City Journal Spring 2014

Current Issue:

Spring 2014
Table of Contents
Tablet Editions
Click to visit City Journal California

Readers’ Comments

Jim Manzi
What Social Science Does—and Doesn’t—Know « Back to Story

View Comments (74)

Add New Comment:

To send your message, please enter the words you see in the distorted image below, in order and separated by a space, and click "Submit." If you cannot read the words below, please click here to receive a new challenge.

Comments will appear online. Please do not submit comments containing advertising or obscene language. Comments containing certain content, such as URLs, may not appear online until they have been reviewed by a moderator.

Showing 74 Comment(s) Subscribe by RSS
" it was clear that we wouldn’t know which economists were right even after the fact"

There are certain fundamentals in the sphere of economics that are indeed predictive, and failure of the stimulus program as it was proposed, and further how it was politically implemented was an unmitigated failure. There were better economic choices that didn't have political luster.
Social science--the real, non-PC kind--knows a lot. But too few Americans, including conservatives, bother with it. Or know how to scrutinize it.

(By the way, "gay" is parent-caused, preventable, and treatable.
Thanks for an elaborate article.
Protease [url=]kamagra pills[/url] inhibitors control the metabolism of sildenafil, effectively multiplying the plasma levels of sildenafil, increasing the amount and mercilessness of side effects. It is recommended that patients using protease inhibitors limit their attention of sildenafil to no more than merged 25 mg weight every 48 hours [url=]viagra pharmacy[/url].
obsuzuoiek pgvbrkvmpn viagra buy
djfhrztqqu uxubkpsjpu cheap kamagra online
Protease [url=]kamagra online buy[/url] inhibitors interfere with the metabolism of sildenafil, effectively multiplying the plasma levels of sildenafil, increasing the situation and cold-bloodedness of side effects. It is recommended that patients using protease inhibitors limit their select expediency of of sildenafil to no more than a bride 25 mg amount every 48 hours [url=]viagra pills[/url].
viagra cheap
cheap brand kamagra
As a social scientist, I suspect that Manzi's core argument is exceedingly important and probably largely accurate. However, care must be taken in lumping together all social sciences in one fell swoop.

Let us not forget that although James Lind conducted an experiment with multiple treatment groups in 1747, it wasn't until the second half of the 20th century that the modern randomized controlled clinical trial was truly codified and regulated.

Further, Manzi does not seem to have a basic understanding of the challenges of experimental vs. observational research. Randomized controlled experiments, by their very nature, try to provide a high degree of internal validity (i.e. the ability to make causal inferences). However, due to cost and logistics, external validity (i.e. the ability to generalize findings to other populations) is often compromised. This isn't new information. In the social sciences, if you truly want high internal and external validity, you're going to have to recruit a large multistage clustered sample, assign participants randomly, and remain ever vigilant against threats to validity.

The social sciences can do better on self-appraisal and evidence-based practice. But to imply that experiments represent a novel savior for social sciences is inaccurate.
seroquel for limbic dysfunction. buy cheap seroquel online. seroquel periactin erectile dysfunction. buy cheap periactin online. buy doxycycline hyclateThis is an unusual occurrence in studies of erectile dysfunction, so it casts some doubt on the study results. in diabetic men with erectile dysfunction who had not responded well
then human proof-read reconstruction of the 1992 edition. The content The electronic conversion of this document was funded in part by the New ZealandWouldn't be half as bad. It wouldn't be half as sad. If each and everybody in it had, yeah Wouldn't be half as bad. It wouldn't be half as sad. If each and everybody in it had. Just a
noun food for feline kindeny problems. food for feline kindeny problems. ran food mold identification. food mold red exam erection nurse doc. exam erection nurse doc. interest wonder woman sex cartoonPretreatment and 3-year prospective patient reported outcomes of urinary, bowel problems, and sexual function after radical prostectomy and external beam radiotherapy in patients with localized prostate cancer. - ASCO - The American Society of
metal erection woodbridge ontario
can't get a hard erection
erection after circumcision surgery

viagra for muscle building viagra blind cheap viagra pills cheap 25mg viagra
taztia impotence emotion behind impotence⠤
best buy online buy discount viagra,how to get without buy viagra pharmacy,cheape online snorting buy cheap generic,generic no prescription cod purchase online ems delivery buy generic drugs,order cod next day delivery not expensive prescriptions online drugstore,generic drugstore,ordering online without prescription where can i buy online online pharmacy,overnight delivery of in US no prescription needed generic rx drug, watson saturday delivery cash on delivery no rx
Engaging and cogent explanation of a complex subject.

