Tracking the Country: Railways and Population
Change in Industrializing England, 1845-1914
An Application of Historical Geographic Information Systems (GIS)
Robert M. Schwartz
Department of History
Mount Holyoke College
South Hadley, Massachusetts
|
Is there no nook of English ground secure From
rash assault.[of railways]?. . . William Wordsworth, 1844. |
The iron rail proved a magicians' road. The locomotive gave a new celerity to time. It virtually reduced England to a sixth of its size. It brought the country nearer to the town and the town to the country... It energized punctuality, discipline, and attention; and proved a moral teacher by the influence of example. Samuel Smiles, 1859 |
Railways and Internal Migration
Did railroads facilitate internal migration? Did they hasten rural depopulation? Accelerate rural to urban migration?
Did the railways facilitate migration from countryside to town? Did they hasten rural out migration?
My main question: To what extent did the extension of the rail system into small towns and villages retard out migration? Rather than hastening a rural exodus, to what extent did that the coming of railways and the commercial opportunities they brought open new jobs for local people, and thereby serving to stem or postpone the pace of rural out migration.
Historical GIs
Helps identify and interpret the relationship between migration and railways.
The indispensable source for this inquiry is the Historical GIs of the British Isles. This is a vast data base created by: Humphrey Southall and Ian Gregory at Queen Mary and Westfield College, University of London. Thanks to their efforts and generosity, I am able to drawn on their data and geographic coverages, to which I add data on the growth of the railway system. Hence, the essential sources for this study are:
Preliminary Results
Growth of the Rail System


Explaining the Pattern of Railway Expansion


The Growth of the Rail System in Rural Districts and Rural Out Migration


Patterns of Rural Out Migration and Rail Density in the 1880s. Within
rural registration districts, was there a geographic correspondence between
the rate of out migration and the availability of rail service, as indicated
by levels of rail density? A correspondence between
lower rates of out migration and higher rates of rail density would be consistent
with the hypothesis in question: that
the coming of rail service to rural districts served to retard out migration
by fostering new economic opportunities,
at least for a time. Such opportunities manifested themselves in various forms:
These opportunities brought some period of relief to rural settlements that
faced economic decline and depopulation.
The advantages for rural settlements were more or less temporary, for what
the railway gave with one hand, it eventually
took away with the other. By shrinking space and increasing long-distance trade,
rail transport intensified competition
throughout the country, which eventually undermined the livlihoods of small-scale,
local producers. The agrarian depression
of 1873 to 1896 was one example. Partly because of effecient rail transport
in the United States, arrival in England of
cheaper grain from the U.S. pulled the rug from under many a wheat farmer, forcing
reductions in output and in labor
costs as well as other economies.
The search for a possible geographic correspondance between greater access
to rail service and lower rates of
out migration can begin by inspecting the following map. With two exceptions,
all the rural districts lost population in
the 1880s, with the highest rates concentrated in central Wales, in Somerset
(SW), and in the peripheries of northern
Yorkshire and East Anglia. The visual comparison with the pattern of rail density
in the rural districts shows that there a
correspondance in some cases where lower rates of out migration seem to go hand
in hand with the higher levels of
rail density. On the other hand, the highest rates of out migration show no
regular correspendance with the lowest
levels of rail density, as in central Wales and on the edges of East Anglia.
In short, the visual comparison is suggestive
but inconclusive.

A second attempt using visual comparison shows in blue those rural districts
that recorded lower than the median
level of out migration during the 1880s; those in gray sustained higher rates.
Here the rail lines open in 1876 are
overlaid on the rural districts as another way of searching for corresponding
geographic patterns. Once again the
results seem inconclusive.

Although inconclusive, the visual comparisons prompt useful reflections. First
and foremost, the problem
in play is much more complex than the simple relationship between two variables.
The varying rates of rural out migration
resulted from many causes. All were geographic related; some prominently so,
such as: the proximity to localities with greater
economic opportunities, or the regional distribution of unemployment and economic
distress--the "pull" and "push" of economic
factors on decisions to move.
Lacking the data for a complex analysis, the exploration of two variables across
geographic space is nonethless worthwhile. It
shows that the retarding effect on rural out migration of rail service in the
countryside is plausible and deserves further
further investigation. Second, it illustrates the limits of simple visualization
as a means for identifying spatial relationships. Other,
more sophisticated methods are needed.
Chart Junk
Indicator: rural districts in the lower quartile of rail line density, kilometers of rail per square kilometer
Indicator: rural districts in the upper three quartiles of rail line density.
Grouping: LQRAIL76 (rail2.sta)
Group 1: RRLow : Rural Low Qrt Rail Den
Group 2: RRHigh : Rural 75% Rail Den
|
|
Mean |
Mean |
|
|
|
Valid N |
Valid N |
|
Variable |
RRLow |
RRHigh |
t-value |
df |
p |
RRLow |
RRHigh |
|
DEN7151 |
-1.667 |
.0903 |
-2.55 |
213 |
.011475 |
60 |
155 |
Grouping: LQRAIL14: lower quartile for low rail density rural
Group 1: RRLow : Rural Low Qtr Rail Den
Group 2: RRHigh : Rural 75% Rail Den
|
|
Mean |
Mean |
|
|
|
Valid N |
Valid N |
|
Variable |
RRLow |
RRHigh |
t-value |
df |
p |
RRLow |
RRHigh |
|
DEN1481 |
-4.475 |
-1.871 |
-2.50 |
231 |
.0123 |
71 |
162 |
Further Research