Trust in Numbers

by Theodore M. Porter

This investigation of the overwhelming appeal of quantification in the modern world discusses the development of cultural meanings of objectivity over two centuries. How are we to account for the current prestige and power of quantitative methods? The usual answer is that quantification is seen as desirable in social and economic investigation as a result of its successes in the study of nature. Theodore Porter is not content with this. Why should the kind of success achieved in the study of stars, molecules, or cells be an attractive model for research on human societies? he asks. And, indeed, how should we understand the pervasiveness of quantification in the sciences of nature? In his view, we should look in the reverse comprehending the attractions of quantification in business, government, and social research will teach us something new about its role in psychology, physics, and medicine.


Drawing on a wide range of examples from the laboratory and from the worlds of accounting, insurance, cost-benefit analysis, and civil engineering, Porter shows that it is "exactly wrong" to interpret the drive for quantitative rigor as inherent somehow in the activity of science except where political and social pressures force compromise. Instead, quantification grows from attempts to develop a strategy of impersonality in response to pressures from outside. Objectivity derives its impetus from cultural contexts, quantification becoming most important where elites are weak, where private negotiation is suspect, and where trust is in short supply.

Genres: Science, Nonfiction, History, Philosophy, History Of Science, Economics, Social Science

328 pages, Paperback

First published February 17, 1995

Reviews

[Alex · Rating 4 out of 5]

'trust in numbers' asks why administrative and scientific organizations quantify, and begins by claiming that the standard story, namely that quantification and formal models are inferentially more powerful than verbal theories, is insufficient. porter does not claim that quantification doesn't work, and instead sidesteps this discussion entirely to flesh out the administrative side of the story. the first several chapters essentially discuss measurement: administrative difficulties due to lack of information, and lack of information due to lack of standardized measurement. for this portion of the book 'statistics' essentially means 'counts of things,' and porter discusses extensive pushback in accounting and medical communities against counting, in favor of richer, more nuanced understand dependent on expert judgement.

the claim is then that national administrations became more powerful, and this increased power and bureaucratic scope allowed nation-wide measurement and standardization, which porter describes as social technologies. throughout this argument porter hints that these social technologies are shaping reality in addition to measuring reality -- i would have like to see connections drawn between his argument the performativity hypothesis, for example.

after this we arrive at what seems to me to be the core of the argument: that quantification is not an epistemic tool, but rather a social one. quantification is supposed to produce objective information, and objective information is necessary for policy decisions, or really any decision where interested parties will try to steer towards a given outcome. this claim is supported by a comparative study of doctors, accountants and engineers in europe (mostly britain and france) and the united states -- in europe, the experts are largely insulated from powerful political adversaries, and institutions that house experts have social and political status that allows them to operate autonomously and without much oversight. in these settings, experts quantify, but do so very informally, and do not share their results publicly; they are trusted to be objective already. in the united states, experts do not have this political cover, and they deploy an extreme form of quantification to certify their disinterestedness. in a related study of academics, porter shows similar effects of disciplinary disunity or newness, and the consequences of this institutional frailty.

the argument is compelling, and i would love to see it applied to more modern phenomena: the decline of universities and expertise in general, and more particularly the replication crisis and modern pushes for reproducibility. for example, the book suggests that our current focus on reproducibility is not so much about improving scientific practice (where, for example, is the theory work on how to conduct science correctly that would accompany such an effort?) as it is about certifying objectivity by making scientific results reproducible by literally everyone. i would like to see this idea taken further.

i also find porter's len an amusing framework to investigate the replication crisis. porter discusses the rote use of statistic tests (against all statistical advice!) in psychology as a tool to grant credibility to psychological experimenters in the face of disciplinary disunity (i am reminded of brian caffo and roger peng's "p-values are just the tip of the iceberg" paper). anyway, we see the same thing happening now: "solutions" to the replication crisis like pre-registration, which do not actually accomplish good inference, merely transparency. there are hints here of a futile feedback cycle: psychologists adopting mechanical procedures for inference to bolster their credibility, mechanical procedures for inference being impossible, such that the resulting inferences are bad, and this negatively affecting the credibility of psychology.

all in all, fairly dry writing, but full of interesting information. absolutely essential in that it considers how inference practices are responses to social pressures, rather than choices based on epistemic virtue. more data analysis needs to consider the goals, influence, and power of parties interested in data analytic outcomes, and this is a promising start. i look forward to work that can discuss this sociological aspect of data analysis, as well as technical and epistemic matters simultaneously

[Rodrigo Medel · Rating 4 out of 5]

This is a classic book in history of science. By comparing the role of numbers and the acceptance of objectivity criteria in France, England and USA, the main thesis of the book is that, contrary to what is commonly thought, quantification in the different spheres of society did not arise from the science of nature and then passed to social activities. Rather, the arrow seems to be exactly the reverse. Accountability and control of expert and elitist groups by society seemed to have played a critical role, with statistics and model building being only secondarily accepted as a regular practice by scientists.

[Chunyang Ding · Rating 3 out of 5]

A very thorough but dry book. Porter does a very good job laying out his argument, that the usage of numbers in several disciplines (economics, sciences, accounting, engineering) originates from the growth of a field, and the inability for gatekeepers to otherwise trust the participants of that discipline. However, during the transitionary process, numbers are often used as weapons to demonstrate objectivity, even if the process in which those numbers are created (ie, the methods for accounting for unquantifiable properties, the instruments by which data is collected and created, the statistical methods used for analyzing that data) is tightly controlled and bent to whatever that field already considers to be their "Truth". More than anything, Porter's book seemed to me to be a warning against implicitly trusting numbers, or people trying to say "Just the facts", because there is *always* some kind of social implication to it.

However, the presentation of this research is incredibly dull. The entire book is presented very academically, where minor details are as important as the primary theses of the manuscript. The portion on French engineering was especially difficult for me to follow, and even the section on the American Corps of Engineers, a topic that I have found fascinating since reading Desert Cadillac, was made to be grueling with the sheer quantity of details included. Porter makes an excellent resource for those studying philosophy and history of science, but this isn't a great book for the casual reader.

[Judd Street · Rating 4 out of 5]

Apparently when I read this book in 2016, I gave it five stars with the pithy review below. None of that is wrong - this is a classic text and it does a nice job of making some points regarding standardization - but with an additional seven years of reading STS texts, I'm not sure this is actually the best encapsulation of these arguments. Or, at the very least, I found it a real slog to get through a second time when re-reading it for a course, and am not sure it's the most accessible or approachable articulation of these ideas.

A core project of the book, of course, is to justify why it is that decision-makers are so compelled to work in quantitative worlds. As he argues in the introduction,

The appeal of numbers is especially compelling to bureaucratic officials who lack the mandate of a popular election, or a divine right. Arbitrariness and bias are the most usual grounds upon which such officials are criticized. A decision made by the numbers (or by explicit rules of some other sort) has at least the appearance of being fair and impersonal. (p. 8)

I'm not persuaded by Porter's view that this is a trap for unelected officials in particular, given that politicians seem equally likely to use rhetorics of impartiality and quantitative reasoning to garner support. But, the point holds in general. These quantitative narratives also have a "compatibility [with] positivism with the pursuit of control over nature" (p. 19) which fits with broader societal logics.

Quantitative logics aren't all rhetorical, though. For instance, Porter refers to J. S. Hunter's work on the US National Bureau of Standards, pointing out the ways that mandates (in theory) constrain producers from choosing the most favourable ways of self-quantifying (p.28). They also only hold water because people accept them, such as the way that publics accept school grades or accounting practices as standard and real (p. 45). And, they significantly shape professional practice, as discussed later (p. 116).

This can be particularly problematic when there's disagreement on how to count. Porter explores this with a great example from medicine and ways that normalization needs to be based on disease rather than bed days, lest results be misleading (p. 83) - very a propos in our COVID world.