[url=]tramadol shipped overnight no prescription[/url]
[url=]Buy Prednisone no prescription[/url]
[url=]Trazodone without prescription[/url]
[url=]Ultram Overnight No Prescription[/url]
[url=]Xanax cod next day delivery[/url]
[url=]No prescription generic viagra[/url]
[url=]Xanax without prescription canadian[/url]
[url=]Buy Soma with no prescription[/url]
[url=]Buy Viagra without prescription[/url]
[url=]buy cheap Acyclovir overnight[/url]
[url=]Order Soma cod overnight[/url]
[url=]soma no prescription next day delivery[/url]
[url=]Buy Clomid with no prescription[/url]
bukmachermilenium April 04, 2011 at 5:08 AM
Good work, it is pleasure to read your articles. Waiting for more
platne typy bukmacherskie
Victoria Secret Gift Card February 07, 2011 at 9:33 AM
Hi : )
You shop on the web or in-store? which usually would you like? really wondering lol.. i prefer in-store since i don't really like waiting it to come!
History provides so much of data which the social scientists need to take into account before predicting.
The problem for social scientists is that their predictions or solutions are constrained because they operate within the parameters of established ideology. Karl Marx's predictions about capitalism continue to be relevant because he moved away from the established ideology of his times and predicted on the basis of honest analysis. That the world has problems in accepting his predictions is again because of the ideologically imposed restrictions on free thinking. If social science ideas are allowed to fall as freely as an apple can fall from a tree with just one pull that is human welfare pulling it down, one would surely get good results from social scientists too.
I find it fascinating, and supportive of Jim Manzi's article, that so many of the commentators here are reacting to what they perceive as an attempt to either break their rice bowls or alter their fixed ideological/theological world views.

In a rational world they'd dump whatever science wasn't working, back off, try to ignore their prejudices, and take an objective look at the problem. But that seems too much to ask for economists or social scientists who have dedicated their lives to a theory, instead of objective truth. To them it's a religion, not a science, and they are the faithful priests and followers of its prophets.
Dallas Morning News Article 9/5/2010-I have long held scattered thoughts and beliefs about that social science subject and I thank you for brilliantly putting it all together for me. I agree completely with that article. Bless you and the DMN for printing it.
Letter to the Monitor, American Psychhological Association

Copy Sent to Jim Manzi [Please Forward]

Sent: 9/5/2010 11:44:58 A.M. Central Daylight Time
Subj: Deciding Who Belongs

Re: Deciding Who Belongs

Arizona's immigration law will likely harm police-community relations

and increase stereotyping and prejudice, psychologists say.

By Christopher Munsey

Monitor Staff

September 2010, Vol 41, No. 8

Unfortunately, Christopher Munsey's treatment of Psychology’s take on Arizona’s immigration law, an attempt to deal with the serious social problem of illegal immigration, does little to buttress respect for Psychology as a discipline devoted to the science of human behavior. The article seems to be entirely based upon his predisposed prejudices. In fact, I suspect his subtitle reflects his "politically correct" bias which fostered both his inspiration and his conclusions without any serious consideration given, quite ironically, to either reason or an impartial examination of the facts.

Here are the obvious issues he neglected to address:

1. Are the selective findings of a few social scientists a reliable foundation for making judgments about social policy? As Jim Manzi has recently written, “At the moment, it is certain that we do not have anything remotely approaching a scientific understanding of human society. And the methods of experimental social science are not close to providing one within the foreseeable future. Science may someday allow us to predict human behavior comprehensively and reliably. Until then, we need to keep stumbling forward with trial-and-error learning as best we can.”1 There wasn't anything one might call tentative in Munsey’s discussion.

2. Are the majority of American citizens and the growing number of state legislatures now in favor of laws like the one passed in Arizona simply seeking empowerment “to use their stereotypes and prejudices to characterize someone as undocumented, and feel justified in performing acts of discrimination by calling the police”? If so, what has occasioned the alarming wave of American bigotry?

3. While the foundation for any scientifically warranted claim that the Arizona law would give rise to the unintended consequence of indiscriminate racial profiling undermining our cherished tenet of equal justice for all seems especially suspect, where is there any consideration given to the unintended consequences of existing social policies and enforcement procedures which, having been so grossly exploited by those entering this nation illegally, arguably undermine the fabric and economic well being of our society?

I wonder also how the editors failed to notice such flagrant omissions.

Paul I. Munves, Ph.D.
5741 Prestwick Lane
Dallas, Texas 75252
Tel. 972-380-2396

1 What Social Science Does—and Doesn’t—Know Our scientific ignorance of the human condition remains profound. Jim Manzi. City Journal, Summer 2010, Vol. 20, No. 3, The Manhattan Institute. []
My one question is long-winded but I think critical: what percentage of the social programs and policy innovations alluded to tackled their problem areas with overwhelming force holistically directed at all or most of the expected core causal factors to the problem?

Those expected core factors could be roughly generated at the onset by a committee of experts. And by overwhelming force, I mean far, far more than a simple rehabilitation program for a criminal. The criminal’s friends, family, housing, job, health, etc. would also need to be addressed with vigor to impact recidivism. Obviously, this is often too costly with today’s technology, so I’d expect the answer to be zero or close to zero. With tomorrow’s technology, we may have a chance at that desired rigor and impact.

My point here is that history is a terrible guide for predicting the future of social sciences. Just as the ancient Greeks had a terrible grasp of physics, we still have a lot of work to do.
Dear Mr. Manzi,

You seem to be suggesting that it is not clear that lower tax rates provide a stimulus to investment, or that free competition does not reduce prices and raise quality; whereas, monopolistic conditions raise prices and reduce quality. It is hard to believe you truly believe that.

Yet people seem to understand. We witness a profound growth in free market policies here over the past couple of hundred years. I live in South Korea, and there is no confusion among the populace regarding their relative affluence and freedom vis-a-vis that of North Korea. You don't see people moving into North Korea.

We compare social policies on a broader scale. (And yes, the human mind can do that, while it may be that no computer ever has that capacity. This is how cultures wax and wane, not by accident, but by human design.) And, we look at them in relation to religion.

It merely takes a broader outlook to evaluate them, but people do it every day, and vote with their feet (emigrating to places of better perceived opportunity) or striving to instate similar policies of admired cultures at home (which is more difficult, and perhaps more admirable).

Love, Nathaniel
Urstoff, I agree with your aside that a theory without evidence is a fiction. Furthermore, this insightful article by Jim Manzi reminded me of an essay by media critic Neil Postman called "Social Science as Theology," where he described social science as a form of storytelling, only one that adopts the language and methods of science (models of closed systems, mathematical equations, statistics, formalized language, etc.) to assume science's epistemic authority and to convey intellectual rigor, during the telling of the author's story, that is, in the presentation of the social scientist's conclusion and theories. The article can be found at:
Like all science, the "social sciences" need a research base.
We badly need a research-based social science.
In reality, the social sciences have uncovered a huge amount of useful knowledge about humanity. The vast field of cognitive testing, for example, which has revolutionized American society over the last century, is a triumph of the social sciences.

What social scientists can't do is the same thing as physicists can't do: the impossible. Just as social scientists have failed to eliminate many social ills, physicists have failed to make possible many desirable technologies lovingly imagined in the science fiction novels of my youth.

Where's my teleporter? Where's my faster-than-light spaceship? Where's my anti-gravity device? Where's my time machine? C'mon, physicists, what's the matter with your methodologies?
Social "science" is anything but science. It is a category of psychology and sociology which themselves are subject to politics and culture and which are tools for social control.

Re3member when homosexuality was a disorder? Psychologists also had terms for the "disorder" of slaves who escaped and who wanted to flee captivity ("drapetomania").

Newer manifestations of this are curent and in our recent past. Remember the psychologists who studied alien abduction in the 80s and 90s? "Repressed Memory Syndrome"? Current ideology posits that only women can be victims of domestic violence and mothers who kill their children or husbands must be suffering from one or more "disorders" or "syndromes".

The Caucasian witch doctors continue to practice their primitive trade.
Without ever saying so, Dr. Manzi, nearly all the measures you mention are associated with generational poverty; where the criminal justice system and poverty of mind, body, and environment are intricately wound.

I say we invest sustained billions on universal pK-PhD for all comers from all lands. And that would mean for our poorest, school-zone housing that includes nominal rents, security, medical clinics, merchants... We would be forging new communities where education is central and heralded. Buy now. Nuisance abatement indeed.
On 'causal density', David Ricardo in a letter to Malthus in 1815 remarked, "If I am too theoretical, which I really believe is the case, - you I think are too practical. There are so many combinations, - so many operating causes in Political Economy, that there is a great danger in appealing to experience in favour of a particular doctrine, unless we are sure that all the causes of variation are seen and their effects duly estimated."
richard goldwater August 06, 2010 at 8:58 PM
more at
Social science operates as an analogy of physical science, using the same math and the same assumptions of how reality works. The social sciences labor at a disadvantage however because as the author admits, the basic model has not changed in 250 years -- the time of Isaac Newton. Physical science has changed much since Newton, but social scientists have not noticed, a failure of our educational system to be sure.
Social scientists (with the worst implications for economic science) are ignorant of the implications of the Second Law of Thermodynamics, which reduces Newton to a special case.
Newton's math describes the universe as a frictionless mechanism. Thermodynamics describes the universe as the consequences of an explosion, having built a world view from the study of steam engines. Thermodynamic literacy can bring social scientific thinking closer to the objectivity they seek.
More at
A further 'problem' is that the 'objects' (or should that be 'subjects') of social scientific research — namely people.

