LINGUIST List 13.1351

Tue May 14 2002

Disc: Falsifiability vs. Usefulness

Editor for this issue: Karen Milligan <karenlinguistlist.org>

Directory

1. Greg Matheson, Falsifiability vs. Usefulness
2. Dan Everett, RE: 13.1348, Disc: Falsifiability vs. Usefulness
3. Robert Whiting, Re: 13.1334, Disc: Falsifiability vs. Usefulness
4. Mark Douglas Arnold, Re: 13.1348, Disc: Falsifiability vs. Usefulness

I am interested in this discussion as a teacher of English as a
foreign language and as a pragmatist, perhaps (in both senses of the
term), but not as a linguist, which I probably am not (in either sense
of the term, and certainly not that of this list).

Two observations on usefulness. The discipline of applied linguistics
seems almost independent of its parent discipline, pure
linguistics. The relationship appears to be different from that of
applied mathematics with pure mathematics, for example. Is the naming
of the body of knowledge or discipline that developed around the
desire for a principled approach to the study of language teaching
just a historical accident?

The second observation is about the maxim, "There's nothing as
practical as a good theory", of Kurt Lewin, the progenitor of action
research, the idea that practitioners can, should or do do research. I
believe this maxim leads to a corollary, "There's nothing as good as a
practical theory", given two axioms. The first is that theories are
just not good or not good, but that there is a cline of goodness
between the bad and the good. The second is that the cutoff point
between the good theory and the not good theory is not fixed, but
varies according to desire. With these axioms, we can see that the
better the theory, the more practical it is. The monotonicity of this
relationship between goodness and practicality allows us to conclude
that the more practical a theory is, the better it is.

Which is the conclusion I (and Lewin?) was looking for. Of course the
maxim may still be a false maxim. I don't think however that it is
equating goodness with usefulness.

- 
Greg Matheson 
Chinmin College
Taiwan Penpals Page: http://netcity.hinet.net/kurage

Message 2: RE: 13.1348, Disc: Falsifiability vs. Usefulness

Sharifian, Bouchard, and Lotfi (I finally spelled Ahmad's last name
correctly! Apologies to him for misspelling it in previous postings)
are all concerned about the issue of falsifiability and practice in
one way or another. It seems to me that they each take a position to
the effect that *falsifiability is more useful* in linguistic theory
than usefulness. But each response conflates falsifiability as an
epistemological notion with falsifiability as a matter of
procedure. The first is what we are discussing.

The responses posted in 13.1348 basically say that some people have
abandoned their hypotheses because of counterevidence. Maybe not much.
But they have. Therefore falsifiability does work. But this is not
falsifiability. To falsify something is to show it to be wrong. It is
not merely to convince someone that another way of doing things
handles the facts better than their original approach (that is
usefulness). And to show something to be wrong, one must overcome the
problems inherent in such an enterprise that the epistemologists I
have cited on this list have raised. None of the postings in 13.1348
do this. Instead they talk about falsifiability in practice,
i.e. someone changing their mind based on data, vs. epistemological
falsifiability, which is showing something to be wrong. The former
idea is harmonious with usefulness. The latter is not because, again,
one cannot show things to be wrong.

Let's consider some of their actual statements. Sharifian says: "I've
seen the arguments but they have not been cogent enough." Well, it is
more than possible that I have failed to be cogent. So I will concede
this. But now consider: "Popper's demarcation criterion has worked
very well in a lot of cases. And above all, Popper's ideas largely
influenced the way a lot of us now think about science and scientific
method." This has nothing to do with the epistemological issue. Has
*the idea* that something has been proven to be wrong had a practical
effect? Perhaps. But a minimal one. There is a huge literature on
this that argues that falsifiability has played an incredibly small
role. In fact, that was the lesson of Gale's previous posting, namely,
that such cases are so rare as to be both noteworthy and subject to
alternative explanations. Sharifian also says that: "Even if "one
simply finds no (or few, if Gale is correct) examples of
falsifiability playing a role in acknowledged advances in science",
which is definitely an oversimplification, still it doesn't harm the
LOGIC." This fails to respond to the last point of my last letter,
namely, that either sciences advances illogically or falsifiability is
not logically necessary to the advance of science. Therefore, and this
is vital from a pragmatist point of view, the discussion has failed to
unearth an clear exemplar that falsifiability makes a difference to
practice.

Lotfi's principal point is to remark that I confuse theories and
statements. No, I do not. Both are meant to fall under falsifiability.
And my previous remarks apply to both. He raises no new objections.

Finally, Denis Bouchard's remarks. Consider his rebuttal on my urging
that we consider usefulness as an alternative to falsifiability: "In
fact, isn't this just talk, and don't linguists generally present what
is the most useful descriptive tool they have come up with?" I must
say that agree with both of Bouchard's rhetorically expected answers
here: First, yes it is just talk. But, after all, that is all we can
do about anything on this list, I think (on the other hand, there is a
huge body of literature on pragmatism and my ideas on the matter are
shared to one degree or another by philosophers e.g. Quine, Rorty,
Putnam, Wittgenstein, James, Peirce, CI Lewis, Dewey, and many, many
others. For a simple introduction, I recommend the excellent and brief
book by Harvey Cormier, 2001, _The Truth is What Works_). What else
did Bouchard think we could do? Second, he is absolutely correct in
the portion of the sentence following 'and'. Read that carefully,
because it sums up my thesis well. (However, this then renders
Bouchard's next remark a nonsequitur: "Moreover, isn't this usefulness
an impediment to the progress of linguist theory?")

