Coding riots: A note of self-criticism

In a recent article I tried to explain how the practice of rioting in Britain had changed over the nineteenth and early twentieth centuries. The empirical material came from archival work conducted as part of my PhD, in which I used a variety of primary sources (newspapers, Home Office disturbance reports and local police records) to compile a catalogue of riots from Manchester, Liverpool and Glasgow from 1800 to 1939.

As well as making historical arguments based on textual descriptions of riots, I also attempted to quantify my catalogue in various ways to produce measures of rioting over time (see below).

One of the challenges in producing these quantifications was how to determine the boundaries of my cases, i.e. how to determine what counted as a single ‘riot event’. Following John Bohstedt, I tried to reproduce the logic of the original reporting as much as possible. This meant that I counted each riot as a single event, if it was reported that way at the time. Therefore, if a riot was reported as consisting of one attack in the morning and a later retaliation in the afternoon, I counted the two incidents as part of one single event.

This is obviously an imprecise art, not least because we have to follow the categorisations made by the author of the source, rather than having direct access to the experience the rioters themselves. But as I have gone back through my sources for a new essay, I have also come to question one of my earlier coding decisions. When faced with a series of riots throughout a strike wave, I coded them as a single event. This is because they were normally reported as such: as a series of connected “outrages” which formed part of a (single) industrial dispute.

However, this has the effect of reducing the importance of industrial rioting in my catalogue and giving equal weight to events with far fewer participants - a serious problem for all protest event analysis. In hindsight, splitting them up would have been (at the very least) an equally defensible decision.

In terms of my arguments about the changing practice of rioting, changing the coding procedure in this way would have two main effects:

First, it would inflate the number of riots in the early nineteenth century (particularly in Manchester), thus making the decline in urban rioting more noticeable and making Glasgow even more of an outlier compared to my two other cities.

Second, it would exaggerate the change in the form of rioting over time. In my paper I argued that, over the nineteenth century, “riots went from being an autonomous tactic to one which was largely subordinated to other protest logics [noteably the strike and the demonstration]” and that “the way rioters chose their targets changed: instead of targeting individuals with whom rioters had concrete relationships, they started targeting people as tokens of some wider type”. In both cases, changing my coding procedure would exaggerate these trends over time by giving more weight to early nineteenth century industrial riots in which violence was used as an autonmous tactic and directly targeted at employers and managers.

For example, it would mean giving more numerical emphasis to riots like the ones reported in these remarkable accounts from 1812:

On perceiving [the mob]... coming down the road, I assembled the children and nurse in the parlour, and fastened the windows and doors; the gardener presently rushing into the room and conjured us to fly that moment, if we wished to save our lives... each snatching up a child, we escaped at the great gate just in time to avoid the rabble. We proceeded to Mrs Syke’s; but before we reached our destination we saw our cottage enveloped in flames. Every thing, I have since learnt was consumed by the fire and nothing left but the shell. The mob next proceeded to the factory, where they broke the windows, destroyed the looms and cut all the work which was in progress; and having finished this mischief, they repeated the three cheers which they gave on seeing the flames first burst from our dwelling...

It is now nine o’clock at night, and I learn the mob are more outrageous than ever at Edgeley. Fresh soldiers have been just sent there. Another troop of horse are expected to- night. (Extract from a letter published in The Times, Friday April 17, 1812)

and

It is one of the most painful duties of our province to observe and record the desperate excesses of those of our countrymen who have so far suffered the machinations of disaffection to triumph over their good sense and every principle of rectitude, to prompt them to acts which reduce them to comparison with uncultured barbarians — with the very Goths and Vandals of antiquity. Let our artisans enquire of the aged, what were the number fed by our different branches of manufacture, before our machinery had acquired such perfection (for in most districts it is yet within the recollection of many), and they will soon be satisfied that the accumulation of goods in our warehouses is to be ascribed to the present deranged state of commerce throughout Europe, and not the multiplying power of machinery...

Monday, (as noticed in our last) the weaving mill of Messrs. Burton’s, at Middleton, was attacked by an infatuated mob, who had assembled in immense numbers at a distance, from various quarters. They came nearly in one body in density up to the building, fired a pistol, and then suddenly threw a shower stones, which was so violent, that those who defended the place, began directly to fire over the heads of the rioters, in hopes they would desist... finding their expectations disappointed, and as [the] assault continued with desperate fury and loud cries of “break in and murder them,” they were obliged for the preservation of their lives, to fire upon the mob, till in consequence three men being killed and several wounded, two of whom afterwards died, they withdrew from, the factory.— They continue in the town destroying the furniture belonging Messrs. Burton’s servants, and threatening their lives, till a troop of Cavalry arrived... On Tuesday [a Mill] was surrounded by about fifty of the Militia, the mob... after ransacking and dividing the spoil, they set fire to the house, barn... [illegible] which consumed the whole, with the furnishings, fixtures &c. The damages is supposed to be upwards of £2000... The conduct of the Military has been highly praiseworthy. The number of the wounded is we understand 27, nine of whom very severely. (Manchester Mercury, Tuesday April 28, 1812)