Unlike the social science, the 'objects' of social scientific analysis are able to speak back. And not only that, they are able to read the findings of social scientific research and alter their behvaiour in light of its finding.

The English sociologist Anthony Giddens noted this toward the end of his book The Constitution of Society, published in the early 1980s. Giddens noted that the where the natural sciences are involved in a subject-object relationship (scientist and rock, for example) the social sciences are involved in a subject-subject relationship (social scientist + average person).

In Giddens' terms, the social sciences are always involved in a double process of interpretation. A social scientist examines a social world that is always and already meaningful to those who inhabit it. In other words, that world has already been interpreted by social actors as part of their daily lives. Social scientists offer another layer of interpretations which, though being perhaps more systematic and revealing things that previously were hidden from view, remains yet another interpretation. Social actors can themselves read this research and alter their behaviour.

For example, social research on child development changes the way people rear their kids. Psychological research into people's emotional lives changes the way they comport themselves. Of course, not everyone does so, and it may take time, but the basic point is that the social sciences nevertheless alter the 'objects' that they study.

This includes experiments. While the effects of the researchers on the research subjects can be minimised, but it can't be eliminated. Moreover, the research findings — assuming it become widely known — can alter people's behaviour.

This seems such a basic point, but it's amazing how much social science is presented as if it confidently revealed fixed truths about human beings — forgetting its own role in the constitution of socieities and individuals.

I forgot to add that part of the debate involved a lot haggling over "operationalizing": do we mean short term or long term? Do we mean local effects or global?

Social sciencing is risky business because the phenomena under its purview are multidimensional.

It's my feeling that economists speaking about the economy should be treated like people on the evening news explaining the weather.
Something similar happened when Mt Pinatubo erupted in the Philippines. There was a debate between two eminent atmospheric scientists. One said, "The average global temperature will now drop because energy from the Sun will be blocked by particles in the atmosphere." The other said, "The average global temperature will increase because heat re-radiated by the Earth will be trapped by the particles in the atmosphere." What were we, the non-experts supposed to think if these two experts were in complete disagreement?

Anyway, the global temperature stayed the same.
I think this is a brilliant article. In the public sector there is next to no experiment, because the "producers" of any government program are a protected species. I can never see these brilliant insights successfully applied to government - unless through private contractors. However, private contractors would use lobbying, etc. to stop innovation.
congratulations on clarity and lucidity of article. much enjoyed
Lucid and realistic, I look forward to Mr. Manzi's forthcoming book
Social scientists want more than anything else in the world to be considered to be real scientists, with all kinds of mathematical formulas they can use to determine the future.

In 1970, the University of Minnesota's Graduate School of Public Affairs began to require differential calculus for admission. This was at a time when all kinds of budgeting techniques were being devised to predict outcome of programs.

The requirement was later dropped. The problem surely was the fact that most of the "unknown" factors in the formulae were essentially unquantifiable.
The article is dead on. The only (ONLY) way to get evidence for a causal connection in any system that has more than two objects (so, all systems) is by randomized experiment. The more complex the system, the weaker the evidence each experiment provides. A priori reasoning and theory can lead you to suggest an experiment, but they certainly cannot substitute for them. A theory without experimental evidence is just a narrative. Possibly fictional, with no evidence for it. Coherence of a theory is a necessary condition for justified belief in one, but it is nowhere remotely sufficient.

So, science is devilishly difficult. Social sciences even more so. If you meet someone that's confident about something in these areas, they're probably trying to sell you something.
Chaos is a biotch
the solutions offered today start with a preferred solution and shoe horn the "problems" into fitting those solutions.
Until we actually ID the problems the solutions will continue to fail.
A case in point: The new financial regulation bill doesn't say or do anything about Fannie Mae and Freddie Mac. You can debate the amount they contributed to the crisis but it is impossible to deny they were significant factors. The bill to avert a future crisis ignores the causes of the one we just had.
Now that's leadership ...
The Galileo experiment was to take 3 equally weighted balls and drop them. They of course will reach the ground and the same time ... now repeat this experiment and bring 2 of the balls closer and closer together until they finally touch ... at no time would we expect the 3 balls to reach the ground at different times. In the final test weld the 2 balls together. It is only logical that the 2 balls welded together would fall no faster than the same 2 balls a millimeter apart.
does this mean you reject school vouchers?
See Steve Sailer's response here
1. Bayesiam approaches handle the incremental plausibility changes that true science brings to ideas. Dogmatic (true/not true) thinking is a major weakness in human epistemelogical efforts.

