Electricity engineers and the happening of behaviour: lessons from a real-scale experiment

ABSTRACT

This paper addresses the question of the constitution of behaviour and its effects on the observers and on those observed, in the case of a ‘direct load-control’ experiment in the electricity sector in France. Typical of the rise of the behavioural approach to understanding demand in the electricity sector, that experiment lasted seven years and involved several hundred households. It aimed at sorting out human behaviours from that of heating systems and of buildings in order to prepare for ‘demand-response’ market designs. The paper proposes to consider behaviour as happening. Speaking of the happening of behaviour takes inspiration from performance art and serves to emphasize two crucial elements. First, it indicates that behaviour is more the product of a staged performance of those observed, than of a classical scientific instrumentation of an external reality. Second, it refers to the element of surprise that such a performed behaviour provokes on the part of observers. The surprise belongs to a quite different world from that of the machine-like action and well-delineated unit of the behaviour. Speaking of the happening of behaviour is a way to hold together two contradictory takes on life: that of the richness of the singular experience and that of the powerfulness of the generalized abstraction.

KEYWORDS:

• Abstraction
• experimental subject
• behaviour-specific action
• smart grid
• direct-load control

Acknowledgements

The content of this paper does not reflect the official opinion of EDF. Responsibility for the information and views expressed in the article lies entirely with the author. I warmly thank Pierre Scolan of EDF R&D, Olivier Thiery of Olivier Thiery Conseil and Régis Suteau and his team at Ipsos who contributed to sociological fieldwork on the PeakHeating Programme. I also want to specially thank my colleagues at EDF R&D with whom I worked on this programme, and especially Philippe Charpentier and Isabelle Debost who shared their knowledge with me and were interested in mine. This paper has been unusually long in the making – I more than gratefully thank those who have given feedback and advices on various stages of this paper – Celia Lury, Noortje Marres, two anonymous reviewers, Thomas Berker, Elizabeth Shove, Gordon Walker, Peter Karnoe and of course and foremost: Fabian Muniesa.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes on contributor

Catherine Grandclément is a researcher at EDF R&D. Her research combines elements of science and technology studies and economic sociology to examine the constitution of people as consumers in the energy sector.

ORCID

Catherine Grandclément http://orcid.org/0000-0003-0287-4425

Notes

1 I choose not to use the corporate name of the company in order not to ‘personalize’ it and the stories recounted here. Elements of the organization that feature in my empirical material as points of origins of decisions and actions aren't isoperimetric with the company: they don't represent the totality of what the company is, they are only a part of it.

2 ‘End-users’ is the most neutral term I have found in lieu of the vernacular words of consumers and customers which have replaced that of the ‘subscriber’ in the post-liberalization era of competition in the electricity sector.

3 Part of the empirical material on which this paper relies was also exposed in (Escoffier et al. 2015).

4 The grid in Brittany is notably weak and is called in electric engineers’ parlance an ‘electric peninsula’ because the grid is not ‘interconnected’. The shape of the grid there is that of a tree and not of a web, meaning that the current has only one path to reach its destination making the congestion stake all the more crucial. An ‘electric peninsula’ presents much higher risks of not attaining the all-important goal of ensuring the constant flow of electricity, the always-here, always-on quality of electricity that users take for granted and that is the absolute standard of value of good management of the electricity system. Contrarily to a web-shaped grid, a tree-shaped grid is one-way only, meaning that any default effectively cuts electricity supply downstream of the line.

5 While securing the participation of householders in smart grid demonstration projects is usually difficult, it turned out that the PeakHeating Programme attracted more households than expected. Its local dimension in a place with a strong regional identity and the trustworthiness, public service and technological excellence and the image of the company contributed to this unusual achievement.