Porter perhaps deviates some from Scott, who makes similar arguments, because of Porter's focus on knowledge production (versus social control). "But the bureaucratic imposition of uniform standards and measures has been indispensable for the metamorphosis of local skills into generally valid scientific knowledge," argues Porter (p. 21). This is linked, of course, with the desire to achieve objectivity, something quantification promises to deliver on (p. 74)

Porter also links this, in some ways that vaguely rhyme with Collins & Evans, to the nature of expertise. Many of his examples illustrate ways that quantification is an antithesis of expertise, the latter relying on discernment versus "mechanical objectivity" (p. 91). Standardization also frequently seeks to erode individual discretion, for better or for worse, which is often opposed by those like actuaries (p. 111); "In place of precision they offered a profession," Scott recounts (p. 113). Some professions (actuaries) were more vulnerable than others (engineers) to this professional erosion because of social and political power (see p. 138, 142). Put more directly,

Quantification was never merely a set of tools. Making up numbers in deference to political necessity was unacceptable to these engineers. It compromised their status as a disinterested elite and violated standards of mathematical integrity that they took seriously. (p. 118)

Finally, Porter's exploration of the rise of cost benefit analysis as championed by professionals to increase social standing and in the rise of increasing social distrust (p. 189) is also fascinating (Chapter 7), and features a few great quotes (e.g., "Hammon was aware, though, that Adam and Eve felt temptation even before the economic serpent presented them with this apple," p. 187). It also aligns with some remarks on the rising of statistical analysis due to distrust in medicine as well (p. 208-209).

All told, yes this is a good book... but, damn, it's been a while since I've struggled so much to get through a text due to lack of engagement. Adjusting previous 5 star rating to an average of 5 + 3 as a result.

--Dec 24, 2016 Review--
Trust in Numbers joins Scott's "Seeing Like A State" as quintessential texts of STS. Numbering, ordering, and standardizing the world is ultimately part of making it, and Porter does an excellent job of touring various disciplines and histories in order to illustrate these points.

[Amy · Rating 4 out of 5]

I skipped over some of the sections on quantification in engineering, but otherwise, this was the book I had been looking for for a while.

There were a lot of good points: how quantification a tool that comes in when trust in a community is eroded, how data can be used as a tool of accountability (but always a bit antagonistically), how social nature and socialization of the science community give it ~neutral authority, how the professions that rely on statistics the most are the ones that otherwise have little credibility (psychology, economics), how professionals will resist quantification and standardization bc of a belief they "expert subject knowledge" (and how more prestigious professions have been able to resist it better, e.g. medicine v. education.)

Niche history, but good.

[Tin Alvarez · Rating 4 out of 5]

I removed a star for the tedious prose in dense chapters. Otherwise, I would've rated this a five if the writing were as consistently lucid and lively as the prefaces, because it is a wonderful historical treatise tracing the prestige of quantification in governing
knowledge and social life.

[Sam]

I read this book on David Sposito’s recommendation, who in turn deploys it in his crusade against the social sciences. David’s point is that the social sciences are obsessed with quantification as a semi-mystical language that gives them authority to ‘scientifically’ deal with plainly human questions. These human questions are so controversial that society is unwilling to trust them to anyone who lack scientific pretenses – at least not anymore. Scientific quantification, however, neuters social scientist and makes their arguments pedantic at best and statistical jiggery pokery at worst.
Surely Porter does support this point, though not as directly as I expected. The riff on the general social sciences and their statistical consolidation contra the atomizing sixties is mostly Sposito’s. But Porter does expound on this necessary and sufficient truth: that to measure anything it must be measurable, and this is a quality like any other, with a particular flavor and characteristic set of deformities.
And what’s more, Porter does convincingly establish the public’s role in pressuring so-called experts to quantify – especially a suspicious democratic public. For any group of experts relies on trust, which must come from somewhere, whether it is inherited in the marrow of society or proven in practical applications, and while numerology certainly deforms their art, and often they do protest it, when experts are under such pressure that they seek cover, numbers can turn their assertions into so many impersonal observations, easily communicated across time and space, that can pretend to public trust. But even this, Porter reminds us, requires a standardized system of measuring and observation across time and space (for now observations exist abstractly, without tools of observing), and in this way the world is deformed to serve numbers, and these observations always rely on arbitrary judgement calls (especially in the social sciences), but by making these before wading into further deliberations, the experts can check the most thorny issues at the door.
In my opinion, the most obviously appealing element of the history of sciences is that it makes a very certain truth feel contingent. In school our teachers present science as a set of deep and fundamental facts about the universe. The history of science gives you the pleasure of seeing these models in their full contingent form, not less true for it but suddenly particular. Certainly, in this joy there is a bit of that perverse gratification of corrupting something so pristine. But it is more than that, it is a reminder that science, and those who pretend to its almighty power, is/are not excused from the course of history.
A personal aside: nowhere is this clearer than the hypothesis, which in the scientific method is an idea of the world that seems to spring up fully formed from nowhere. For we must pre-register our hypotheses in order to test them against the facts, implying that the hypothesis does not come from the facts themselves. But where does it come from then? It is intimately linked with and dependent on the facts. It comes from theory, yes, but this only delays the question one step further, for theory must also come from facts. And unless it is a theory with clear application to a completely new set of data, which is rare, the theory and subsequently the hypothesis must be modified in order to apply to the data at hand – but again, this is impossible to do if you are to form the hypothesis while ignorant of the facts. When reading the history of science you see in the hypothesis, and in theory, the stubborn presence of history, of course it does not come from nowhere, it comes from ideas about the world, about science, about facts past and present, and not from some all-pervasive scientific either, form which we can pluck potential-truths to verify or discard against the facts.
Porter summons a tension which I think he does not fully resolve, and I am curious to think about more. Even among the numerologist there are different kinds. To be specific, there are those who try to speak in the language of math for its own sake, abstract modelers who say that even if the models are naive they can at least be useful for their internal consistency. There are also practical statisticians who apply statistical principles to count and measure their observations. These two are both bean counters, but they are not obviously aligned, and might never be, separated by the imposing gulf between theory and practice.
Both these camps of numerologists fit Porter’s argument about trust in different way. The mathematical theorists retreat to ground so abstract and impeccable that external observers can easily verify it. They must also intervene in all sorts of ways to make it have any baring on the world. The statisticians also generate publicly verifiable trustworthy knowledge, through their care in measurement and observation. Despite these similarities, however, I doubt if these camps map on to Porter’s argument so similarly. The public image of the abstract economist who deals only in abstracted rational agents is quite different from that of the statistically-minded political scientist who measures weather patterns to determine casual patters in voter turnout. Each have their pathological deformities in response to a suspicious public, but I do feel intuitively that they are meaningfully different. They are reminiscent of a perpetual tension in scientific reasoning between induction and deduction, and even the last few decades’ reconciliation has not fully remove this tension. I do not have here the energy or expertise to push on this door further, but I suspect that if I did it would yield.
For me this book creatively overlaps with the professionalization class I took junior year (one of the BIG three or four lasting classes from college) and Abbot’s The System of Professions, my own supplement. The perpetual theme is authority in a democratic society, which is a product of institutions, politics, culture, culture downstream of politics and vice versa, and so forth, as much as it seems – and to some extent it – a product of some objective truth about reality. Porter resonates with me because he clearly applies these ideas to numerology in the social sciences. David Sposito is ultimately correct to call on Porter to provide the theory, if not the substance, of his anti-social science crusade.
Coda 1, chapter-wise notes:
Preface: This book stems from Porter’s quest to demonstrate that neoclassical economics, far from its air of objectivity as the most scientific of the social sciences, is a mathematical game with little bearing on practice. The book turned out quite differently, of course, but maintains the spirit of dethroning what might seem an intuitive sense of the objectivity of numbers, not just in the social sciences but also in the natural ones (which not too long ago looked more like social sciences…). Chapters 3 and 5-7 are the most properly historical ones and the others lean into the sociology, theory, etc.
Intro: Disciplinary objectivity is when something is true because a consensus of experts agree it is. Mechanical objectivity is when something is true because anyone can follow a clear set of rules to obtain the same result. The book is centrally concerned with the tensions between the two in social and natural sciences, and how experts shift their objectivity claims in response to outside pressure. Mechanical objectivity can never be purely mechanical because it must confront the messy world, but this form of objectivity gives a premium to quantitative logic, which is at least more structured than other logics.
1: Tools make the world countable – even natural scientific measures like temperature, weight, and length, do not appear out of nowhere. Europeans developed them first and foremost as part of a bureaucratic project to count and control the world. Administrators and scientists must standardize measurements and measuring devices, all of which necessitate imposing structure.