Anyway, on to his example. In the example he gives, a fairly standard
kind of justification of falsifiability among generative linguists, he
conflates parsimony & falsifiability. In the example he gives, neither
analysis has been falsified, but one is more parsimonious. Personally,
I would normally take the most parsimonious explanation, because I
usually find those easier to understand and easier to apply, and
because my head was made in the West. I have no objection to that at
all. Except that we should recognize that Bouchard has (i) given no
evidence for falsifiability in his example and (ii) overlooked a large
part of linguistics history in which hypotheses have been abandoned
and gone back to time and time again, even after having been
supposedly falsified. We go back and say, "Wait, if we interpret
hypothesis A to mean x + n, instead of just x like we thought, then it
is not falsified and becomes useful to us again." But this just means
that falsifiability is nonlethal to hypotheses, i.e. that it can
always be circumvented, which makes it less than useful.

Ultimately what I wanted to do in beginning this discussion is just
this - to show that falsifiability is a dubious notion, not nearly as
straightforward as linguists all too often think. This has been done.
Notice that not a single response has yet attempted to deal with the
serious objections I pointed out from Hull, Hempel, and
Lakatos. Because to answer those objections would be quite
difficult. Just so. Falsifiability is not easy to defend. And if it
isn't then it is not *automatically* the best way to proceed. You may
(want to) say that is how you go about your business but that, again
to quote Bouchard, is "just talk". It is, to be sure, talk that
emanates from the oracular names of the discipline, but that doesn't
make it any less likely to be used unreflectively.

Dan Everett

Message 3: Re: 13.1334, Disc: Falsifiability vs. Usefulness

I somehow have the feeling that I am being led around by the nose-ring
in this discussion. Because no matter how you word Dan Everett's
claim, whether "falsifiability has no application ever" or
"falsifiability is less than useful," it falls into the category of
such self-falsifying statements as "all sweeping generalizations are
invariably false" or "every rule has an exception." Such statements,
of necessity, include themselves in their domain. Therefore, if they
are true, they must be false, but if they are false then they prove
their own truth.

So Dan Everett's claim that falsifiability is useless is simply a
colossal waste of time. For if he proves his statement true by
falsifying falsifiability (with any useful result) then he will have
proved his statement false by producing a useful application of
falsifiability.

I say forget about Popper and Lakatos and Henry James for the time
being and go back to Sherlock Holmes: "... when you have excluded the
impossible, whatever remains, however improbable, must be the truth."
Regardless of whether one considers "the truth" to be the domain of
science or of religion (or even if one just considers science another
kind of religion), if there is such a thing as objective reality
independent of observation or perception, then anything that has been
excluded as impossible (falsified) does not remain a candidate for
objective reality.

Falsifiability lies in the simple fact that a proposition and its
opposite (or P and ~P) cannot both be true at the same time (although
both may be false). This means that if P is true then ~P cannot be
true and vice versa. If a P has a ~P that can be investigated by
direct observation, then P is falsifiable. If P is not falsifiable,
then it cannot be excluded as impossible and must be kept open
(forever) as a possible candidate for "the truth."

But neither falsifiability nor usefulness moderates the truth value of
a statement. Falsifiability merely determines how the statement can
be investigated. The current epistemological paradigm says than a
non-falsifiable claim cannot be investigated scientifically (i.e., it
is unscientific). But whether a statement is scientific or not does
not moderate its truth value either. Scientific does not mean true
and unscientific does not mean false. Scientific simply means that it
can be investigated using the tools and methods of science.

The value of falsifiability lies in being able to decline to
investigate non-falsifiable claims. A non-falsifiable
theory/hypothesis/statement/opinion is a "just-so" story and can be
put with Santa Claus, the Easter Bunny, the existence of Atlantis, the
location of Noah's Ark, and the latest translation of the Phaistos
Disk, that is, shelved (= ignored) indefinitely. In short, scientists
can decline to waste their time refuting irrefutable (non-falsifiable)
claims and be understood by other scientists (if not by the makers of
the claims or by the media).

But despite all this, it still appears that falsifications are not
called falsifications. One can think of the Michelson-Morley
experiments or the development of relativistic theory (which arose
from a falsification of Newtonian mechanics under certain
circumstances; of course, all this was pre-Popperian). Instead,
successful falsifications result in "paradigm-shifts," but this is
just a matter of terminology. One is reminded of the lines of Sir
John Harrington (1561-1612):

 Treason doth never prosper: what's the reason?
 For if it prosper, none dare call it treason.

So, whatever the reason, a successful falsification becomes a
paradigm-shift. But just because it is called something else, that
doesn't mean that there are no successful falsifications or that
falsifiability is useless any more than there is no such thing as
successful treason.


Bob Whiting
whitingcc.helsinki.fi

Message 4: Re: 13.1348, Disc: Falsifiability vs. Usefulness

Though the following might seem flippant, it is not meant to be.

Dan Everett, in 13.1334:

"So either science progresses illogically, or falsifiability is not part
of its logic."

Presumably Everett takes the apparent absurdity of the first proposition
to set up the coup de grace in the second. I humbly submit that the first
proposition is not as obviously false as many of us (seem to) want to
believe.

If that's true, it strikes me that Denis Bouchard has, in 13.1348, offered
the most penetratingly astute commentary thus far:

"In fact, isn't this just talk..."

Respectfully,

Mark Arnold