Abstract: Governing Gas: Energy, Security, Circulation

Thesis submitted December 2016. Abstract below.

Natural gas is a troublesome and ‘wayward’ material (Bridge, 2004; 396). Amongst other qualities, it is invisible, intangible, naturally odorless, highly inflammable, and constantly resistant to the forces that contain it. This thesis provides an account of how these qualities both introduce a series of insecurities to everyday social environments, and also make it a challenging material to govern. Specifically, I examine the way that security is performed around gas circulations in the UK’s transmission and distribution pipelines, and I describe how a range of specialised security practices have been developed according to the particular challenges that gas’s materiality presents.

In developing this account, I make two claims. First, I argue that performances of security cannot be adequately understood without attending to the specific qualities of the circulating elements around which it is practiced. Here I develop upon Dillon’s (1996) observation that security has tended to be treated as a noun that is independent of the elements that it is practiced in relation to. As a consequence, it has typically been framed as a broadly transferrable set of practices that can be more-or-less unproblematically applied to very different elements. I suggest that this abstraction has resulted in the further reduction of security into two broad practices: acts of circulatory filtration (in which risky elements are separated from flows of safe bodies, materials and things), and acts of circulatory maintenance (whereby security is performed by ensuring the continuity of particular circulations). It is my contention in this thesis that security scholars need to pay better attention to the ways in which the specific material qualities of circulating elements are generative of particular forms of securing practice. Indeed, by examining the way that security is performed around gas, I describe a series of practices that far exceed those described in accounts that present security as a matter of circulatory filtration or maintenance.

My second claim is that the spaces and scales at which security is analysed need to be expanded. I demonstrate how the critical security studies and energy security literatures have both tended to focus on security’s practice within particular nodes, at the exclusion of the performances of security (and forms of insecurity) that develop across the journeys of circulating elements; as they move between nodes. Indeed, I suggest that circulation has often been reduced in these accounts to thin, straight, and featureless lines that are largely inconsequential for performances of security. I seek to trouble this reduction, following gas as it travels through the UK gas transport infrastructures, tracing the various forms of (in)security that develop across these journeys.

As a consequence of these two claims, security takes quite a different form in this account to its various depictions in the existing security literatures. I describe it as consisting of a series of ontological projects that are enacted across the lengths and breadths of gas’s circulations, and through which the material reality of natural gas is constantly (re)organised in attempts to facilitate, ‘compensate for’, and ‘cancel out’ particular kinds of perceived potential phenomena (Foucault, 2007; 36). Significantly, these performances are shown to be structured, or ‘programmed’ (Latour, 1991), through the coming together of multiple interests that pertain to a variety of heterogeneous actors and manifold referent objects. Different interests are shown to come together across gas’s journeys, and to undergo ongoing processes of negotiation that result in a variety of security performances, through which different imperatives are pursued. As such, I suggest that gas becomes ‘modulated’ (Deleuze, 1992) – it is constantly transformed from moment to moment, across the full duration of its circulatory journeys.


Longform: A History of Gas Governance (I)

The Mine as Laboratory

“And oft a chilling damp or unctuous mist,
Loosed from the crumbling caverns, issues forth,
Stopping the springs of life
….To cure this ill
A philosophic art is used to drain
The foul imprison’d air, and in its place
Purer convey.”
Jago (Cited in Galloway 1882)


The earliest European accounts of governance techniques being systematically employed upon gaseous matter appear in 16th century texts on coal mining operations. These texts describe underground encounters with a variety of air-like substances, each of which displayed markedly different qualities and presented significant threats both to human life and to the continued extraction of coal resources. Coal mining was a burgeoning and highly lucrative industry during this period, and it was as a consequence of these threats that the first documented practices of systematically governing gaseous matter emerged.

Human encounters with gases in coal mines and attempts to classify them pre-date scientific understanding of the existence of multiple distinct gases. As Galloway (1882) notes, whilst industrial coal mining was practiced in the United Kingdom from as early as the 12th century1, it was only in the 16th century, when demand for coal was rapidly increasing and supplies of surface coal were nearing depletion, that collieries began to extend their works further underground and gases began to accumulate in perceivable and threatening volumes (due largely to decreases in ventilation as mine workings became deeper). Until this point, frequent and consistent human encounters with gases in somatically perceivable ways (through effects on bodies and on the gas’s surrounding environments), were rare. As such, the coal mine presents a unique site in human-gas history, in which distinctions were made between different kinds of gaseous atmosphere for the first time.

