and results in geology, psychology,
chemistry, and other fields. I suspect we could all cite examples in

	Bruce Nevin
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue

Message 3: hooked on phonics

Date: Mon, 24 Feb 92 09:33:34 CShooked on phonics
From: (Bruce E. Nevin" <>
Subject: hooked on phonics

Just because I am suspicious of a $200 instant learn-to-read program
does not mean I'm against educating the poor. But I don't think that
high levels of encyclopedia sales in poor neighborhoods is reason to
give Hooked on Phonics a second look. I'm all too familiar with
encyclopedia sales strategies (at least those of 20 or so years ago)
where college students working summer jobs were forced into making
high pressure installment sales (no money down, easy financing,
no payments till March) to gullible residents of lower and working
class neighborhoods (play on their guilt, their hopes for their
children). Most people in this country would probably admit to
wanting to read /use language / write better than they already do
(literacy is, after all, an apple pie issue). But actually
increasing/improving literacy turns out to be a bit more complex;
it takes more than an advertising budget, and more than good intentions.

It's easy, as well, to belittle NCTE (though that organization works
hard for language rights) or any other arm of the education bureaucracy.
What I object to most about Hooked and Phonics and schemes like it
(and I repeat, I am not familiar with the program, only with its ads),
are the following:

1. the attempt to turn a perceived literacy crisis (and it is a _perceived_
crisis, as well as a real one) into big bucks
2. the suggestion that there is an easy solution to language problems
(like the eat-all-you-want, exercise never, thigh-master/tummyciser,
lose weight, copper bracelet, grapefruit diet)
3. the suggestion that entrepreneurs are better than professionals
(a very American notion, this, that the experts are fools--a feeling
that the popular-press treatment of linguistics reminds us about
all the time)

There is a common notion abroad that the schools have failed to educate
our children (in many cases this is true; in many it isn't). But of course
the schools can't do everything we tell them to do, and education isn't
as simple as sending children to school. Mass public schooling is an
assembly-line process, not hand-crafting. We shouldn't be surprised that
the line breaks down. The experts--teachers, cognitive psychologists,
reading specialists, the passionate literacy volunteers, anthropologists,
advocates of the poor and powerless, and, of course, the vast body of
readers [and nonreaders] themselves disagree over the nature of the
literacy problem and what to do about it.

So if you trust little ads in the backs of magazines and 30-second spots
on the radio, then by all means call and find out more. My initial
reaction is to put HOF in the category of replacement windows and
basement de-watering systems. I don't think it's the millennial
solution to the problem any more than Literacy 2000 is (don't tell
me the millennium turns in 2001, either; I saw the movie; I know
why Arthur C. Clarke chose the name). But I've been wrong before.
And the number is so easy to remember. 1-800-ABCDEFG.

Dennis Baron
Dept. of English office: 217-244-0568
University of Illinois messages: 217-333-2392
608 S. Wright St fax: 217-333-4321
Urbana IL 61801
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue