Re: Substance and Form

Aaron Lynch (
Mon, 01 Jun 1998 13:57:34 -0500

Message-Id: <>
Date: Mon, 01 Jun 1998 13:57:34 -0500
From: Aaron Lynch <>
Subject: Re: Substance and Form
In-Reply-To: <>


>> I don't have the 1971 edition, but Rogers 1983 definitely talks about not
>> just artifacts and behaviors, but also propagating information in human
>> hosts. (Though he does not use the term "host.") To quote from the=
>> by Rogers himself, (p. XIX), "...The diffusion of innovations, thus, is
>> essentially a social process in which subjectively perceived information
>> about a new idea is communicated." Later sections talk about the spread=
>> "knowledge," "software information," "opinions," etc. These all qualify=
>> mnemons in my usage.
>Now that is interesting, because he doesn't say that in the 1971=20
>edition. Obviously Rogers was influenced by the growth of ideational=20
>anthropology in the late 60s and 70s. Apparently there's a new edition=20
>out now (which will be the 4th, I think) so it will be interesting to=20
>see what Rogers currently thinks.
> =20
>> One mnemon that probably co-propagates with the behaviors and artifacts=
>> a belief that cowrie shells are precious, though I do not have a relevant
>> study at hand.
>> Back when you were very young, someone probably explained to you =A31=
>> were more precious than other pieces of paper. Explanations such as these
>> probably affected the information stored in your brain. On the other=
>> send 100 kg. of =A31 notes into some societies, and you might be able to
>> observe them being used for cooking fuel--unless you are very careful to
>> spread some ideas along with the papers. The great social distance from
>> those using cowries in Africa may make the co-propagating ideas less
>> apparent to people in Europe and North America, but this does mean that=
>> ideas are absent.=20
>I also have reservations about the issue of transmission/replication of=20
>beliefs. Even accepting for the present that belief is actually=20
>transmitted (which I don't accept - but that's another argument),=20
>there is still a substantial body of evidence, beginning with the=20
>classic work of LaPiere in the 1930s which suggests that professed=20
>beliefs are of little or no relevance to subsequent behaviour=20
>(reviewed by Cronk 1995, pp.183-185). Belief, even if we could=20
>say that it is transmitted (which I don't think we can, what is=20
>transmitted is a language statement eg. 'I believe x' - a=20
>performative utterance in the jargon of linguistic philosophers),=20
>is largely irrelevant to the material and behavioural=20
>manifestations of culture in the real world.

"Largely irrelevant" admits of further definition. Such salient behaviors
as turning to face Mecca several times a day versus attending a church on
Sunday really call out for explanations. Also, a belief in the importance
of Islamic law may also clash with, say, a belief in the validity of Hindu
caste systems. Let's hope this doen't lead to nuclear war.=20

>> What you can count are hosts of mnemons, each of which corresponds to a
>> single mnemon instantiation. This does involve making functional
>> definitions, as with surveys and polling methods. Yes, there are cases
>> where counting hosts is more difficult than counting objects, but I have
>> also pointed out problems in usefully counting instantiations of certain
>> artifacts. Using Mormon membership data is indeed an indirect gauge of
>> belief prevalence, but indirect measurements are performed in all sorts=
>> sciences. Ultimately, we want to design a survey that asks questions such
>> as "Jesus visited North America: [Y/N/?]," "The Book of Mormon was=
>> in golden tablets to Joseph Smith: [Y/N/?]," etc., with only respondents
>> answering in certain ways being considered "Mormon." Such surveys are=
>> indirect measures of the prevalence of ideas, but this does not mean that
>> the ideas do not exist (or that they are useless abstractions).
>This is precisely the approach that Quine (1960) uses to define=20
>linguistic behaviourism. Quine refers to =91linguistic behaviour=92,=20
>the tendency to affirm or deny when presented with linguistic stimuli. =20
>The =91true=92 or =91false=92 markers are then simply the subject=92s=
>to the stimuli (the memes as bald concepts without value markers).=20
>However, Quine is not so easily brought into the orthodox=20
>memetic fold. The word =91behaviour=92 is significant in his philosophy. =
>For the Quine of the early 1960s, there was little prospect of any=20
>knowledge of internal states. Individuals utter sentences and other=20
>individuals affirm or deny them. The intentions or belief states of=20
>the transmitter are not under consideration, merely the memes themselves=20
>as transmitted.
>In any case, is it not an invalid assumption that members of=20
>religious groups can actually give answers to the sort of=20
>questions you ask? You seem to be defining possession of the=20
>religious 'mnemon' to be productive of theological knowledge. If=20
>you take 100 Presbyterians and ask them to explain predestination=20
>or irresistible grace, the majority of them will look at you=20
>blankly. Likewise, in my experience, many Western Buddhists seem=20
>to confuse reincarnation with metempsychosis.

Specific memes that fail to contribute much to the long-term spreading of
the overall meme package may evolve to lower prevalence, so that there
could be rather little difference between a Christian who calls herself a
"Presbyterian" and one who calls himself a "Lutheran."=20

>Bear in
>> mind that quarks, magnetic fields, etc. are also measured indirectly. You
>> cannot "actually count" quarks, for instance. Even cowrie shells are
>> measured indirectly, too, based on the opinion of a human observer who
>> might sometimes count counterfeit cowries, similar shells, damaged=
>> too small shells, etc. When using a survey to classify respondents as
>> Mormon or non-Mormon, you can also ask questions such as "how many=
>> do you have?" The result will again be an indirect measurement of actual
>> numbers of children. Methodology will always be an important topic for
>> empirical research.
>Yes, and methodology in memetics is the biggest problem we have. =20
>If you adopt your questionnaire approach, you are: a) performing a=20
>behaviourist experiment, which b) will probably reveal that all=20
>those who are labelled as Mormons constitute a highly=20
>heterogeneous group with regard to the beliefs attributed to=20
>Mormonism as a whole, with the majority (as with any religious=20
>group) largely ignorant of even the most basic theology of the=20
>group. In any case, few of these professed beliefs will affect=20
>their behaviour (Cronk 1995, pp.183-185).

I agree that relatively few beliefs in the overall official package of
Mormon teachings affect behavior in a serious way. But it only takes a
handful of beliefs to affect important things like reproduction rates, for
instance. How many beliefs in the Islamic faith specifically influence
adherents to face Mecca 5 times a day? Rather few, again.=20

>> Bear in
>> mind that quarks, magnetic fields, etc. are also measured indirectly. You
>> cannot "actually count" quarks, for instance.
>But memetics is not built around a theory analogous to quark=20
>theory (maybe it should be, if it will solve the problem of how=20
>to count mnemons). Rather it is analogous to=20
>genetics/epidemiology. In those two theories, you do have to count=20
>infections or genes directly. Therefore we have to try to stick=20
>to things we can count directly.

Correct me if I am wrong, but I was under the impression that genes are
never observed directly. Rather, restriction enzymes, oligonucleotides,
etc. are used to infer the presence of certain nucleotide sub-sequences,
the presence of which is then used to infer the existence of a longer
sequence. It is not a simple matter of turning on the electron microscope
and having a look. Even with infections, the observation is often quite
indirect, as with the testing for antibodies whose presence leads one to
infer the presence of a particular microorganism.=20

>Cronk, L. (1995) Is there a role for culture in human behavioral=20
>ecology? Ethology and Sociobiology 16, 181-205.
>Quine, W.v.O. (1960) Word and Object. Technology Press, MIT.

--Aaron Lynch

This was distributed via the memetics list associated with the
Journal of Memetics - Evolutionary Models of Information Transmission
For information about the journal and the list (e.g. unsubscribing)