To put this in perspective, scientific research distinguishing individual gases and their respective properties did not begin until the 17th century, and the term ‘gas’ itself was not to come into common parlance until over a hundred years later. Instead, miners called these gases ‘damps’, a derivation of the German ‘dampf’, meaning ‘vapor’ (Freese 2003). Multiple forms of damp existing in mines were identified, each distinguished by the manner in which they were encountered and the perceivable qualities that they displayed.

Despite the qualitative nature of the observations of these gases, up until the beginning of the 18th century the classifications of damps employed by miners more accurately described these gaseous materials than the classifications employed by scientists. This was in no small part due to the deep coal mine possessing two specific geographical features that enabled unique interactions between humans and gases to take place.

The first of these features was the coal mine’s ability to effectively constrain atmospheric volumes. The deep mine is effectively a vessel; it is a site of containment in which gases can become trapped by layers of rock which prevent them from mixing with larger atmospheres. This enabled unfortunate miners to encounter gases in sufficient volumes to witness certain clearly-observable gaseous actions (explosions, asphyxiation, poisoning). As mines became deeper and ventilation became poorer, the volumes in which gases were able to accumulate became greater, and their effects became more pronounced. In this way, deep drift collieries achieved something that was to later hold back scientific discovery for several decades; they managed to contain gases in significant volumes within a vessel (albeit a very large vessel), enabling humans to witness the behavior of these gases under a range of conditions.

The second feature of the deep colliery was its ability to roughly isolate certain gases into broadly distinct chemical forms. The labyrinthine geography of the colliery with its variable depths and gradients served to effectively separate gaseous mixtures into discrete materials based upon their relative densities. Certain gases such as carbon dioxide (which is more dense than normal air) would sink to low parts of pit workings, often causing the miners working in these areas to suffocate, whereas other gases such as methane (which is less dense than air) would rise to ceilings and high points, often igniting when miner’s candles came into contact with it, or setting alight when met with sparks produced by worker’s tools hitting the coal face.

These distinct features enabled different kinds of gas to be encountered in close to chemically pure states and in sufficiently large volumes for particular properties of these different gases to be expressed in clearly observable (and often devastating) ways. Because of this, miners were able to categorize different kinds of gas based upon their observable effects, and were able to begin to form governance strategies for these gases based upon their knowledge of each gas’s specific characteristics.

The Different Types of Mine Gases

The four most commonly encountered forms of mine gas were firedamp, chokedamp, and afterdamp (Rosner & Markowitz 1987; Freese 2003). Each of these gases displayed markedly different properties and presented distinctly different problems for mining operations.


“…a terrible explosion occurred, making its way up the pits, destroying men, horses, and all in its passage. The noise was heard for three miles around, and the blast of fire from the shaft was as visible as a flash of lightening.”
(Description of the 1766 firedamp explosion at Lambton Colliery, Chester-le-Street – Fynes, 1873 p11)

Of all the gases encountered in mines, firedamp was the most destructive. Firedamp is what is now referred to as natural gas – a gaseous mixture consisting primarily of methane. In coal mines it would seep out of cracks and fissures in the coal face and would accumulate at the ceilings and high points of mine workings. It was invisible, typically odorless, and had little perceivable effect on the body². As such, it was very difficult for miners to somatically detect it prior to it’s ignition. This combustibility made it extremely visible however, and assisted in its classification. Confined in large volumes in the workings of mines, and brought into contact with the oxygen drawn from the earth’s surface, firedamp could cause sizable explosions, single incidents sometimes disabling complete mine systems and killing large numbers of workers.

Whilst miners were familiar with firedamp and its flammable properties as early as the 1500’s, the scale and frequency of firedamp related incidents increased throughout the 16th century as mining intensified and pits became deeper. The first recorded firedamp explosion was in Gateshead in 1621 (Verakis & Nagy 1987), but by 1681 explosions were commonplace in British collieries, and by the turn of the 18th century, major explosions resulting in large numbers of fatalities were being widely reported (Galloway 1882). This capacity to instantly (and without warning) extinguish large numbers of lives and destroy colliery infrastructures made firedamp the most significant threat to miners and mining operations during this period.