2: Social statistics are controversial (and if you trust their early history inherently political), and respond to official pressure by prioritizing standardization, so that their method is examinable and not simply based on trust, even at the cost of obvious discontinuity with reality.
3: In this chapter Porter discusses the tensions between abstract minded and practical minded economic measurers, for more of which see above. French administrators tried to apply economics, both abstract and practical, to practical matters with various levels of success and support.
4: Those who present morally salient arguments in a quantitative way can claim to be more impartial than those who rely on non-quantitative evidence, because the quantitatives can appeal to the clarity and durability of their method. These figures, however, obviously always have a political salience and their creators must make irreducible politically salient judgements.
5: Professionals, in this case accountants, long relied on a sense of gentlemanly trust to support their judgments. It was government investigation, an outside force, that pressured them to appeal to quantitative rigor. But those accountants Porter discusses mostly had the cohesion and strength to resist the outside pressure.
6: French planners and engineers could apply economic quantification in varied and often contradictory ways, but their status as an elite class imbued with public trust gave them the leeway to resolve these disputes internally and mostly use quantification only as they saw fit, without over standardization.
7: Many American public works agencies created standards of cost-benefit analysis that most fit their mandate and agenda. The quantitative rigorous pretenses of cost benefit analysis, while not complete, allowed these agencies to claim their analysis was impartial. The Army Corps of engineers is the greatest example of this phenomenon, and used its standard of cost-benefit analysis to muscle out its competitors in constructing great western American water projects.
8: American political culture especially cultivates a standard of demonstrable public evidence, which has a Midas-like quantifying effect on everything American politics touches. This includes medicine with its clinical trials, and psychology with its standardized testing, both contingent forms of evidence contra judgment by disciplinary experts like doctors, teachers, and psychologists.
9: There is a popular idea of science and the scientific community as the model of objectivity. The history of sciences demonstrates that this is all much more contingent. Science involves, must involve, socialization into a set of norms. Different scientific communities differ in their norms and standards for each other – high-energy physicists are a small tight-knit group who all know each other and so papers are mostly post-hoc, with all the real science taking place in informal discussions. Invoking Khun, rules are less important when the dominant paradigm is secure, but more important when there is much squabbling among scientists who seek to displace the paradigm, or alternatively to defend it. All these phenomena that Porter has been discussing, attendant to outside pressures, shape the discourse within different scientific communities. Science may provide a model of democratic community, as some of its glossy-eyed admirers claimed, but it is “not the stable, organic Gemeinschaft, but the impersonal and suspicious Gesellschaft, requiring a form of knowledge that, in important ways, is genuinely public in character” (231).
Coda 2, choice quotes:
“Mathematical and quantitative reasoning are especially valued under [the circumstances of mechanical objectivity]. They provide no panacea. Mapping the mathematics onto the world is always difficult and problematical. Critics of quantification in the natural sciences as well as in social and humanistic fields have often felt that reliance on numbers simply evades the deep and important issues. Even where this is so, an objective method may be esteemed more highly than a profound one. Any domain of quantified knowledge, like any domain of experimental knowledge, is in a sense artificial. But reality is constructed from artifice. By now a vast array of quantitative methods is available to scientists, scholars, managers, and bureaucrats. These have become extraordinarily flexible, so that almost any issue can be formulated in this language. Once put in place, they permit reasoning to become more uniform, and in this sense more rigorous. Even at their weakest point—the contact between numbers and the world—methods of measurement and counting are often either highly rule-based or officially sanctioned. Rival measures are thereby placed at a great disadvantage” (5-6).
“It is important to note that the form of knowledge resulting from this relatively rigid quantitative protocol is decidedly public in character. It embodies, and responds to, a political culture requiring that as much as possible should be brought into the open” (97).
“The opposition to the give and take of politics is shared by technocracy and practical quantification. But the reference to ‘a solution’ bespeaks the emphasis on impersonality of militant quantifiers. Technocrats in the French tradition have insisted that a cultivated judgement is required to solve social problems, and would be hard put to explain why different experts should not on occasion reach somewhat different decisions” (146).
“Though tools like [cost-benefit analysis] can scarcely provide more than guide to analysis and a language of debate, there has been strong pressure to make them into something more. The ideal of mechanical objectivity has by now been internalized by many practitioners of the method, who would like to see decisions made according to ‘a routine that, once set in motion, by appropriate value judgements on the part of those politically responsible and accountable, would—like the universe of the deists—run its course without further interference from the top.’ This, the ideal of economists, originated as a form of political and bureaucratic culture. That culture has helped to shape other sciences as well” (189).
“[Clinicians] are particularly bad at probability calculations, as the new studies of judgement under uncertainty have shown. It is not clear why professionals with graduate degrees including training in statistics should have so much difficulty solving elementary Bayesian problems. But it is no mystery why such problems do not succumb to the abilities of the ‘intuitive statistician’ once thought to be within us all. Apart from a few games of chance—and even these are arguable—no human ever confronted a stable, quantified probability value, or even the data to construct one, before the seventeenth century. Probabilities are in every case artifacts, created (but not arbitrarily) by instruments and by well-disciplined human labor. By now, an economist, doctor, or psychologist who cannot comprehend statistical argument involving variances and probability values will work less effectively on that account. This is not because the world is inherently statistical. It is because quantifiers have made it statistical, the better to manage it” (213).
[This one is just quite funny]: “The notion of scientific community has become a commonplace. Partly this is a matter of emptying of content from the concept of community: we find in everyday journalistic usage now such locutions as the ‘business community’ and the ‘black community.’ More inspired voices have spoken of the ‘intelligence community’ (spies) and the vaguely oxymoronic ‘international community’” (217-218).
Coda 3: I wonder how prediction markets fit into all this...