2. Multiple regression models of ideas absorbs causal density.

3. Social science is hypothetically under the influence of theological factors, which have in recent decades been proven, beyond reasonable doubt, to exist using the most mature or orthodox scientific methodology. Because these factors actually include disinformation and scientific interference, attention to them is a sine qua non to successful social science. Your comments constitute yet further confirmation that this last statement is true beyond reasonable doubt. Another prediction is that anyone dogmatically opposed to giving theological hypotheses the needed attention will remain stuck in "stumbling," and only "forward" in delusion or deception.
One wonders if St. Thomas Aquinas could somehow persuade science and religion to join and decide the real fate of our human condition in this secular society today?
Great points; I shudder to think how apt they are for educational research as well, and the giant, "moon shot" policies that result.
Douglas Nance,


Great article. In the areas where governments spend the most on highly uncertain interventions (healthcare, education, financial stimulus, etc) what do you think the appropriate mix is between spending on experimentation/learning and applying existing techniques that have not yet been firmly established? E.g. if we were going to dedicate n% of Medicare's budget to comparative effectiveness trials and the infrastructure to support them, what should n be?
unfortunately, this article will be read by far too few people; and yet its wisdom is crucial to so many in this age of attempted behavior modification. it is humbling, as mr. manzi notes, to realize how little we know; and frightening to realize how much damage can be done by misguided social and government policy. as a great 20th century American wit once observed, 'it ain't what we don't know that scares me, it's what we think we know that just ain't so'
"But sophisticated experimentalists understood that because of the issue’s high causal density, there would be hidden conditionals to the simple rule that “mandatory-arrest policies will reduce domestic violence.” The only way to unearth these conditionals was to conduct replications of the original experiment under a variety of conditions. Indeed, Sherman’s own analysis of the Minnesota study called for such replications. So researchers replicated the RFT six times in cities across the country. In three of those studies, the test groups exposed to the mandatory-arrest policy again experienced a lower rate of rearrest than the control groups did. But in the other three, the test groups had a higher rearrest rate."

Not mentioned is this experiment uses real people with real lives being subjected to arrest/no arrest/whatever .

These actions (WHIMS) profoundly effect those subjected to them- "oh, that wasn't working so we'll do this or that to them now and see what happens"

When Huxley was informed of Darwin's theory of evolution, he is said to have exclaimed "of course, how stupid of me not to have thought of it." In other words, even the greatest theory of biology, the one that orders all the rest of the discipline, is obvious once it had been expounded.

Likewise, most social science theories are like that -- seemingly obvious after the fact, but hardly so before hand. Take, for example, Mancur Olson's theory on the logic of collective action, one result of which is that groups with one massive player and a host of small ones will be likely to maintain a coalition, because the big player will gain enough from collective action to tolerate 'free riding' from its smaller associates. That is born out in practice, even in those hated 'regression studies'. Seems an obvious result once explained, but was hardly so in 1965 when it was proposed.

There are literally hundreds, thousands, of results like Olsen's.

As for the controlled experiment method, even some medical sciences (particularly nutrition) find these inadequate, because in real life not all factors can be controlled, and indeed large scale regression analysis and data exploration more generally in order to suss out previously unsuspected factors -- the unknown unknowns if you will.

Finally, the author ignores consilience, the ability to move up and (esp) down the level of analysis, and indeed across borders of disciplines. To go back to Darwin, the macro effect is species evolution, something that can be seen with the naked eye, but get closer to real 'truth' the 'gene' concept was needed, and eventually the chemistry had to be worked out. Processes had to be traced, etc. Some of this involves experiments, some of it simulations, some of it extrapolations from case studies.

In short, the author needs to get current with social science results, modern social science methods, and the philosophy of science more generally. EO Wilson's 'Consilience' is a good place to start.
Check this out
Bah. Humbug.
It was refreshing to read an incisive article followed by intelligent comments, a rarity on the web these days. I work for a company that studies social policy and for several decades has conducted large random assignment (and other) evaluations involving welfare, medical care, crime, jobs programs, child care, education initiatives—virtually everything that government and foundations spend money on to produce positive outcomes.

