The most recent issue of Environment & Society explores different practical, theoretical and conceptual problems associated with measuring and measurement across diverse environmental issues. In this blog we have reproduced the introduction to that collection, which is available on open access here. The arguments here should whet your appetite for more, and given some idea as to why this issue is so important.

The imperatives of measurement seem particularly prominent in today’s social environmental concerns. This is partly because of the problems different societies and international bodies have set themselves to solve. Whether these be the monitoring required to meet climate change targets or prevent biodiversity loss, the formulation of indicators to track progress to meet the targets set for the Sustainable Development Goals, concern about the return of high levels of inequality or crises in the validity of political polling, how we measure and monitor the world is increasingly on the agenda.

The prevalence of measurement is also partly due to the new possibilities and ways of knowing that are now being revealed through “big data” that track social media records, mobile phone use, and new abundances of machine-read satellite data. These data are themselves complicated constructs; they combine proxies, signals, indicators, and diverse interpretations and inferences. But this does not make them any less “factual”—after all, that is what fact has almost always constituted. The result is that many parts of the world look different now because we have new tools to see them with.

Measuring, monitoring, and counting have come to be close to the heart of research and academia. In Britain, academics’ research is evaluated periodically in national comparative procedures (now called the Research Excellence Framework), and their teaching is about to be also (the Teaching Excellence Framework). Metrics are increasingly likely to dominate these scorings (Wilsden et al. 2016). And independently of any official surveillance, many academics have a fairly good idea as to our citation scores and h-index. In some disciplines these will even be printed on the back of business cards. The fact that these measures can become such driving goals is a good measure (indicator if you prefer) of the social power of indicators.

Getting measurement right is important because it can become a means of spotting patterns and of holding powers to account. “Without measurement and standards, organizational agents operate under the tyranny of cronyism” (Power 2004: 774). But the means of seeing that measurement affords can also become ways of not seeing, of invisibilizing, of failing to recognize people, places and ways of knowing. As Paige West and colleagues (2006: 254) observed over a decade ago, with reference to a then newly comprehensive database of protected areas, new machines for measurement can become “a way of seeing the world with blindspots and blurred vision not easily perceived by its operators, but these blindspots become darker and fuzzier as the machine becomes better.” Michael Power (ibid) argues that when measurement becomes a means of comparing performance in organizations, “new secret organisational worlds” form that create and exploit invisibility in order to appear better in performance metrics.

More than that, we must understand measurement not just as a way of seeing but also as a creative and organizing project that co-produces the very world it aspires to describe. Theodore Porter (1994) is quite clear on this. Understanding measurement (this discipline is called metrology) is not just about how measurement describes society and nature. We have to understand how it reconfigures relations between and with them, and how it imposes and erases meaning. Understanding this is what metrology is all about. As Mark Cooper (2015: 1789) puts it: “[i]t invites us to question the social, political, and scientific conditions under which agreements about measurement and commensuration do or do not occur, and the consequences or effects of particular metrological systems.” Careful historical approaches are required to reconstruct how measurement came to be and what effects it has had (cf. Jacob 2001).

Measurement was integral to colonial and imperial projects. The quest to govern and the quest to measure and classify were intimately related (Freidberg 2007). Histories of the failures and dramas of colonial rule are replete with failed and violent attempts to create new categories of people or nature and ways for these newly identified categories to behave. Similarly, new forms of auditing and rule by standards are likened to new forms of neo-colonial endeavor.

But the transformations of measurement can also be experienced in subtler ways, and closer to home. Porter gives a useful example of the case of the US Corps of Engineers, which was instructed to ensure that the benefits of flood control schemes exceeded the costs. This sounds innocuous enough, except that the means by which costs and benefits were to be calculated was not explicated. Determining the costs and benefits of different proposed schemes then became the stuff of political battles between corporate interests and different branches of the state. More revealing still, when attempts to reconcile competing visions began, quantifying actual costs and benefits proved to be “startlingly elusive” (Porter 1994: 395). Even the agreed format for doing so that eventually came out was still added to and interpreted in idiosyncratic ways. Measurement could not contain the difference it was meant to arbitrate between.

The case illustrates a broader point. Measurement is not about describing the world. It requires building an apparatus that makes that description possible. As Porter (1994: 404) puts it:

‘To quantify a quality is not merely to solve an intellectual problem. It is to create what Latour calls a center of calculation, surrounded by a network of allies . . . The quantification of qualities is as much an administrative accomplishment as an intellectual one. And no matter what the skeptics may say, many social qualities have already been successfully quantified, in a variety of ways. Those who seek to do it differently, or to spread the net of quantified qualities still wider, need to consider not only epistemological questions but also moral and political ones. There is strength in numbers, and anyone who proposes to wield them more effectively must ask not only about their validity but also about how the world might be changed by adopting new forms of quantification.’

