The Dictionary of Human Geography

non-scientific. For Popper, the division
between falsifiable and non-falsifiable state-
ments was inviolate, allowing consistent sep-
aration ofsciencefrom non-science
Popper first proposed falsification in his
1935 bookLogik der Forschung(translated as
The logic of scientific inquiry; Popper, 1959). It
was written as a response to discussions at the
Vienna Circle oflogical positivismthat he
attended, although it was a group to which
he was never admitted as a full member.
Logical positivists argued for the principle of
verification, according to which the truth of a
scientific statement was given by its corres-
pondence to real-world observations. Popper,
dubbed by one member of the Vienna Circle,
Otto Neurath, as ‘the official opposition’,
claimed to the contrary that scientific state-
ments were never verified, only falsified.
Because no one is omnipotent, it is impossible
to know whether in the future a scientific claim
will be disproved by a disconfirming instance
(and a single disconfirming instance is all that
is required to refute a general claim). For
example, the verified truth of pre-eighteenth-
century natural scientists ineuropethat all
swans were white was invalidated when
Captain Cook sailed toaustraliaand found
black swans. The growth of scientific know-
ledge, as Popper (1963) would later claim, was
based not on verifying theories, but on falsify-
ing them, and in the process developing alter-
natives that were less worse; that is, not yet
falsified. This was one of the basal proposi-
tions of Popper’s development of critical
rationalism.
In The Structure of Scientific Revolutions,
Kuhn (1970 [1962]) provided the root of a
trenchant critique of Popper’s position. Kuhn
portrayed the trajectory of science as punctu-
ated by periods of fundamental change he
calledparadigmshifts; for example, the move
from Newtonian to Einsteinian celestial mech-
anics. During such shifts, everything was in
flux, Kuhn argued,including criteria of falsifica-
tion. Popper had made such criteria fixed and
constant and yet, as Bernstein (1983, p. 71)
later put it, ‘data or evidence do not come
marked ‘‘falsification’’.’ Criteria of falsifica-
tion are moving elements, not existing outside
of debate, and settled only after change has
happened.
Falsification should have been an important
criterion for geographers during the period of
thequantitative revolution, when the dis-
cipline modelled itself most explicitly on the
natural sciences. But it wasn’t. If anything,
that disciplinary move was backed by logical

positivism, the very philosophy Popper

thought he had dispatched in the 1930s.

Alan Wilson (1972, p. 32), one of the leaders

of geography’s quantitative revolution, did

suggest the importance of falsification in a

1972 statement: ‘The essence of the scientific

method ... is an attempt to disprove theory –

to marshall observations to contradict the

predictions of the theory.’ But it was program-

matic, and never realized in Wilson’s practice,

or the practices of anyone else in geography.

This was precisely Kuhn’s point, and later

developed in science studies: falsification

could never be realized. tb

Suggested reading

Popper (1963, pp. 33–9).

family reconstitution A method used in

historical demographyto create measure-

ments in the absence of data on demographic

stocks and flows made available through

censuses and vital registration that applies

nominative linkage techniques to baptisms,

marriages and burials recorded in parish

registers (see Fleury and Henry, 1965). The

technique can also be adapted for use with

genealogies (see Henry 1956). It starts from a

marriage and links baptisms and burials of

children born to the couple as well as their

subsequent marriages. Family reconstitution

rules devised originally for use with French

parish registers by Louis Henry have been

adapted for use with registers elsewhere, with

a view to establishing a population at risk

or under observation so that age-specificmor-

talityandfertilityrates can be calculated.

Marriage age can be derived by linking bap-

tisms and marriage dates and information

can be extracted that will enable fecundity,

birth and pre-nuptial pregnancy rates to be

computed (see Wrigley, 1966a). Linkage is

far more successful wheremigrationis low,

and consequently cities and large towns have

rarely benefited from the technique, which has

been used primarily to reconstruct the popu-

lation of villages and smaller market towns

(see Wrigley, Davies, Oeppen and Schofield,

1997). rms

Suggested reading

Wrigley (1966a); Wrigley, Davies, Oeppen and

Schofield (1997).

famine A relatively sudden event involving

massmortalities from starvation within a

short period. Famine is typically distinguished

from chronic hunger, understood as endemic

Gregory / The Dictionary of Human Geography 9781405132879_4_F Final Proof page 239 31.3.2009 1:20pm

FAMINE