LINGUIST List 16.970

Thu Mar 31 2005

Disc: Historical Ling: Punctuated Equilibrium Model

Editor for this issue: Naomi Fox <>

To post to LINGUIST, use our convenient web form at


        1.    Marinus Van der Sluijs, Dixon's model of punctuated equilibrium

Message 1: Dixon's model of punctuated equilibrium

Date: 29-Mar-2005
From: Marinus Van der Sluijs <>
Subject: Dixon's model of punctuated equilibrium

Fund Drive 2005 is now on! Visit to donate now!
R. M. W. Dixon's 'The rise and fall of languages' (Cambridge University 
Press, 1997) has aroused much interest and justly so. In this concise
volume Dixon applies the biological model of punctuated equilibrium to the
theory of language evolution. The core of his argument is that languages do
not evolve gradually at a uniform rate, but alternately experience long
periods of equilibrium, punctuated by brief episodes of drastic change.
During phases of punctuation, languages diverge, differentiating into
groups that can be modelled as family trees; during phases of equilibrium,
languages in contact converge, obfuscating their genetic relationships as
modelled in family trees.

Dixon's discussion offers much food for thought, but I believe his model is
fundamentally flawed. In terms of Dixon's model, the global community is
currently in a period of punctuation, as rapid changes in communication
take place and the English language spreads its influence abroad. Dixon
also claims that in times of punctuation, languages split into family
trees. This is at odds with the observation that languages are seen to
converge now, rather than to diverge: in many parts of the world where
English is spoken in a bilingual context, local languages assimilate
themselves to English, sometimes modifying the English language itself to
the effect that creoles or pidgins are produced. From the perspective of
the diffusion of English, then, we seem to be looking at a period of
punctuation, but from the perspective of the ongoing fusion between English
and local languages the state of equilibrium would be a more apposite
description. Clearly, something is not quite right with the model.

In my view, one has to acknowledge that a language can be in states of
'equilibrium' and 'punctuation' at once, depending on the context in which
its evolution is placed. Convergence to neighbouring and unrelated
languages by definition means divergence from genetically related languages
if a language is diffused into a new territory. When the speakers of
Proto-Indo-European dispersed towards their historical homelands, they all
carried a form of Proto-Indo-European with them. Subsequent centuries saw
an increasing diversification within Proto-Indo-European, ultimately
leading to the formation of the daughter languages. This happened for the
most part under the influence of local 'substrate' languages that were
being absorbed into the Indo-European satellite communities. Thus,
Proto-Greeks modified their form of Proto-Indo-European under the influence
of the indigenous people in Greece, whilst Proto-Celts adjusted it under
the influence of the native population in their own area. In terms of
Dixon's model, no one would hesitate to qualify the differentiation of
Proto-Indo-European as a process of punctuation, in which a 'family tree'
was produced in a relatively short time. But that is not the whole story.
Zooming in on each particular homeland, a state of equilibrium is initiated
as soon as the speakers of Proto-Indo-European arrive: in this state the
respective substrate languages gradually merge with Proto-Indo-European,
undoubtedly going through a phase in which a Sprachbund is formed, followed
by the complete absorption of the local languages into Indo-European.
Chronologically, convergence and divergence happen at the same time: the
further local forms of Proto-Indo-European are adapted to local languages,
the more diversified the Proto-Indo-European language becomes. Exactly the
same can be argued for the formation of Romance languages from Latin. In
situations like these, 'punctuation' and 'equilibrium' cannot be described
as sequential phases of development, but have to be seen as synchronous
phenomena, different perspectives on the same development.

Marinus van der Sluijs
Port Moody, British Columbia, Canada

Linguistic Field(s): Anthropological Linguistics
                            Discipline of Linguistics
                            General Linguistics
                            Genetic Classification
                            Historical Linguistics
                            Language Description
                            Linguistic Theories
                            Philosophy of Language

Subject Language(s): None ()

Respond to list|Read more issues|LINGUIST home page|Top of issue