 “Suddenly his lamp went out as if extinguished by a soft breath and at the same moment Pat Reedy choked and lay quietly down beside him. Not water this time. Black damp.”
(Extract from ‘The Stars Look Down’ – Cronin, 1935)

Chokedamp, also known as ‘blackdamp’ or ‘stythe’ (and known today as carbon dioxide), formed through oxidization processes that occurred as a direct result of mining operations. These processes included the miners’ own respiration and the use of fire in mines, but most significant was the reaction of carbon trapped in the coal with oxygen drawn from the earth’s surface (Unwin 2007). Similarly to firedamp, this gas was invisible and odorless, but unlike firedamp, carbon dioxide was a far more potent asphyxiant³. When encountered in large volumes it could cause rapid suffocation and death, and because it was incombustible and would extinguish flames (such as those used by miners for light), miner’s ability to navigate the mine workings and evade the chokedamp’s suffocating atmosphere before succumbing to it was often severely impeded.

Being heavier than air, chokedamp would sink to low, poorly ventilated locations in mines and could accumulate in deadly concentrations. It was this capacity of chokedamp to displace oxygen that presented risks to miners; unlike the other gases referred to here, it was not so much the properties of carbon dioxide itself that were directly threatening to life (indeed, as Barbara Freese (2003: 182) writes, “Its hard to think that a gas as friendly as carbon dioxide can be a pollutant […] It isn’t noxious, or caustic, and it doesn’t damage lungs, poison ecosystems, or destroy vistas”), but it was instead the absence of oxygen that posed threats to life. This density also meant that chokedamp was one of the first gases to be identified by miners, for unlike firedamp which in shallower pits would simply rise up  and exit the workings via the main shaft, chokedamp would settle and displace the air in even relatively shallow workings.


“…he had not been working more than half-an-hour before his head was like to split; and, ultimately, he was carried out insensible, and lay in his bed three days.”
(Description of an encounter with afterdamp in Thornley colliery, Durham, 1844 – Fynes, 1873 p66)

Afterdamp, or ‘whitedamp’ as it was sometimes referred to, is known today as carbon monoxide. Whilst an exact scientific understanding of the process of its formation was unavailable to miners during the 16th century, the circumstances under which afterdamp formed were well known. Afterdamp was so-called because its effects were often observed following incidents where firedamp ignited, carbon monoxide forming as the result of the incomplete combustion of trapped methane (Rosner & Markowitz 1987). This gas could accumulate in significant volumes after an explosion and had perceivably different qualities to either firedamp or chokedamp, enabling it to be accurately categorized as a different gaseous entity. Whilst it was similar to the other gases in that it was invisible, odorless, and like chokedamp, could cause asphyxia, it was quite different in that it was poisonous and had enduring effects on the body. When a person in a mine successfully escaped an atmosphere of chokedamp, they experienced no persisting negative effects upon their health. But when sufficient quantities of afterdamp were inhaled, miners who had been exposed often subsequently died, or took considerable time to recover following extraction from the hazardous atmosphere. This is because when carbon monoxide is inhaled it is absorbed into the bloodstream more readily than oxygen and can remain in the body for extended periods of time. As a result of this preferential adoption, carbon monoxide displaces oxygen and reduces the amount of oxygen that critical body tissues can receive, ultimately causing asphyxiation (Penney 2008). Moreover, in addition to a number of associated bodily indicators that made it perceivably distinct from chokedamp, such as headaches, muscle weakness, nausea, dizziness, fainting fits, convulsions, and comas (Bour et al. 1967), afterdamp also occasionally visibly presented itself upon victim’s skin, colouring it a cherry pink. In these ways, afterdamp ‘spoke’ of its presence, enabling its discrete classification.

Further contextual factors

The deep coal mine had a number of further specificities that help to explain why the first forms of gas governance developed in this location. The first significant feature regarded the types of gas found in coal mines. As is described above, these gases displayed fearsome  properties that readily expressed themselves upon contact with bodies and with sources of ignition. These threats were exacerbated by the continued deepening of coal mines in response to increasing demand for coal both for domestic use and for export, which led to consistent reductions in ventilation and the accumulation of larger volumes of hazardous gases (Galloway 1882). Secondly, due partly to the total darkness experienced underground; sources of ignition were ubiquitous in mines. Until the mid-1700’s, open candle flames were the most common form of lighting used within collieries, and other spark-producing devices such as picks and shovels constituted the main tools used underground. These devices dramatically increased the frequency of firedamp incidents, which in turn greatly increased demand for strategies to govern this gas. Thirdly, the financial costs of gaseous incidents were high. Structural damage from fire-related incidents could be extremely expensive, and could put mining operations out of service for months. This incentivized investment into finding ways to govern gas and reduce the number of gas related incidents. Finally, the cost to human life was also extremely high. Due to the large work force required for such a labor-intensive industry, sizeable numbers of bodies became exposed to these gases. Such numbers were not considered significant simply for their intrinsic human value, but also because they had political consequences, for it is highly probable that without such large numbers of casualties being frequently reported in newspapers between the late 1600’s and the early 1800’s, the forms of governing practice that did eventually come into practice would not have emerged.


