Re: JASSS Critical Review of Thought Contagion

Aaron Lynch (
Mon, 26 Apr 1999 10:42:40 -0500

Message-Id: <>
Date: Mon, 26 Apr 1999 10:42:40 -0500
From: Aaron Lynch <>
Subject: Re: JASSS Critical Review of Thought Contagion
In-Reply-To: <>

At 01:07 AM 4/26/99 PDT, Paul marsden wrote:
>For those that have not read Aaron Lynch's Thought Contagion,

It is important that you note that this review really is intended for those
who have not yet read the book. It contains much misinformation that would
only be plausible to someone who has NOT read the book, and the review was
indeed presented to readers who mostly have not read the book. It contains
numerous criticisms that you did not present on the memetics list, where so
many have already read Thought Contagion that the remarks would have seemed
far less plausible--except, of course, to those who already oppose "thought

> the Journal of Artificial Societies and Social Simulation
> (JASSS) have published a critical review at
>Aaron will be publishing an undoubtedly vigorous response to my review in
the next issue.

Below is a rough-draft preview of what I will be saying in my vigorous

Reply to Marsden Review of _Thought Contagion_

by Aaron Lynch

Marsden's review of _Thought Contagion_ in Volume 2, Issue 2 of the Journal
of Artificial Societies and Social Simulation has been (published at
<>) fails to mention that
it comes on the heels of a contentious debate between me and him on the
definition of the word "meme." I have advanced a definition for which
memetics theory can be expressed using propagation event diagrams and
differential equations. To do this, I keep the word's meaning fairly
specific: a meme is only an interpersonally copied memory item in the brain
(see Lynch, 1998). Marsden and his associates Blackmore (1999) and Gatherer
(1998), on the other hand, advance definitions that allow for "memes" to be
instantiated in a variety of media or behaviors and for which propagation
dynamic quantification methods are unclear. Yet instead of advancing
propagation dynamics models for his less specific definition of "meme,"
Marsden vaguely criticizes my differential equations as "arbitrary
mathematical manipulations" and my propagation event diagrams as
"permutations and combinations" (Marsden 1999a). As I do not discuss
"permutations" at all, I must conclude that Marsden either does not
comprehend the symbolic and quantitative model I have offered, or that he
is deliberately misrepresenting it to make it look silly. His present
review seems to touch on the transmission event diagrams yet again, this
time falsely labeling them as "laws of combination and separation." Once
again, this shows an abject lack of comprehension in that I do not discuss
anything called "separation" events at all, let alone "laws" of separation.
Marsden does not attempt to identify any section of my paper he takes as
discussing "laws" of meme separation, nor does he quote anything to this
effect. Instead, he implies to the unwary reader that such material must
exist, even though it does not. I do, however, discuss meme combination
events and introduce a way of diagramming them. Yet the diagrams are only
representations of some of the vast number of possible types of
transmission events; they are not presented as "laws." (Similarly, the
chemical diagram N2 + 2O2 --> 2(NO2) is not a "law" of combination of
elements; it is merely a representation of one of many possible
Nitrogen-Oxygen reactions.) Finally, despite the presentation of
non-metaphoric memetics in the paper, Marsden refers to a so-called
"thought contagion metaphor" and goes on to claim that I take it literally.
The high irony is that nothing in the paper depends on metaphor to either
genes or biological contagions--again demonstrating Marsden’s vociferous
lack of comprehension. While classical social science curricula do include
statistics courses, they usually omit methods of quantitative population
dynamics. Unfortunately, that can sometimes lead to anti-mathematical and
anti-quantitative commentaries that need not be taken at face value.

Marsden's review of my non-technical book _Thought Contagion_ demonstrates
that by carefully choosing which sentences and sentence fragments to quote
out of context, and how to mislabel them, one can make any book look
totally silly. So I invite JASSS readers who might be misinformed by the
Marsden review to read chapter 1 of Thought Contagion online at
<> and judge for themselves whether the
book is really as silly as Marsden makes it look.