But even after meticulously planned and controlled studies, with teams of prominent economists, social scientists, statisticians, and experts in public policy contributing their formidable knowledge and intelligence, it is often still very difficult for researchers to conclude anything definitively. (Sex education based on, versus not based on, abstention is just one example.) I had not encountered the term "causal density," but I will find it supremely useful every time somebody tells me "such-and-such city tightened (or loosened) its gun laws and crime went (up, down, sideways)" or "people who eat yogurt live longer."

Most of the people I work with don't believe anything until it's undergone a random assignment study ("the gold standard," not always possible, of course).
Social science: there is no better oxymoron.

A group of self-important; self-appointed "experts" identifying as absolute a handful of criteria producing some predetermined result-only to find later that their conclusions are non-repeatable because the variables in any significant social activity are in fact infinite and such that any number of combined factors may produce similar results independently of the preselected criteria.
Great piece, puts science into perspective and I hope this sees a lot of classrooms this Fall. It'll see my Philosophy of Science seminar in the Spring; I think its a very readable and thought provoking introduction to the limits of what we do as social scientists. I really wish that more of the electorate understood our many limitations before they buy into the Fad of the Minute.
This article very cogently sets forth the reasons why the scientific method works in the physical sciences but does not work in the social sciences. From this, the author concludes that we should try -- EVEN HARDER -- to make the experimental method work in the social sciences. Wouldn't it be more logical to conclude that human affairs are different from the physical sciences, and a basically different methodology is needed. Take a look at Ludwig Von Mise's big book, Human Action. Von Mises is the acknowledged master of the Austrian School of Economics. The Austrian School is almost universally rejected by mainstream economists. Why? Are its predictions worse than those of mainstream economics? Do its policy ideas make even less sense than the norm in this field? No. The Austrians are ignored and rejected, because they reject the methodology of modern economics. I am not sure that Von Mises is correct. I do find very appealing, however, his approach of rejecting the imitation of natural science and seeking an entirely different methodology.
Curmudgeon Killjoy August 02, 2010 at 7:29 PM
A "scientific understanding" of human society likely will remain beyond the grasp of science until human beings (i.e., the masses) behave like masses (e.g., bodies that remain at rest until an outside force, preferably an enlightened one, moves them in a progressive direction). Until scientists know how to strip the twin delusions -- thought and choice -- from the minds of individuals, scientists may have to be content with just predicting tendencies in human behavior.
As a sociologist, I agree with much of what you said, since its usually covered on the first day of an introductory social science class.

However, I don't think experimentation is the answer. First of all, experiments are extremely artificial, and second, IRB requirements for testing of human subjects preclude investigation into much of what scientists might be interested in. Perhaps this is why Capital One and Reality Television can develop more "precise" results - because they aren't limited by academic ethics and can manipulate people to their hearts' content.

I think a bigger problem is the lack of rigorous theory in social science - a prerequisite for any science. Often social scientists equate "method" with "theory." The problems you described are most especially problems in economics which suffer from a number of flawed assumptions - notably an explicit interest in producing laws comparable to natural sciences, and allegiance to a number of outdated theoretical paradigms - namely neoclassical and Keynesian economic theories that rely too heavily on a rational choice perspective.

This doesn't mean that all social sciences don't know about this flaw. We have a sub-discipline of "Sociology of Knowledge" for example that explains the way paradigms and ideologies constrict awareness of anomalous behaviors. As in an example from economics, consider the theory of George Soros, which I have yet to see economics depts. take seriously. In sum, Soros suggests that economists' models are wrong because they do not include economists as causal variables. Consider Jim Cramer, of any of the economists you mentioned who says this or that action needs to be taken now, and then, en masse, people take that action. This leads to irrational crowd-like behavior that produces asset bubbles. Like Capital One in your example, Soros has put his theory to the test, making billions off of this simple flaw in human "nature," which is actually a flaw in human "science."

Another example of ideological blindspots: in the 1970s a massive economic experiment with "Basic Income" was done in the US and Canada, known as GAI. This was a resounding success, demonstrating that universal basic income does not lead to decreased work ethic, but rather gave individuals the freedom to retrain and select jobs of their own choosing on their own time frame. But this would amount to a complete change in socio-economic organization (and power). Why didn't anyone hear about it? Why is this "proven" experiment in the public sphere as an approach between the "left" and "right" economists (all of whom are market fundamentalists)? From a sociological perspective, the overwhelming economism of US discourse, which operates on an IDEOLOGICAL, not economic, level is a reasonable explanation that sociologists have the theoretical tools to capture.

The fact that people do not want to hear much of the conclusions that critical sociologists offer is not a demonstration of their worth.