1 The first explicit evidence of coal being mined rather than collected from surface deposits comes from an approval of a grant for the construction of a Monk’s colliery near Blackness, signed by King William. Whilst this record contains no reference to a specific date, King William’s reign ended in 1214AD (Galloway, 1882).

2 Methane can cause asphyxiation, but the atmospheric ratio between methane to oxygen that is necessary for asphyxiation is significantly higher than its explosive limit. Gas reaching such volumes was therefore much more likely to ignite than cause suffocation.

3 See note 2

This post is part of a series. Next up – Early Forms of Gas Governance: Coal Mines and Damps

Short: The Frost on the Pipe

There are no fish in Chile’s Quintero Bay. According to local fishermen, the water in this part of the Pacific Ocean is extremely toxic, polluted with high concentrations of mercury and copper tailings. It also exhibits a dramatic and life-extinguishing thermocline, caused by industrial processes that deposit both heated and chilled seawater back into the ocean within only a few hundred metres of each other. The result has been the formation of a stretch of coastline that is exceptionally inhospitable to organic life, a phenomenon that constitutes just one part of a wider environmental crisis that is affecting the Puchuncavi area.

It is the role of natural gas in this ecological catastrophe that I am interested in here, and how we, as academics interested in the social lives of materials, might understand gas to ‘speak’ in and through this crisis. Quintero Bay is the site of a large reception terminal for liquefied natural gas (LNG), which receives LNG from container ships, stores it in holders, and converts it back into gas ready for overland pipeline transmission. It is in this latter process of conversion that the materialities of gas and seawater collide, resulting in the ecologically significant jettison of chilled seawater described above. The driver for this process lies in the considerable thermal energy that seawater contains, energy that is used for warming LNG to a temperature above -161°C; the point at which it begins to convert back into a gaseous state. However, as the water exerts its heat upon the LNG, so the gas simultaneously exerts itself upon the water, forcing the water to undergo a dramatic reduction in temperature. In what follows, I propose that this might be one way in which we can understand the gas in Quintero Bay to perceivably ‘talk’.

Gas is unlike many of the materials that have formed the focus of vital materialist enquiries. It is, for example, distinctly different from Jane Bennett’s1 black plastic glove, her discarded bottle top, or her dead rat. This is because, unlike these materials, gas refuses to present itself readily for human somatic or sensorial experience. During our visit to Quintero Bay, we did not see gas, nor did we smell, touch, or taste it. We were not moved, seduced into contemplation, motivated to write, or spurred into action by simply observing it glinting in the sunlight, by being alarmed by its odor, or by feeling it brush against our skin. Gas, from our position as humans, was completely invisible, intangible, and odorless2. It was immune to human sensorial perception; it exceeded our senses and refused to grant us a direct audience.

But gas still speaks. Whilst beyond human sensation, it is not a formation of primordial matter that belongs to a world entirely inaccessible to human experience. Gas exerts itself upon other things that are sensible by human bodies (miner’s candles, scientific apparatuses, ultrasonic meters), and it is in these indirect expressions that we might find one understanding of how gas makes itself audible. For, whilst the gas in Quintero Bay did not directly press upon our senses, it did exert itself upon other materials in ways that we were able to experience. Through the chill in the water, the absence of fish, the audible vibrations in the pipeline’s components, and the formation of frost on the outside of pipes, we can understand gas to have ‘spoken’ and to have made itself heard.

