The Myth of Knowledge

14 Apr

People frequently overestimate how much they know. They also confidently share things they don’t know.

But why?

One reason is social desirability—the motivation to appear knowledgeable in front of others and perhaps ourselves (our ability to fool ourselves deserves closer inspection). Another is that conversations are typically not carried out for any epistemological purpose but for emotional and social purposes. So we make up things because the truth is unimportant, at times even a hindrance. For example, envision a conversation where both parties profess their ignorance. The conversation will be pretty short. Fencing yourself within what you know means you can’t discuss many salient topics.

For two, lay conjecture often suffices when the audience is ignorant. More generally, the propensity to make up something depends on the speaker’s assessment of chances of getting caught, which is a function of the audience’s knowledge, sympathies—towards the speaker or the topic being spoken about, and volubility, which in itself may depend on the status difference, among other things. But as any professor will tell you—students happily make things up even when they are in the presence of experts. So the rate of decline in propensity to make up stuff as the chance of getting caught increases is low, and the absolute level of propensity to make up stuff is high. It is also likely that the confidence with which people typically say such ‘lies’ is an attempt to cover up their ignorance as confidence is taken by others at its face value: as a sign of surety about facts. Thus, never seriously threatened by their ignorance, people build a somewhat more positive assessment of how much they know.

But can it also be that people think that lying about their ignorance is only a minor transgression? Do people have this innate belief (which rarely gets challenged) that somehow when they speak, they will be able to be right; some sort of a ‘God bias’ — that they will be the exception to making sense without knowing?


The modern era of knowledge production has brought its own challenges. First, the rate of production of knowledge has exploded. And as the rate of production of knowledge accelerates and the rate of learning stagnates, relative ignorance increases, which I take is the case. But not only has the rate of knowledge production exploded but so has the complexity. Increasingly substantive discussions about important areas of human activity (public policy more broadly but say health, fiscal policy, etc.) need more sophisticated thought and deeper immersion in the wealth of knowledge that has been produced.


It is often said, often without a whoop of surprise, ‘the more you know, the more you realize how much you don’t know.’ But why would that be? Are limits of knowledge so obscure that they cannot be known by the proverbial (and now the literal) average Joe? In fact, it is easy to deduce one’s ignorance. For example, we are surrounded by phenomena that we can’t describe well, much less explain. To infer our mean level of knowledge, we may want to do the following: recognize the fact that even in areas where we claim expertise we often fall short, hence we must really know very little about the things we don’t spend time learning.

The point about the limits of our knowledge is broader and not there to malign the average Joe. There are real limits to what humans can achieve. Think about the following: If one were to read a book a week for the next 50 years, one would end up reading 2500 books. If the smallness of the number surprises you, then let that be a lesson. The point allows me to segue into the next one—the myth of being well-read.

The Myth of Being Well-Read

Often, people confuse being well-read to mean reading a few bad books poorly. To be well-read, one must satisfy three criteria: 1) have read at the very least ~250 (arbitrary low number) books; 2) a substantial majority, if not all, of which ought to be good—literature or non-fiction; 3) and they ought to have been read well.

The 7-step program for correctly classifying yourself as well-read or not

  • Only a few people (<< 1%) are well-read. Do you think you are in such an elite company?
  • Force yourself to list as many books that you have read in the past year (or life). If that number is less than 10, you may not be well-read. Apply this to specific areas as needed. For example, you may ask yourself whether you are well-read in political science.
  • Alternately, think hard about how many books you bought or checked out from the library last year.
  • If you catch yourself citing one book repeatedly, whenever the topic of books comes up during conversations, you are unlikely to be well-read. In the U.S., it is typically The Catcher in the Rye, though that may be changing — Harry Potter, Dan Brown, Twilight series, etc. may be the new ‘go-to’ books. For a colleague of mine, it is White Noise. In some circles, it may be some Malcolm Gladwell or Thomas Friedman’s book.
  • Relatedly, if Harry Potter, Blink, Atlas Shrugged, Catcher in the Rye, etc., are among your favorite books, it is unlikely that you are well-read.
  • If you are younger than 21 (or typically 25), you couldn’t have read much. There are a few exceptions. They help make the rule.
  • Count the books you own. If you don’t own more than 50 books, you are unlikely to have read much.

Lifting the Veil on Some Issues Around The Burka Debate

30 Mar

For the unfamiliar, the BBC guide to Muslim veils.

The somewhat polemical:
Assuming that God has recommended that women wear the burka, assuming that burka has no impact on a woman’s ability to communicate or quality of life, as has been suggested by its supporters, then here’s a suggestion—to all men, who haven’t been ordered by God to wear a burka, and who don’t see a downside to wearing it—why not voluntarily commit to wearing the burka, since no law opposes such a voluntary act, to show solidarity with the women. My sense is that even the French would come to support the burka if Muslim men en masse chose to wear it.

More considered:
‘The interior ministry says only 1,900 women wear full veils in France, home to Europe’s biggest Muslim minority’ (BBC). If the problem is interpreted solely in terms of women wearing the veil, then it is much smaller than the dust in its wake.

There are three competing concerns at the heart of the debate: Protecting the rights of women who voluntarily want to wear it, protecting the rights of women who are forced to wear it, and protecting (French) ‘culture.’ Setting aside cultural concerns for the moment, let’s focus on the first two claims.

People are incredulous of the claim that women will voluntarily choose to wear something so straightforwardly unpleasant. Even when confronted with a woman who claims to comply voluntarily, they fear coercion, or something akin to brainwashing at play. There is merit to the thought. However, there is much evidence that women subject themselves to many unpleasant things voluntarily, such as wearing high heels (which I understand are uncomfortable to wear). So it is very likely indeed that there is ‘voluntary compliance’ by some women.

Assuming there exist both, voluntary compliers, and those forced to wear the niqab, wouldn’t it be pleasant if we could ensure the rights of both? In fact, doesn’t the extant legal framework provide for such a privilege already? Yes and no, mostly no. While it is true that women forced to wear the niqab can petition the police, it is unlikely to happen for a variety of reasons. Going to the police would mean going against the family, which may mean doing something painful, and risking financial and physical well-being. Additionally, the laws governing such ‘coercion’ are likely to carry modest penalties and unlikely to redress the numerous correlated issues including inadequate financial, and educational opportunities. Many of the issues raised here would seem familiar to people working with domestic abuse, and they are, and the modern state hasn’t (tried to) found a good solution.

Perhaps both camps will agree that wearing a niqab does dramatically limit the career opportunities for women. Of course, people in one of the camps may be happy that there are limits to such opportunities but let’s assume that they would be happy if the women had the same opportunities. Part of the problem here then is the norms of dressing in business environments in the West. Entrepreneurs in Saudi Arabia recently brought to air a television talk show in which both of the hosts wore the niqab. The entire effect was disturbing. However, that isn’t the point. The point is that there may be ways not to reduce career opportunities for women based on the dress code, which after all seems ‘coercive.’

Time considerations mean a fuller consideration of the issue will have to wait. One last point – One of the problems cited about the burka is that it poses a security threat, which has some merit, given its long history in being used a method of escape, including by militant clerics.

Idealog: Opening Interfaces to Websites and Software

20 Mar

Imagine the following scenario: You go to NYTimes.com, and are offered a choice between a variety of interfaces, not just skins or font adjustments, built for a variety of purposes by whoever wants to put in the effort. You get to pick the interface that is right for you – carries the stories you like, presented in the fashion you prefer. Wouldn’t that be something?

Currently, we are made to choose between using weak customization options or build some ad hoc interface using RSS readers. What is missing is open-sourced or internally produced a selection of interfaces that cater to diverse needs, and wants of the public.

We need to separate data from the interface. Applying it to the example at hand – NY Times is in the data business, not the interface business. So like Google Maps, it can make data available, with some stipulations, including monetization requirements, and let others take charge of creatively thinking of ways that data can be presented. If this seems too revolutionary, more middle of the road options exist. NY Times can allow people to build interfaces which are then made available on the New York Times site. For monetization, NY Times can reserve areas of real estate, or charge users for using some ad-free interfaces.

This trick can be replicated across websites, and easily extended to software. For example, MS-Excel can have a variety of interfaces, all easily searchable, downloadable, and deployable, that cater to specific needs of say, Chemical Engineers, or Microbiologists, or programmers. The logic remains the same – MS needn’t be in the interface business, or more limitedly, doesn’t need to control it completely or inefficiently (for it does allow tedious customization), but can be a platform on which people can build, and share, innovative ways to exploit the underlying engine.

An adjacent broader and more useful idea is to come up with a rich interface development toolkit that provides access to processed open data.

Expanding the Database Yet Further

25 Feb

“College sophomores may not be people,” wrote Carl Hovland, quoting Tolman. Yet research done on them continues to be a significant part of all research in Social Science. The status quo is a function of cost and effort.

In 2007, Cindy Kam, proposed using the university staff as a way to move beyond the ‘narrow database.’

There are two other convenient ways of expanding the database yet further: alumni, and local community colleges. Using social networks, universities can tap into interested students, and staff acquaintances.

A common (university-wide) platform to recruit and manage panels of such respondents would not only yield greater quality control and convenience but also cost savings.

An Appetizing Question: Why Don’t We Read Nutrition Labels?

24 Feb

A single wheat tortilla has 18% of the daily value of sodium. I discovered it just recently. The discovery followed my reading an article discussing serious negative consequences of excess sodium for heart health. In turn, the discovery was followed by two others: I realized that I insufficiently assimilated information from food labels in general and that I was curious to know why that was the case.

One way to understand this shift is the following: attention is a limited resource and only allocated to things deemed important. Once the importance was established, attention followed.