Marsden did actually read enough of _Thought Contagion_ to discover some of
its topics. For example, there really is a section on the evolution of
memes pertaining to homosexuality. In a nutshell, it says that adherents of
the taboo out-procreated more tolerant people over the course of many
generations in ancient times, leading to increased prevalence of the taboo.
Then _horizontal_ transmission kicked in as people maligned homosexuality
to "prove" their adherence to the taboo. As the taboo becomes extremely
widespread, most homosexuals live heterosexual lives, leading them to
reproduce any genes involved. As the genes gain prevalence, the rate of
taboo dropout increases. Gene carriers who have dropped the taboo are more
sexually and socially motivated to spread acceptance of homosexuality than
are non-gene carriers who drop the taboo. So the rising gene prevalence can
lead to a self-sustained propagation of pro-gay memes. (Horizontal
transmission, again contrary to Marsden's claim that I ignore horizontal
transmission.) That, in turn, can lead to lower gene prevalence in the next
generation, and even lower prevalence of pro-gay memes. All of this leads
to potential fluctuations over long time spans. Incidentally, there is a
brief mention of how beliefs about anal sex become involved with
homosexuality taboos. Yet despite an multi-faceted discussion of
homosexuality taboos, Marsden excerpts a sentence on anal sex and presents
it as if I had offered it as the answer to the question "What memes deter
homosexual behaviour?" He has thus taken a complex argument and
misrepresented it as childishly silly. Marsden uses similar misleading
tactics in all of his other indented excerpts.

The "appendix" in which Marsden supposedly lists "Lynch's Seven Modes of
Memetic Transmission" in the same indented format is not based on excerpts
at all. There, the wording of each mode is Marsden's, and phrased so as
make a mockery of what is actually said in Chapter 1 of the book. Finding
out what I actually said, however, is as easy as reading chapter 1 online.

In order to find _Thought Contagion_ a "spectacular failure" in putting
memetics on a more serious footing in the sciences, Marsden demands that
the book serve a far broader function than the one it claims to serve. In
the first sentence of the preface, I state that the book "introduces a new
branch of science dealing with ideas that program for their own
retransmission." That is, ideas that gain propagation by influencing or
manipulating adherent's idea-propagating behaviors. It is NOT offered as a
general theory of cultural evolution, nor a general theory of social
contagion. Even within the subject of memes, the focus is not on the field
in general, but rather, on the small but important subclass of memes that
play a particularly active role in causing their own retransmission. Even
within this sub-subcategory, I mostly leave previously well-studied topics
such as urban legends up to other researchers.

Some take a very expansive definition of the word "meme" and envision
"memetics" as covering almost all topics in social science and social
philosophy, such far flung topics as whether the self is an illusion
(Blackmore, 1999). For them, Thought Contagion may indeed be a
disappointment. I, on the other hand, hold only the more modest and
scientifically conservative agenda of demonstrating that there are a
reasonable number of phenomena that really call out for the evolutionary
replicator analysis of self-spreading brain-stored information. Like Rogers
(1995), I narrow my focus to discuss only a subset of cultural evolution
phenomena, and this does not call for a survey of literature in the broader
field as Marsden seems to imply.

Now the book does say on its cover "THOUGHT CONTAGION: How Belief Spreads
Through Society: The New Science of Memes." While this may seem to claim a
far wider territory than the book actually covers, I should explain that
the subtitle "How Belief Spreads Through Society" is only intended to
identify the subject area in general terms to people who have never heard
the word "meme." The line "The New Science of Memes" is only intended to
place the book in the subject of memetics to those who *have* already heard
of memes. I suspect that reading too much into the cover could be what lead
Marsden to the rather inflated hope that the book would somehow cover both
of the vast territories of Evolutionary Culture Theory and Social Contagion.

_Thought Contagion_ is indeed filled with hypotheses awaiting empirical
investigation, and warns the reader of this in advance. In the preface, I
express the hope that publishing hypotheses will stimulate the research
needed to test them. Likewise, the publication of hypotheses about
biological evolution, biological contagion, etc. has historically played an
essential part in stimulating empirical studies.

As for the more technical definition of "meme" mentioned by Marsden's
review, It can be found in the paper Units, Events, and Dynamics in
Memetics Evolution (Lynch, 1998). I quote it below:

"MEME: A memory item, or portion of an organism's neurally-stored
information, identified using the abstraction system of the observer, whose
instantiation depended critically on causation by prior instantiation of
the same memory item in one or more other organisms' nervous systems.
("Sameness" of memory items is determined with respect to the
above-mentioned abstraction system of the observer.)"