But these ‘once-removed’ presses upon human senses are just one way in which we might understand the gas to be vocal. Whilst there may of course be many kinds of gaseous expression that are beyond human ability to perceive (as vital materialists we are aware that we may only be privy to a fraction of a material’s utterances), there are further means by which we might understand the gas in Quintero Bay to have a wider perceivable vocabulary than simple expressions upon direct material neighbors. As we arrived in Quintero Bay, we rapidly became aware of gas’s presence without ever touching the sea, without hearing any vibrations, and without seeing any frost. Rather, gas made itself known through the presence of a vast complex of pipes; through gas company plaques on gates and fences; through guarded barriers to entry. It declared its existence through large yellow signs warning us of its dangers, through personnel in red jumpsuits and plastic hardhats, in the enormous ships delivering the gas to shore, and through the giant concrete gasholders that dominate this section of the bay. Gas spoke in the way that it gathered around itself a complex and contingent collection of heterogeneous parts. As Lorraine Daston puts it, this “capacity to call […] a society of things into existence is as much a part of a thing’s thingness, of its reverberations in the world, as its material properties like weight and chemical composition”3. In other words, the ability to rally a diverse assemblage of people, materials, technologies, ideas, and things into being is as much a material vocal expression as the visual allure of a glove, or the observable formation of frost on the outside of a pipe.

But what is it about gas that demands the formation of such a complex assemblage? Whilst gas undoubtedly has other influential qualities, one significant aspect of its communicativeness lies in its destructive vitality. Gas’s expressiveness through its capacities to expand, ignite and explode is a key motivator in the calling together of this diverse attentive community. People, pipes, gates and gasholders are drawn into a constant performance that is predicated upon the imperative of securely containing gas and preventing it from expressing its destructive capacities. This performance takes considerable effort, particularly in the face of the geological instability that characterizes this region of Chile. Indeed, during our visit I was struck by just how much work was undertaken to ensure that the pipes retain their integrity in the face of the ground’s unpredictable movements. Far from the subterranean being a space of protection (as in many infrastructural systems elsewhere in the world), the underground here in Puchuncavi is a site of vulnerability that threatens to tear steel apart. Pipes are consequently not buried; they are suspended in protective cradles above the ground, and the entire infrastructure is supported by vast arrays of shock absorbers. As such, what initially appeared to be a rigid and stable infrastructure revealed itself to be a flexible assemblage that speaks of the extreme efforts required to prevent gas from realizing its explosive capacities.

Rather than understanding this assemblage to be predicated upon the strategic silencing of gas through inhibiting its explosive agency however, we might instead understand gas to speak of itself within these formations. For through the signs, the restricted areas, and the arrays of shock absorbers, gas communicates its destructive potential. And in so doing, it speaks of another aspect of its role within Puchuncavi’s fraught environment; that in spite of the constant work done to contain it, its mere presence presents significant risks to the humans, animals, materials, and things that populate and constitute this troubled ecosystem.

It is interesting to consider why such a community has formed around this particular aspect of gas’s potential environmentally damaging vitality however, and not around its thermal impacts upon the maritime environment. Indeed, the question of how objects come to acquire certain ‘powers of engagement’ through which they encourage the formation of publics is currently a source of academic debate4. However, in this instance there seem to be certain forms of silencing that are taking place. For example, when pressed about the environmental impacts of gas within Quintero Bay, our guide (a spokesperson from the gas company), claimed there to be practically no negative ecological impacts of gas’s commercial exploitation. Indeed, by his account, the gas company in question was one of the area’s most environmentally sensitive organizations. Whilst in relative terms this might be the case (it seems hard to equate the discharge of cold seawater with the illicit dumping of radioactive materials), gas’s chilling exertions upon seawater received no mention, despite the claims of local residents and a distinct absence of fish. Admittedly, within this complex assemblage where copper, mercury, gas, water, and organic bodies collide, it may be difficult to attribute singular blame for the ecological crisis that currently grips Puchuncavi’s maritime environment. But gas coolly speaks of its role in this ecosystem, and its utterances have not been completely stifled. For, at the very least, it has drawn me, my computer, and my keyboard into conversation with it. It has forced me to contemplate it and it has compelled me to write. And in that, gas has spoken.


1 Jane Bennett (2010) “Vibrant Matter: A Political Ecology of Things” London: Duke University Press

2 The smell often associated with natural gas is the result of an artificial additive. Natural gas typically has no natural odor, and the gas in this part of the system is not odorized.

3 Lorraine Daston (2004) “The Glass Flowers” In: Daston, L. (Ed) “Things That Talk: Object Lessons from Art and Science” New York: Zone Books (p. 228)

4 Noortje Marres & Javier Lezaun (Eds). (2009) “Materials and Devices of the Public: An Introduction” [Special Issue] Economy & Society 40:4