However,

  1. Attention is not that limited a resource.
  2. The marginal cost of attention to sodium is small given I generally do look at the calorie content.
  3. Given that I encounter same or similar choice tasks repeatedly—cost per choice, invested once, is close to zero.

What are the reasons behind this seeming conundrum?

In the domain of food, we care about three things: price, taste, and health. Let’s leave out the price for now. This leaves us with taste and health. While making a choice, we recruit ‘relevant’ (more later) information to assess choices in a manner that maximize our utility.

Assume we have a strict preference for taste over health; more narrowly —taste always wins whatever the health information; health information comes into play only when the taste is equivalent. Health information in this scenario is immaterial, and one only needs to focus on information about taste to maximize his/her utility. So, one way to explain my inattention to relevant health information is just that.

Expanding upon the toy example, orderings of things we care about (utilities) dictate the way we seek information, and what information is sought. However, observation tells us that the ordinal structure of utilities is manipulable in the domain of food. For example, we prefer taste strongly but if health information were to be made salient, we would be liable to choose something healthy. One inference which we can draw from such manipulability of order is that the initial preference ordering must not have been strong. But that doesn’t seem right given our strong preference for taste ‘explains’ subdual of information seeking on health.

Rational choice assumes that if information acquisition costs are zero, more information should always be sought, and used in decision making. Rational Choice seems inadequate to the task of explaining, hide and not seek.

Let’s assume we have subconscious and conscious preferences (aside from assuming a subconscious). Subconsciously we greatly prefer taste more than health. Consciously we prefer the reverse. Taste wins if health information is not made salient at the time of purchase. Assuming subconscious controls behavior, health information is deliberately not sought.

Another way to think about underlying preference structure is the following— aside from preferring tastier food, we also prefer feeling good about our decision. Feelings about a decision are evaluative and emerge from whether we chose wisely given information. So one way to include good feelings is to choose healthy food but that sacrifices our preference for taste. Another way this is resolved is ignoring information about health, which is much more easily ignored than information about taste.

Given this, we suppress health information. Interestingly this suppression doesn’t extend to information seeking about health on all fronts but applies only during decision time about a food.

Another interesting psychological thing to note here is that we have negative affect associated with decisions that lead to negative long-term consequences, but we also have ways to prevent this negative affect pathway from being triggered at all. Additionally, the information suppression isn’t a one-time-only but long term because we want to repeatedly ”sin.” This, in turn, means that we firstly somehow ‘know’ that the food is unhealthy and hence not look at the health information, otherwise wouldn’t it just help boost one of the reasons for consuming something tasty, but don’t consciously acknowledge this information.

Yet another way to think about the problem is to assume that we have preferences for health but they are somewhat lower in order of priority. For lower order preferences (here health), information seeking becomes more passive and increasingly depends on how easy it is to acquire information, for example, how prominently it is displayed. Social desirability pressures may also play a larger role in moderating information acquisition when importance is low. For example, in US people frown upon those who look at labels in a supermarket. Thus cowed, people may be less likely to look up information. Though it was always possible to look up the information once home, and now given ease and convenience of anonymous information gathering (Internet), it is likely that social desirability issues are less of a factor (it is likely that social desirability pressures continue to apply when one is alone.) However in cases where other lower order preferences predict same choice information about them is likely highlighted. For example, if a tasty thing were healthy as well, it is likely that one reminds oneself of the health benefits while making the choice.

But why is taste implicitly prioritized over health? One explanation is that preference for taste is evolutionary – the positive immuno-response from eating calorie-rich food is biologically potent. Another is that consequences on health from choosing unhealthy food are long term while gratifications from taste are instantaneous. Given that, it allows us to more readily imagine the consequences of one which in turn is perhaps define one of the key ways we decide our preferences. Lastly, the preference for taste in matters of food has become likelier due to advertising and its constant valorization of taste over everything else.

It is still awe-inducing to see to what degree our brain is lazy and inhibits acquisition of reasonably readily available information.

The above analysis assumed considerations dictating choice at the point of purchase. Once we have bought something, however, another consideration applies — we have invested in x, so now enjoy it. What is the point of reading information now that I have already spent money?

A few straight forward policy proposals emerge, given what we know about how people behave,

  1. Front of package labeling
  2. Prominent, easy to read, comprehend, labeling
  3. Priming aim — be healthy
  4. Priming habit — look for information when buying food
  5. Priming consequences

Link:

http://www.wnyc.org/shows/radiolab/episodes/2008/11/14

Note:
The word ‘we’ is in quotes in the title because I do not have data to show how widespread the tendency is.

You Got (Fraudulent) Mail!

22 Jan

Republican [Con-]census

The party that dislikes the census, put a hold on the nominee for census bureau, is now sending out fraudulent mail surveys that seem as if they were from the census bureau. Read here, here, and here. Accusation for being ‘fraudulent’ stems, not only from the use of `census’, but also from it being an attempt at “frugging”, the practice of cloaking a fund raising appeal in what appears to be a research. (“Suggers” sell using surveys.)

Who Knew
One of the people who has benefited the most from Macaulay’s reforms recently sent an email, part of a larger email chain that now generally implies some travesty, which quoted Macaulay as having spoken the following (on Feb 2nd, 1835 in India, the same day his gave his Minute on Education in Britain) –

I have traveled across the length and breadth of India and I have not seen one person who is a beggar, who is a thief. Such wealth I have seen in this country, such high moral values, people of such caliber, that I do not think we would ever conquer this country, unless we break the very backbone of this nation, which is her spiritual and cultural heritage, and, therefore, I propose that we replace her old and ancient education system, her culture, for if the Indians think that all that is foreign and English is good and greater than their own, they will lose their self-esteem, their native self-culture and they will become what we want them, a truly dominated nation?

Of course, Lord Macaulay never said such a thing. He said many other tawdry things but never did he eulogize the absolutely hokum positive images of India. While we all want glorious histories, past isn’t as glorious. Neither are confessions of colonial rapacity generally so naked.

Suggestion: For greater imagined glories one must go further back than just 1835, when the ‘Muggles’ had had their way (as Hindus may point out), writing had been invented, into the mists of more obscure pre-history where one can have his way with the truth. How about the crowning glories of Lord Rama and his return on an airplane like ‘pushpak’ vahan?

Flawed Analyses in Deliberative Polls

14 Jan

A Deliberative Poll works as follows: A random sample of people are surveyed. Out of the initial sample, a random subset is invited to deliberate, given balanced briefing materials, randomly assigned to small groups moderated by trained moderators, allowed the opportunity to quiz experts, and in the end surveyed again.

Reports and papers on Deliberative Polls often carry comparisons between participants to non-participants on a host of attitudinal, and demographic variables (e.g. see here, and here). The analysis purports to answer whether people who came to Deliberative Poll were different from those who didn’t and to compare participant sample to the population. This sounds about right, except the comparison is made between participants, and a pool of two likely distinct sub-populations—people who were never invited (probably a representative, random set), and people who were invited but never came. Under plausible and probable assumptions, such pooling biases against finding a result.

The key thing we want to measure is self-selection bias—was there a difference between people who accepted the invitation, and who did not. The right way to estimate the bias would be as follows:

(Participant/Didn't come) ~ socio-demographics (gender, education, income, party id, age) + knowledge + attitude extremity

Effect sizes can be provided to summarize the extent of bias. This kind of analysis can account for the fact that bias may not occur at first marginals (gender), but at second marginals (less educated women). (This all can be theory-driven, or more descriptive in purpose). The analysis also allows for smaller effects to be seen, as variance within cells is reduced.

p values
When the conservative thing to do is to reject the null hypothesis, think a little less about p-values.

Assuming initial survey approximates a ‘representative’ sample of the entire population and assuming we want to infer how ‘representative’ participants are of the entire population, it makes sense just to report mean differences without the p-values.

The survey sample estimates stand in for the entire population. Entire population census numbers are without standard errors or very low s.e. so comparisons are always significant.

By comparing to an uncertain estimate of the population, one cannot say whether the participant sample was representative of the entire population. That estimation is without bias but suffers the following problem – the more uncertain the population estimate, the less likely one can reject null, and more likely one is to conclude that the participant sample is representative. One way to deal with this is to do the following – Have 95% conf. band of sample estimate of population and then calculate max and min difference between the sample and report that.

Name calling
Calling the analysis ‘representativeness’ analysis seems misleading on two counts:

  1. While a clear representation question can be answered by some analysis, none such question is answered and can be answered by the analysis presented. Moreover, it isn’t clear if it relates to some larger politically meaningful variable. For example, one question that can be posed is whether participant sample resembles the population at large. For answering such a question, one would want to compare population estimates to census estimates (which have near zero variance, so t-tests, etc. would be pointless.)
  2. In a series of papers in the 1970s, Kruskal and Mosteller (citations at the end) rightly excoriate the use of `representativeness’, which is fuzzy and open to abuse.

Citations
Kruskal, W; Mosteller, F. (1979) Representative sampling I: non-scientific literature. Intern Stat Rev. 47:13-24.
Kruskal, W; Mosteller, F. (1979) Representative sampling II: scientific literature. Intern Stat Rev. 47:111-127.
Kruskal, W; Mosteller, F. (1979) Representative sampling III: the current statistical literature. Intern Stat Rev. 47:245-265.

Idealog: Internet Panel + Media Monitoring

4 Jan

Media scholars have for long complained about the lack of good measures of media use. Survey self-reports have been shown to be notoriously unreliable, especially for news, where there is significant over-reporting, and without good measures, research lags. The same is true for most research in marketing.