I completely agree that the majority of social scientists running bivariate linear regression produce nothing of value, but they are copying economists, who are copying the natural sciences.

Societies are fundamentally different than other empirical realms.
Another FUD article. Why is it with liberal failures there is never any 'fault' to be found? They'll tell you a thousand and one reasons why but never admit error or assume responsibility.
The American Social and Economic Crisis was not addressed in the article. First, history has shown that government under a monarchy, communism, facism or even American capitalism,ad nauseum has not met the social and economic interests of the labor forces, including the small business owner, of the four systems. The purpose of American government was pointed out by Thomas Paine in "Rights of Man": "But how often is the natural propensity to society disturbed or destroyed by the operation of government! When the latter, instead of being ingrafted on the principles of the former, assumes to exist for itself, and acts by partialities of favour and oppression, it becomes the cause of the mischiefs it ought to prevent.(p.108-109) Nothing can appear more contradictory than the principles on which the old governments began, and the condition to which society, civilization, and commerce are capable of carrying mankind. Government on the old system, is an assumption of power,for the aggrandisement of itself; on the new, a delegation of power, for the common benefit of society. The former supports itself by keeping up a system of war; the latter promotes a system of peace, as the true means of enriching a nation. The one encourages national prejudices; the other promotes universal society, as the means of universal commerce. The one measures its prosperity, by the quantity of revenue it extorts; the other proves its excellence, by the small quantity of taxes it requires." (p.113) Apparently never intending his comments to apply to himself the answer to our social and economic problems came from forked-tongued Alexander Hamilton, a pure monarchist, in Federalist Paper 84: "Nothing need be said to illustrate the importance of the prohibition of titles of nobility. This may truly be denominated the corner stone of republican government; for so long as they are excluded,there can never be serious danger that the government will be any other than that of the people." Hamilton's position clearly includes today's Corporatecraters of imagined nobility and usurped power as a governmental entity with hereditary privileges and apparently imagined "entitlement" to manipulate wages in the workplace, to exploit prices in the marketplace and to plunder our Treasury, all done at will with impunity. Founding Guards dreamed of a pure constitutional Republic free of the corruptive forces of the greedy nobilty and corrupt politicians. Closing, I must again rely on the words of the Great Paine to Kingcrafters, Corporatecrafters and Christiancrafters of "charlatanerie" outlined at #3 of his Four Letters on Interesting Subjects written before the Declaration of Independence: "As to Corporations they are without exception so many badges of kingly tyranny, and tend, like every other species of useless pomp, to the oppression and impoverishment of the place, without one single advantage arising from them. They keep up a perpetual spirit of distinction and faction, engross emoluments and advantages to themselves,which ought to be employed to better purposes, and generally get into quarrels and lawsuits with the other part of the inhabitants. They dimish the freedom of every place where they exist...But of all Corporations that of Philadelphia is the most obnoxious, its power resembling that of a hermaphrodite [meaning that Corporations, as usurped governmental entities, have contradictory roles or funcitons], or is at least kind of aristocratical Corporation made hereditary by adoption." -- Blast Dorrough, Fellow New Guard ("new Guards" is a self-explanatory term found in the Declaration of Independence.
Did this study make any determinations on why female officers were far more likely to make a determination that the man is the actual victim, rather than the woman whose claiming it?

Annette's Story: The Other Face Of Domestic Violence

Read this and see what I mean. Please sign the guest book.
Kudos to Mr. Manzi. Excellent article. Well presented, and pleasantly devoid of the rightward slant I often find in City Journal.
Social Science knows what works: Liberty.
The U.S. proved it! Now, if we can just remember the lessons!
Thanks Mt. Manzi and MI/city Journal for a very interesting article on 'Social Scientism' and experimental analysis....keep up the good work.
Interesting points. As a social studies teacher for 50 yrs, I wanted to see Mr. Manzi also take a look at the very liberal culture all these studies use as an a priori belief in a secular approach with govt. policies as the enth degree of all solutions. Then too, these so called experts never, never seem to use any faith initiatives, or theological explanations for either individual or group reactions to social, economic, emotional, cultural even military effects on varied groups of people groups. Science will never explain variety, differences, or lack of taking responsibility for thoughts, actions, motives and propaganda used to enforce solutions on people without liberty, for example.
Would you please share the name of the nurse project that you referred to?
Bravo--I had in the past several days concluded what you have . I got there by looking at the history of economic theories but, what was more significant, reading a short preface to a book on the asiatic nethod of organizing a society written by ernest gellner who was way ahead of everyone on this issue.The idea he just tossed out was that our genetics favor variability of behavior, a trait missing from all but a handful of animals, which permits the infinite variety of ways of doing and thinking things we humans display. I see that your current company affiliation is some sort of attempt at coping with this variety. Good luck
anti-social-science August 02, 2010 at 1:03 PM
Jim Manzi says that social scientific theories based on observation can be checked by experiment to see if they are valid scientific truths. What is meant by 'valid scientific truth' is something that can be known about something else that is true in all places and times in the universe. For example, Planck's constant or the speed or light is true in all places and times in the universe. (Or so we assume in order to conduct modern natural science - we cannot check all places and times, but we assume uniformity in nature).