[Lorin Hochstein · Rating 5 out of 5]

There are two general approaches to decision-making. One way is to make a judgment call. Informally, you could call this “trusting your gut”. Formally, you could describe this as a subjective, implicit process. The other way is to use an explicit approach that relies on objective, quantitative data, for example, doing a return-on-investment (ROI) calculation on a proposed project to decide whether to undertake the project. We use the term rigorous to describe these type of approaches, and we generally regard them as superior.

Here, Porter argues that quantitative, rigorous decision-making in a field is not a sign of its maturity, but rather its political weakness. In fields where technical professionals enjoy a significant amount of trust, these professionals do decision-making using personal judgment. While professionals will use quantitative data as input, their decisions are ultimately based on their own subjective impressions. (For example, see Julie Gainsburg’s notion of skeptical reverence in The Mathematical Disposition of Structural Engineers). In Porter’s account, we witnessed an increase of rigorous decision-making approaches in the twentieth century because of a lack of trust in certain professional fields, not because the quantitative approaches yielded better results.

It’s only in fields where the public does not grant deference to professionals that they are compelled to use explicit, objective processes to make the decisions. They are forced to show their work in a public way because they aren’t trusted. In some cases, a weak field adopts rigor to strengthen itself in the eyes of the public, such as experimental psychology’s adoption of experimental rigor (in particular, ESP research). Most of the case studies in the book come from areas where a field was compelled to adopt objective approaches because there was explicit political pressure and the field did not have sufficient power to resist.

In some cases, professionals did have the political clout to push back. An early chapter of the book discusses a problem that the British parliament wrestled with in the late nineteenth century: unreliable insurance companies that would happily collect premiums but then would eventually fail and would hence be unable to pay out when their customers submitted claims. A parliamentary committee formed and heard testimony from actuaries about how the government could determine whether an insurance company was sound. The experienced actuaries from reputable companies argued that it was not possible to define an objective procedure for assessing the a company. They insisted that “precision is not attainable through actuarial methods. A sound company depends on judgment and discretion.” They were concerned that a mechanical, rule-based approach wouldn’t work:

> Uniform rules of calculation, imposed by the state, might yield “uniform errors.” Charles Ansell, testifying before another select committee a decade earlier, argued similarly, then expressed his fear that the office of government actuary would fall to “some gentlemen of high mathematical talents, recently removed from one of our Universities, but without any experience whatever, though of great mathematical reputation.” This “would not qualify him in any way whatever for expressing a sound opinion on a practical point like that of the premiums in a life assurance.” —pp108-109

Porter tells a similar story about American accountants. To stave off having standardized rules imposed on them, the American Institute of Accountants defined standards for its members, but these were controversial. One accountant, Walter Wilcox, argued in 1941 that “Cost is not a simple fact, but is a very elusive concept… Like other aspects of accounting, costs give a false impression of accuracy.” Similarly, when it came to government-funded projects, the political pressure was simply too strong to defer to government civil engineers, such as the French civil engineers who had to help decide which rail projects should be funded, or the U.S. Army Corps of Engineers who had to help make similar decisions about waterway projects such as dams and reservoirs. In the U.S., they settled on a cost-benefit analysis process, where the return on investment had to exceed 1.0 in order to justify a project. But, unsurprisingly, there were conflicts over how benefits were quantified, as well as over how to classify costs. While the output may have been a number, and the process was ostensibly objective, because it needed to be, ultimately these numbers were negotiable and assessments changed as a function of political factors.