Until recently, the state of the art aggregate media use measures were Nielsen ratings, which put a `meter’ in a few households, or asked people to keep a diary of what they saw. In short, the aggregate measures were pretty bad as well. Digital media, which allows for effortless tracking, and the rise of Internet polling however for the first time provides an opportunity to create `panels’ of respondents for whom we have near perfect measures of media use. The proposal is quite simple: create a hybrid of Nielsen on steroids and YouGov/Polimetrix or Knowledge Network kind of recruiting of individuals.

Logistics: Give people free cable and Internet (~ 80/month) in return for 2 hours of their time per month and monitoring of media consumption. Pay people who already have cable (~100/month) for installing a device and software. Recording channel information is enough for TV, but Internet equivalent of a channel—domain—clearly isn’t, as people can self-select within websites. So we only need to monitor the channel for TV but more for the Internet.

While the number of devices on which people browse the Internet, and watch TV has multiplied, there generally remains only one `pipe’ per house. We can install a monitoring device at the central hub for cable, and automatically install software for anyone who connects to the Internet router or do passive monitoring on the router. Monitoring can also be done through applications on mobile devices.

Monetizability: Consumer companies (say Kellog’s, Ford), Communication researchers, Political hacks (e.g. how many watched campaign ads) will all pay for it. The crucial innovation (modest) is the addition of the possibility to survey people on a broad range of topics, in addition to getting great media use measures.

Addressing privacy concerns:

  1. Limit recording information to certain channels/websites, ones on which customers advertise, etc. This changing list can be made subject to approval by the individual.
  2. Provide for a web-interface where people can look/suppress the data before it is sent out. Of course, reconfirm that all data is anonymous to deter such censoring.

Ensuring privacy may lead to some data censoring and we can try to prorate the data we get it a couple of ways –

  • Survey people on media use
  • Use Television Rating Points (TRP) by sociodemographics to weight data.

Coding Issues in the ANES Cumulative File

29 Nov

I have, on occasion, used American National Election Studies (ANES) cumulative file to do over time comparisons. Roughly half of those times, I have found patterns that don’t make much sense. Only a small fraction of the times when the patterns didn’t make sense have I chosen to investigate the data more closely as a likely explanation for aberrant patterns. The following ‘finding’ is a result of such effort.

ANES cumulative file (1948–2004) carries a variety of indices. In creating some of the indices, it appears pre-election measures have been combined with post-election measures in some of the years. If that wasn’t enough, at least one of the times, the same index in some years has pre-election measure combined with the post-election measure, while using only post measures in other years. Here’s an example –

‘External Efficacy Index’ (VCF0648) is built out of two items –

http://www.electionstudies.org/studypages/cdf/anes_cdf_var.txt

Item 1: Public officials don’t care much what people like me think.

Item 2: People like me don’t have any say about what the government does

Item 2 is asked both pre and post-election in some cycles. In 1996, efficacy was built from

960568 (pre), 961244 (post)

http://www.electionstudies.org/studypages/1996prepost/nes1996var.txt

[you can ID post-election wave questions through the following coding category – Inap, no Post IW]. Post version of 960568 is 961245

While in 2000 it is built out of – 001527 (post), 001528 (post)

http://www.electionstudies.org/studypages/2000prepost/anes2000prepost_var.txt

I have alerted the ANES staff, and it is likely that the new iteration of the cumulative file will fix this particular issue.

Prospect Theory in Retrospect

6 Jun

Tversky and Kahneman, in “The Framing of Decisions and the Psychology of Choice”, “describe decision problems in which people systematically violate the requirements of consistency and coherence, and […] trace these violations to the psychological principles that govern the perception of decision problems and the evaluation of options.”

They start with the following example,

Imagine that the U.S. is preparing for the outbreak of an unusual Asian disease, which is expected to kill 600 people. Two alternative programs to combat the disease have been proposed. Assume that the exact scientific estimate of the consequences of the programs are as follows:

[one set of respondents, condition 1]
If Program A is adopted, 200 people will be saved. [72 percent]
If Program B is adopted, there is 1/3 probability that 600 people will be saved, and 2/3 probability that no people will be saved. [28 percent]

[second set of respondents, condition 2]
If Program C is adopted 400 people will die. [22 percent]
If Program D is adopted there is 1/3 probability that nobody will die, and 2/3 probability that 600 people will die. [78 percent]

They add:
“[I]t is easy to see that the two problems are effectively identical. The only difference between them is that the outcomes are described in problem 1 by the number of lives saved and in problem 2 by the number of lives lost. The change is accompanied by a pronounced shift from risk aversion to risk taking.”

Given the empirical result, they propose “prospect theory” which we can summarize as people exhibit risk aversion when faced with gains, and risk-seeking when faced with losses.

Why is Program A less “risky” than Program B?

Expected utility of A can be seen as equal to that of B over repeated draws. However, over the next draw – which we can assume to the question’s intent, Program A provides a certain outcome of 200, while Program B is a toss-up between 0 or 600. Hence, Program B can be seen as risky.