Experiment might be able to discover the universal laws of the human mind too, according to Manzi (and philosophers going back to Hume and Comte). He cites Capital One mailing people white or blue envelopes and seeing which elicits more responses. Let's say Capital One found that blue envelopes get more responses from Americans in 2003. They "proved" this through experiment. What does that mean? Does it mean that Capital One has discovered a universal law that human beings in all places and times will return credit card application in blue envelopes at a higher rate? Manzi seems to suggest so; this is what scientific knowledge is supposed to be.

But has Capital One really proven that? Would the Japanese public return blue envelopes more than white? How about the Ugandan or Danish people? And what about America people at different times? Say 50 years ago or 50 years from the future. Would Americans still prefer blue envelopes? I dont know, but I would guess that other peoples or the same people at different times would choose different colored envelopes.

You dont have to test all peoples at all times; natural science doesn't test this way because of the uniformity assumption. But if you could find one counterexample of a foreign people at any time preferring white to blue envelopes, then you have disproven Capital One's scientific discovery as much as Planck's constant would be disproven if you could find any counterexample.

The problem with social science is that human action is strongly determined by beliefs, which can change over time and are very different in different places. Yet social scientists deny this role of belief and instead elevate natural scientific methodology to the level of the only way to obtain knowledge, and consequently argue that law and policy must be based on social science. This is in a very real sense the worship and idolatry of natural science. The problem is that social science discovers different things at different times because the way for obtaining knowledge about the natural world is different for obtaining knowledge about the human mind. Beliefs are conceptual, not material, and are therefore invisible to natural science. Any society that bases its laws and policy on social science is putting its foundations on ever-shifting quicksand.

Dont believe me? Go right ahead. Social science has failed for over 200 years and its failures are mounting and getting worse (see economists failure to predict the financial crisis, see post-crisis economists nutty suggestions). See the historical exponential increase in crime over the past 100 years, despite criminologists increasing role in lawmaking. Social science is an utter failure compared to natural science - the problem is that our elites cant admit this because they have a religious faith in natural scientific methodology.
This is sort of a yawner. Sorry.
Francis W Stocker, M.D. August 02, 2010 at 6:00 AM
It may not be possible to predict bad behavior, but the police and trauma teams know what volume of injuries to expect from each neighborhood. Homicide rates are remarkably predictable for ethnic groups regardless of latitude and longitude. (The fatal outcomes are mitigated by the trauma teams in American cities.) The homicide champs for white folks are the Russians and the residents of the Baltic Staes. May be related to alcoholism.
What needs to be studied is the number of offspring from the bad behavors. Look at the homicide rates in Central America, sub - Sahara Africa and look at the fertility rates. It will give a picture of mayhem to come.
The fertility rates amongst illegal aliens from south of the border is higher than in their home counties.
I am a retired social worker in corrections. Your article touched on many things that have troubled me in my work with criminals, drug addicts, and gang members. Over the years, I've seen many social engineering programs funded on "conventional" ideas and assumptions that fail to produce intended outcomes. Of course, the agents and agencies that sponsor and execute these programs have vested interests and investments in the outcomes. Despite poor outcomes, these programs get politicized and the public is duped into continuing funding and doing more of what does no work or even be counterproductive.

Anyway, thank you very much for illuminating the issues and problems you identified.

Jaime Jorquez
Hurricane, Utah

This is an excellent article - very informative. Thanks !
There is a McGuffy's Reader version of basic Economics literature,and it was written by Henry Hazzlit. But logic and common sense have not been complicated enough for the burgeoning field of post-war economists.Now, there are so many opinions on every situation that economists are as divided as opposing medical forensic experts testifying in a homicide case.
For me, a seminal essay on societal assumption;that is that human behaviour can be made predictable on a wide enough scale to be important in the generalization of outcomes.So far not proven nor likely to be.
Something we plain folk suspected all along
bravo! very little to disagree with here...anyone who has poked around social science data sets knows that one can produce the "correct" outcomes without enough tweaking.