Removing the philosophy of science about abstractions (explained in the
paper), this becomes more simply:

"MEME: A memory item, or portion of an organism's neurally-stored
information, whose occurrence depended critically on causation by prior
occurrence of the same memory item in one or more other organisms' nervous

This definition identifies the minimum conditions needed to achieve the
recursive process (or recursive algorithm) that forms the basis of
evolution by natural selection in interpersonally transmitted brain-stored
information. Those trying to narrow the focus to a restrictive or
anthropocentric definition of "imitation" (e.g., Blackmore, 1999) have yet
to show that their proposed restrictions are necessary. This definition
becomes the basis of transmission event diagrams and differential equations
(Lynch, 1999) useful in developing computer simulations of the meme
transmission patterns discussed in the book _Thought Contagion_.

The differential equations, for instance, treat horizontal and vertical
propagation as terms in the equation and give them their own separate
parameters. Thus, it does not assume that horizontal or vertical
transmission will be more prominent in any given case, but treats it as a
matter of empirical and quantitative investigation. For some movements, we
should expect horizontal transmission to predominate, while for others we
should expect vertical transmission to predominate. It is a quantitative
question to be analyzed with quantitative methods. Yet Marsden concludes
his review by recommending the Blackmore 1999, a book that takes horizontal
transmission as faster than vertical transmission without any mathematical,
computational, or empirical methods to back it up. Despite reference to
Lynch (1991), which contains the same mathematical methods of determining
the relative impact of horizontal and vertical transmission presented in
Lynch (1998), Blackmore ironically asserts that it is I who does not
clearly differentiate the effects of vertical and horizontal transmission.
This demonstrates a lack of mathematical comprehension similar to
Marsden’s. (While Blackmore 1999 does discuss some quantitative works in a
section titled "Sociobiology and culture on a leash," this section merely
recapitulates the comments of Dennett 1997.) Further lack of comprehension
is seen in Blackmore’s claim that I treat virtually everything we know as
memes, despite my explicit and formal statements to the contrary.
Fortunately, Blackmore has agreed to correct this particularly egregious
error in future printings. Yet we should feel no surprise that Marsden, in
demonstrating a lack of comprehension of quantitative memetics, concludes
his book review by endorsing another work that also shows lack of
comprehension for mathematical memetics.

The quantitative memetics presented in Lynch (1998) also allow for
falsifiability and animal experimentation. Unlike Blackmore (1999), there
is no anthropocentric definition of the word "meme." The thesis can
therefore be tested with experiments in animal populations without raising
the ethical issues of experimentation on human societies.

Contrary to Dawkins (1999), we should not look for memetics to be "given
best shot" in any non-technical work, whether Lynch (1996) or Blackmore
(1999). We should also not expect a "best shot" to take Zen Buddhism as
part of the theory (Blackmore, 1999) rather than as an object of analysis.
Instead, we should expect the strongest expressions of memetics theory to
come in technical and quantitative terms, since the thesis of differential
survival and propagation is essentially a quantitative as well as
qualitative one. Yet sociobiologists may have a conflict of interest in
directing attention away from such technical memetics works, as this leaves
their own frequently rival hypotheses on safer ground. We may therefore
have to set aside some their opinions from time to time. Likewise, we may
at times have to set aside the opinions of some social scientists who,
though sympathetic to memetics, simply do not comprehend its more technical

One of the observers of the debate between myself and Marsden wrote
privately to express his opinion that Marsden's harsh comments were coming
because he felt threatened by my work--especially with the fact that it has
been expressed in quantitative terms suited to computational and empirical
investigation. Whether Marsden really feels threatened or not, I cannot
say. Yet a distorional effort to dissuade people from reading the work is
not the proper way to seek refutation.


Dawkins, R. 1999. Foreword to The Meme Machine by Susan Blackmore. Oxford:
Oxford University Press.

Dennett, D. 1997. The Evolution of the Evaluators. Paper presented at the
International School of Economic Research, Sienna.

Lynch, A. 1996. Thought Contagion: How Belief Spreads Through Society: The New
Science of Memes. New York: Basic Books. Chapter 1 online:

Lynch, A. 1998. Units, Events, and Dynamics in Memetic Evolution. Journal of
Memetics-Evolutionary Models of Information Transmission.

Marsden, P. 1999a. A Strategy for Memetics: Memes as Strategies. Journal of
Memetics - Evolutionary Models of Information Transmission, 3.

ROGERS E. M. 1995. Diffusion of Innovations, fourth edition, New York: The
Free Press

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)