Looked at more closely, however, the interpretation of Program B is still harder –
Probability is commonly understood as over repeated draws. Here – Given infinite draws – 1/3 of the times it will yield a 600, and the rest of the 2/3 of times a 0. (~ Frequentist) Tversky and Kahneman share the frequentist take on probability (though they frame it differently) – “The utility of a risky prospect is equal to the expected utility of its outcomes, obtained by weighting the utility of each possible outcome by its probability.” (This takes directly from statistical decision theory that defines risk as integral of the loss function. The calculation is inapplicable for any one draw.)

What is the meaning of probability for the next draw? If it is a random event, then we have no knowledge of the next toss. The way it is used here however is different – we know that it isn’t a ‘random event’ and that we have some knowledge of the outcome of the next toss, and we are expressing ‘confidence’ in the outcome of the next toss. (~ Bayesian) Transcribing Program B’s description in Bayesian framework, we are 33% confident that all 600 will be saved, while 66% confident that we will fail utterly. (N.B. The probability distribution for the predicted event emanates likely from a threshold process – all or nothing kind of gambling event. Alternate processes may entail that the counterfactual to utter failure is a slightly less than utter failure, and so on and so forth on a continuum.) Two-third confidence in utter failure (all die), makes the decision task ‘risky’.

Argument about Rationality and Equality of utility (between A, B, C, and D)

According to Tversky and Kahneman, utility of Program A is same as Program B. As we can infer from above, if we constrain estimation of utility to the next draw – which is in line with the way the question is put forth, Program A is superior to Program B. An alternate way to put the question could have been – “over the next 100,000 draws, programs provide these outcomes. Which one would you prefer?” Looked at in that light, the significant majority who choose Program A over B can be seen as rational.

However, the central finding of Tversky and Kahneman is “preference reversal” between battery 1 (gains story) and battery 2 (losses story). We see a reversal from majority preferring ‘risk aversion’ to a majority preferring ‘risk taking’ between the two ‘conditions’. Looked independently, the majority’s support in each condition seems logical, but why is that the case? We have already made a case for battery 1, and for battery 2 the case would run something like this – given overwhelming number of fatalities, one would want to try a risky option. Except of course, mortality figures in A and C, and B and D, are the same, and so is the risk calculus.

For Tversky and Kahneman’s findings to be seen as a testimony of human irrationality, Program A should basically be seen as equivalent to Program C, and Program B to Program D. And the lack of ‘consistency’ between choices an indicator of irrationality. In condition 1, our attention is selectively moored towards the positive, while in condition 2, towards the negative, and respondents evaluate risk based on different DVs (even though they are the same). The findings are unequivocally normatively problematic, and provide a manuscript for strategic actors for how to “frame” policy choices in ways that will garner support.

Brief points about measurement and experiment design

1) There is no “control” group. One imagines that the rational split would be one gotten in condition A, or condition B, or as the authors indicate some version of 50-50 split. There is reason to believe that 50-50 split is not the rational split in either of the conditions (with perhaps 100-0 split in either conditions being ‘rational’. This doesn’t overturn the findings but merely provides an interpretation of the control. Definitions of control are important as they allow us to see the direction of bias. Here – it allows us to see that condition 1 allows for more people to reach the ‘correct decision’ than condition 2.)
2) To what extent is the finding an artifact of the way the question is posed? It is hard to tell.

  • A 50-50 split response condition would be achieved if respondents think that both choices are equivalent, and hence pick one choice randomly. But given respondents are liable to imagine that a ‘unique’ solution exists, given they have been brought into a university laboratory and asked a question, people are likely to try to read the tea leaves. Of course, people systematically reading tea-leaves in one way means something else is perhaps going on, but still it is very likely that deviations from 50-50 split would be much less if one were to provide a response option that both choices are equivalent. This is so because some number will choose 3, and then you can either eliminate that sub-sample, and calculate new percentages of deviations from 50-50 by constraining to the two choices (which will likely yield a larger percentage swing) or include everyone, and find a smaller percentage swing.
  • The stump for condition B (that offers Program C or Program D) is the same as stump for condition A – the disease is ‘expected to kill 600 people’. In light of that, description of Program C (“If Program C is adopted 400 people will die”) offers no information about the other 200. Respondent can imagine that 200 will be saved, but isn’t particularly sure of their fate. On the other hand, with the same stump, information about ‘200 will be saved’ allows us to weakly infer that 400 people will die. This biases the swing in favor of the results that we see.
  • If we are to imagine that results are driven entirely differential risk aversion in losses and profits, and not some other cognitive malaise, then it would have been nice to see clearer enunciation of the outcomes. For example, description of Program A could have reworded as ‘If Program A is adopted, 200 people will be saved, and 400 people will not. Or only 200 out of the 600 people will be saved’. This would have likely attenuated the swing that we see, though it is open to empirical investigation. The larger point perhaps is that there are multiple ‘manipulations’, and that the results we see may be artifactual, coming instead from another source. For example – Kuhberger (1995) noted that outcomes in the Asian disease problem are inadequately specified. When Kuhberger made the outcomes explicit (e.g. stating that 200 will be saved and 400 will die), ‘‘framing’’ effects vanish[ed].”