This way of thinking is clearly vital if we are to understand how markets govern, as Cooper (2015) has argued.

But measurement is not just conjured up through marshaling administrative support; it changes the world because, once established, it also entails altering practices to fit with, or respond to, the measurement. Porter gives the example of the US Forest Service, which had been instructed to cut no more trees than were being renewed through regrowth. Regrowth, however, can be boosted if fertilizers and new varieties of fast-growing tree are introduced—which means that more trees and larger trees can still be harvested (Porter 1994: 401). Indeed, the whole point of incentives offered by diverse forms of neoliberal conservation and environmental policy in the form of payment for environmental services is precisely to change the world by changing the behavior of (rational, profit maximizing) individuals. And some of the individuals can even become rational profit maximizers in the process.

The travesties and distortions required to see the world like a market or state can suggest that managing by counting is inherently flawed. But Power (Power 2004) argues that things are more complicated than that. The social processes that constitute measurement, and how people respond to measurement, create cycles of crisis and reform. He distinguishes between first- and second-order measurement, with the former establishing the classifications that make counting possible and the second combining counts into indices and composite indicators, which can forget the social origins and circumstances that produce them. This can result in multitudes of inappropriate numbers and strange uses of them. And this helps to drive the “cycles of reform” that characterizes measurement. According to Power (Power 2004: 778), the social and political responses to different sorts of (flawed numbers) are not about trust or distrust:

‘Dreams of measurement for control purposes are articulated; these are shown to be defective and/or leading to adverse unintended consequences; new measures and refinements are proposed. Any so-called trust in numbers is tempered by the general cultural acceptance of numbers in all aspects of modern society. Equally, specific episodes of distrust and critique lead to the reconstitution and revision of performance metrics, rather than their abandonment.’

Understanding measurement, therefore, is required to understand the societies demanding that measurement, and produced by it. That was the imperative behind this collection. The call for this volume posed a number of challenges. It asked authors to take on a variety of questions, including: How do we approach, measure, quantify, and qualify socio-environmental issues and phenomena? How does what we measure or the way we measure it affect what we know and how we act? How do particular types of, or approaches to, measurement become embedded in epistemic communities and with what consequences? What new things can we learn with new forms and techniques of measurement?

The response was rich, and the eight articles published here capture some of the diversity of interests and approaches. You can read more about what the authors argued and their different approaches in the rest of this open access introduction (start on page 3). In summary there are a number of abiding themes in this collection that are anticipated, and elucidated, in the metrological literature. The first is that measurement is an administrative achievement. The construction of indicators for market governance, community resilience, proxies of performance, heat, fish stocks, and much more requires complicated governing apparatuses and networks of state, private, and civil society interests. A second is that this has unexpected consequences –which perhaps itself should hardly be unexpected at all. The final point is that counting life, whether in societies or environments, is clearly problematic, if not also violent. It does not solve problems but creates a host of new ones.

But it does not follow from the last point that not counting becomes the solution to the problem. It may be a worse fate still not to appear on any register. The issue is not whether to appear, but on whose list, how, for what purpose, and in what circumstances. We cannot avoid measurement. As social beings, we count, calibrate, classify, and measure. How we see ourselves and others, and how we cohere, depends on such processes. It is difficult to imagine societies that do not do any of that. And it is precisely this inevitability and ubiquity of measurement that makes it so necessary to contest it more vigorously.

References

Cooper, Mark. H. 2015. “Measure for Measure? Commensuration, Commodification, and Metrology in Emissions Markets and Beyond.” Environment and Planning A 47: 1787–1804.

Freidberg, Susanne. 2007. “Supermarkets and Imperial Knowledge.” Cultural Geographies 14 (3): 321–342.

Jacob, Margaret C. 2001. “Factoring Mary Poovey’s A History of the Modern Fact.” History and Theory 40 (2): 280–289.

Porter, Theodore M. 1994. “Making Things Quantitative.” Science in Context 7 (3): 389–407.

Power, Michael. 2004. “Counting, Control and Calculation: Reflections on Measuring and Management.” Human Relations 57 (6): 765–783.

West, Paige, James Igoe, and Dan Brockington. 2006. “Parks and Peoples: The Social Impact of Protected Areas.” Annual Review of Anthropology 35: 251–277.

Wilsdon, James. 2015. The Metric Tide. The Independent Review of the Role of Metrics in Research Assessment and Management. London, Sage.