The Dictionary of Human Geography

Comp. by: LElumalai Stage : Revises1 ChapterID: 9781405132879_4_P Date:1/4/09
Time:15:20:53 Filepath:H:/00_Blackwell/00_3B2/Gregory-9781405132879/appln/3B2/re-
vises/9781405132879_4_P.3d

probability map A probabilitymapdisplays

the estimated likelihood that a pre-defined

event will occur at a specific point or defined

area within a geographical space, during a par-

ticular period of time. For example, probabil-

ity maps are used in meteorology to show the

likelihoods of above-normal rainfall occurring

around the globe (see, for example, the UK

Met Office’s seasonal forecasts at http://www.metof-

fice.com/research/). To interpret probability

maps, it is important to know how the prob-

abilities were calculated and interpolated

across the map space, and the period for which

the event is predicted (a 33 per cent chance of

annual flooding being worse than a 33 per cent

chance of decennial flooding). rh

process A flow of events or actions that pro-

duces, reproduces or transforms a system or

structure.

It was not until the 1960s that modern geog-

raphy was alerted to the complexity of the

concept of a process. Blaut (1961) insisted

that the standard distinction betweenspatial

structureand temporal process derived from

akantianismthat had been discredited by

what he called ‘the relativistic revolution’. It

was now clear, so he claimed, that ‘nothing in

the physical world is purely spatial or tem-

poral; everything is process’. In Blaut’s view,

therefore, ‘structures of the real world’ that

seemed to have a fixity and permanence were

simply ‘slow processes of long duration’. In

principle, most formalizations ofspatial sci-

enceappeared to accept this claim – Golledge

and Amedeo (1968) and Harvey (1969,

pp. 419–32) gave ‘process laws’ a central place

in theirmodelsofexplanationin geography

(seepositivism) – but in practice many studies

simply used distance as a surrogate for process

(hence ‘spatial processes’) and thus confirmed

the geometric cast oflocational analysis

andspatial analysisas these were conceived

at the time (seedistance decay).

Many of these locational–spatial models

depended onformal language systems; that is,

language systems whose elements have

unassigned meanings. Thexs andys in their

equations or the points and lines in their dia-

grams could refer to anything – they were

empty of concrete content – and so analysis

was governed by the relations between these

abstract elements in the language system itself:

by (say) the theorems of geometry, the calcu-

lus of probability theory or the mathematical

theory ofstochastic processes, rather than

by what Olsson (1974) called ‘the things we

are talking about’. These locational–spatial

models of spatial process were reviewed in

Cliff and Ord (1981).

Subsequently, however, there was a turn

towards human geographies based onordinary

language systemswhose elements have assigned

meanings. This allowed more substantive con-

ceptions of process to be utilized in human

geography: cultural processes, ecological pro-

cesses, economic processes, political processes,

social processes and so on. Early examples

included a focus on cognitive anddecision-

makingprocesses inbehavioural geography,

on the dynamics of thelabour processand

cycles of capitalaccumulationineconomic

geography,and onpractices ofsocial and sym-

bolic interaction incultural geographyand

social geography. This in turn, and of neces-

sity, opened up a sustained dialogue with the

otherhumanitiesand social sciences. For, as

Harvey (1973) recognized, ‘an understanding

of space in all its complexity depends upon an

appreciationofsocialprocesses[and]anunder-

standing of the social process in all its complex-

ity depends upon an appreciation of spatial

form’. One of his most pressing concerns was

thus to ‘heal the breach’ between the socio-

logical andgeographical imagination.

The distinction between ‘formal’ and ‘sub-

stantive’ definitions of process was cross-cut

bya seconddistinction drawn by Hay and

Johnston (1983), between ‘sequences’ and

‘mechanisms’:

(1) Process as sequence in time and/or

space. This view of process is character-

istic of the vertical themesof trad-

itional historical geographyand of

mathematical–statistical space–time

forecasting models. In both cases, the

account is morphological and descrip-

tive: compare Darby’s (1951) identifica-

tion of the processes that changed the

English landscape (‘clearing the wood’,

‘draining the marsh’ etc.) with Bennett’s

simple (1979) typology of barrier, hier-

archy, network and contiguity processes.

(2) Process as mechanism. This view of process

is characteristic of bothsystemsanalysis

in geography and of geographies con-

ducted under the sign ofrealism(which

require the identification of the relations

between the ‘causal powers’ of structures

and the conditions under which they are

realized). In both cases the account is

interactive and explanatory: it seeks to

show how – by what means, through

which networks – particular outcomes

materialize.

Gregory / The Dictionary of Human Geography 9781405132879_4_P Final Proof page 586 1.4.2009 3:20pm

PROBABILITY MAP