Further Reading

Bless, H., Betsch, T. & Franzen, A. (1998). Framing the framing effect: The impact of context cues on solutions to the ‘Asian disease’ problem. European Journal of Social Psychology, 28, 287–291.

Druckman, J. N. (2001). Evaluating framing effects. Journal of Economic Psychology, 22, 91–101.

Kahneman, D. & Tversky, A. (1979). Prospect Theory: An analysis of decision under risk. Econometrica, 47, 263–291.

Kühberger, A. (1995). The framing of decisions: A new look at old problems. Organizational Behavior and Human Decision Processes, 62, 230–240.

Levin, I. P., Schneider, S. L. & Gaeth, G. J. (1998). All frames are not created equal: A typology and critical analysis of framing effects. Organizational Behavior and Human Decision Making Processes, 76, 149–188.

The Problem of Pakistan

20 Apr

The following article is by a regular contributor to the blog, Chaste. The article is a comprehensive analytic exploration of the extent of problem that we face in Pakistan, and a considered exploration of the possible alternatives.

I will first sketch the history of and the situation in Pakistan, and then suggest a range of options. Throughout the piece, I will use “Jihadist groups” as an umbrella term for groups including the Kashmiri terrorists, LET, UJC, AQ, Taliban, and the like. My apologies to those who see Jihad as an internal spiritual struggle; I cannot think of a more efficient term.

In 1965, Old Pakistan tried unsuccessfully to annex the part of Kashmir that fell to India during the spoils of independence/partition. Six years later, in 1971, an East Pakistan (Bengali) based party won national elections. Because the party was East Pakistan based, the West Pakistan dominated military refused to recognize the election results and launched a coup. What followed was an incredibly bloody suppression of a Bengali insurgency. The West Pakistani military action caused the killing of about 1.5 million people in the space of nine months; that would average out at five thousand killings every day during the whole of those nine months. These are controversial numbers, difficult to verify, so I have halved the numbers claimed by Bangladesh. Ten million Bengalis became refugees in India. India started to support the insurgency, and Pakistan declared war on India at the end of 1971. With the help of Bengali insurgents, India forced a surrender of Pakistani forces in the area, and East Pakistan became independent Bangladesh.

In nearly forty years of independence, Bangladesh has coexisted peacefully with both Pakistan and India, viewing neither as a threat or even a rival. It should be noted here Bangladesh’s population is roughly equal to that of Pakistan and is about 90% Muslim. On the other hand, West Pakistan took ownership of the identity of Old Pakistan – viewing India as a rival and a threat, and itself as the lands of the Muslims of the subcontinent. It did this despite having just lost about half of its population to Bangladesh, and being reduced to a nation of Punjabis, Sindhis, and a host of peoples in the mountainous Northwest of the subcontinent. West Pakistan chose to overlook its slaughter of up to three million Bangladeshis (West Pakistan would point out in its defense that it murdered and raped a disproportionately high number of Hindus). Instead, West Pakistan held India fully responsible for the breakup of Old Pakistan and the emergence of Bangladesh. West Pakistan (henceforth Pakistan) forged a new Pakistani national psyche, which dramatically reinforced India as Pakistan’s existential nemesis. Pakistan forthwith embarked on its project to acquire nuclear weapons. The nuclear weapons project was easily accelerated in the 80s because the West was happy to look the other way. In exchange, Pakistan channeled Western money to Jihadist forces, among others, to overthrow the Soviet backed regime in Afghanistan. The channeling of vast amounts of money through the ISI made it much more powerful, even as it was radicalized both by its common cause with Jihadist forces and by the Islamization under General Zia.

The end of the 80s saw Pakistan’s acquisition of nuclear weapons, and the successful conclusion of anti-Soviet action in Afghanistan by the alliance of the West, the Jihadist forces among others, and Pakistan’s ISI. The former gave Pakistan carte blanche to pursue any military action against India short of a frontal assault; the latter placed at Pakistan’s disposal, a considerable Jihadist force and the bureaucratic / political structure (ISI) to manage these forces. Pakistan decided to capitalize on the sense of grievance in the Muslim majority state of Kashmir over two instances of opportunistic behavior by the Congress party earlier in the 80s (the Congress had behaved similarly in most other Indian states). Accordingly, it set up extensive terrorist training camps in Pakistani Kashmir, and unleashed a violent, partially Jihadist insurgency in Indian Kashmir. One of the first actions of this insurgency was to carry out an ethnic cleansing of Hindus from the Kashmir valley. This insurgency has continued since, claiming on average three thousand lives every year. Pakistan’s usual rationalization for supporting this partially Jihadi insurgency is that it helps defend Pakistan by keeping a large part of the Indian army tied down in Kashmir. This reasoning is patently absurd: India could retain territorial control over Kashmir during any hostilities with a fraction of the army it currently needs to maintain law and order. Besides, Pakistan’s nuclear deterrent should forestall any hostile intentions India may ever have had. Meanwhile, among the various groups that Pakistan supported to a greater or lesser extent, the Taliban not only consolidated its hold over most of a fractious Afghanistan but also became a reliable ally of Pakistan until 9/11 shook things up a little.