In education, teachers were opposed to standardized testing, but did not have the power to overcome it. On the other hands, doctors were able to retain the use of their personal judgment for diagnosing patients. However, the regulators had sufficient power that they were able to enforce the use of objective measures for evaluating drugs, and hence were able to oversee some aspect of medical practice.

This tug of war between rigorous, mechanical objectivity and élite professional autonomy continues to this day. Professionals say “This requires private knowledge; trust us”. Sometimes, the public says “We don’t trust you anymore. Make the knowledge public!”, and the professionals have no choice but to relent. On the subject of whether we are actually better off when we trade away judgment for rigor, Porter is skeptical. I agree.

[Alex · Rating 3 out of 5]

This was a good book, but it wasn't what I was looking for, so take my rating with a sizable kernel of salt. I was hoping that it would contain more discussion of standardized testing and teacher accountability, but I only got about three paragraphs of that toward the very end. Instead, this was largely a history of the bureaucratization of the Ecole Polytechnique and the U.S. Army Corps of Engineers. Porter's critique of objectivity is definitely worth the read, though you may find similar critiques in more concentrated form elsewhere.

[Konstantin Genin · Rating 4 out of 5]

If mechanical objectivity characterizes a science or discipline, it is because it is a weak discipline that doesn't have the power or prestige to avoid discipline or accountability from mistrustful external forces. So quantification is a consequence of a lack of prestige and a lack of trust. A reasonably interesting point, made at length, and a bit overstated. Porter gives us the feeling that it is a shame that we cannot simply trust experts and their judgement, although the flip side of that coin is not explored.

[Polly Callahan]

from New Yorker article https://www.newyorker.com/magazine/20...
which says that “How Data Happened: A History from the Age of Reason to the Age of Algorithms” (Norton), the Columbia professors Chris Wiggins and Matthew L. Jones ....initial chapters, drawing on earlier work like Theodore Porter’s “Trust in Numbers,” Sarah Igo’s “The Averaged American,” and Khalil Gibran Muhammad’s “The Condemnation of Blackness,”

[Heather Hoyt · Rating 4 out of 5]

I read portions of this for a class and then finished the rest on my own, though I confess that by the end of it, I didn't care very much. I started this book by thinking something like, "Wow, I'm trusting in something that is much more arbitrary than I thought." Numbers can be subjective, just like judgment is. For a moment, I felt like I was looking at the wizard behind the curtain, and but then I find myself not caring about it very deeply and I let the curtain slide back. I trust in numbers, and that seems fine.

[Romy · Rating 2 out of 5]

I had to read this for my college seminar, and while it was incredibly dry, it was very thorough and the concept is an interesting one to think about.

Can objectivity ever really exist?

This book looks at this question and many more like it from a scientific standpoint, and I think I would have enjoyed it infinitely more if it were from a philosophical or humanities standpoint.

I won't give it anything less than 3 stars even though I honestly hated reading it, because I recognize it's better suited for a scholar than an 18 year old, and that doesn't make it a bad book, I just didn't enjoy it.

Edit: I take back what I said, my mind keeps randomly wandering to how much I disliked this book so I'm docking a star :)

[Yisu Z. · Rating 4 out of 5]

观点非常棒；写作风格不太习惯（不断推进到新概念，但又不收回来——。实证的部分也感觉一般般，比较浅、散。其中任何一个实证章节都能单独写一本书。非常帮助开脑洞的一本书。

[Harry · Rating 3 out of 5]

I had pretty high expectations and this mostly delivered, last few chapters and the chapter on the history of cost-benefit analysis were particularly good.

[Travis · Rating 1 out of 5]

philosophy of science

[Edward Fenner · Rating 3 out of 5]

Interesting but not exactly a page-turner.

[Melissa]

The book's structure is a little problematic.