The past two decades in Pakistan have been characterized by an unstable political equilibrium. There are three important political factions: The Bhutto party with its base in Sindh, the Sharif party with its base in Punjab, and the military with its base in a combination of raw power, and popular dissatisfaction with the extreme corruption in the other two political factions. Neither the Bhutto nor the Sharif faction has compromised on their core interest of corruption: neither party has indicated an interest in good governance as a means of staying in power. The military, though more interested in good governance than the others, has not compromised on its core interest of absolute power – has not accepted any power-sharing as a means of co-opting ambitious civilians. All three factions share the national consensus about supporting the violent, partially Jihadist insurgency in Kashmir. All three have a similar approach to the Taliban, namely, that they are a core ally of Pakistan who must be protected even as Pakistan needs to guard itself against some of their excesses. All three also approve of groups, which carry out bombings of movie houses, markets, commuter trains, and the like in major Indian cities every few months. The most “spectacular” of these were the attacks on various targets in Bombay including the Taj Mahal hotel and the Victoria Terminus railway station.

For those inclined to deem this a rather harsh assessment of Pakistani political players, consider the following sample of responses (largely from moderate parties or moderate military figures at a time when Pakistan had great external pressures and incentives to act moderately:

  • General Musharraf cut deals with the Jihadi forces by ceding parts of FATA to them and allowed other Jihadi groups like LET full freedom to operate in Pakistan; all this despite the fact that these groups supposedly tried to assassinate Musharraf himself on at least three different occasions.
  • When Ms. Bhutto was assassinated by Jihadi groups, her party directed blame at elements in the government allied to those groups rather than at the groups themselves. Part of this was because it was more expedient to blame the government (governing party) during an election campaign. But part of it is also because of the reluctance to openly criticize Jihadi groups. This latter construction is reinforced as Ms. Bhutto’s widower recently concluded a deal with the Taliban in which the government ceded the Swat valley to the Taliban. The Taliban has promptly taken control of the lucrative emerald mining operations in the area.
  • For several weeks after the Indian government released the names and addresses of the Bombay attackers, the Pakistani government refused to admit that any of the attackers was Pakistani. Indeed, Mr. Sharif was raked over the coals when, within a couple of weeks, he acknowledged that the captured attacker at least was Pakistani. Along similar lines, in the wake of the recent attack on the Sri Lanka cricket team, a Pakistani minister promptly blamed the Indian government for carrying out the attacks. The MO of the Pakistani political establishment is always the same: while they vehemently condemn terrorist attacks (thereby giving the appearance of sanity), they rarely attribute it to any group (other than the Indian government). This lack of attribution preempts any obligation to publicly condemn the Jihadi groups.
  • When faced with Jihadi groups induced instability: increased suicide bombings in Pakistan including in Punjab, the tenseness surrounding the attacks on Bombay, the attack on the Sri Lanka cricket team, the ceding of Swat to the Taliban after a vicious campaign including wide spread school burnings, and the like crises, the Bhutto faction responded not by trying to undermine the Jihadi groups, but rather by trying to undermine the Sharif faction. Accordingly, they have packed the courts with partisans, and recently both Mr. Sharif, and his brother (the governor of Punjab), were disqualified from running for office. The Bhutto faction backed down only after massive street protests threatened its own hold on power.

The foregoing represents what we can expect from the Pakistani political establishment under any modification of business as usual. The reason for this is simple if rather uncomfortable, namely, that the central and defining element of what it means to be Pakistani is a hatred of India. This is not to suggest that every single Pakistani Muslim hates India (only an overwhelming majority do), or that it is a consuming passion (most have more immediate and concrete matters to preoccupy them). Rather, the view that India is Pakistan’s existential nemesis is assumed to be a self-evident truth. This has made hatred of India the preeminent national value. Establishing extreme anti-Indian bona fides has become a surefire shortcut to legitimation for any group whatever. Any Jihadi group knows that it only has to establish its anti-Indian bona fides (by fighting in Kashmir, bombing Indian civilians, or in any other way) to have carte blanche to do anything in Pakistan (including killing Pakistanis). As long as the extreme anti-Indian sentiment and the legitimation shortcut exists, and it will fall under any business as usual scenario, Pakistani Jihadi groups will be ineradicable.

But is the current unstable equilibrium satisfactory for the primary stakeholders? I will focus only on Pakistan, America, and India (Afghanistan is powerless to effect any change, so I will ignore it). All three stakeholders have grounds for dissatisfaction. Pakistanis are dealing with a deteriorating and unstable equilibrium. Extremists have struck in the heart of Punjab, the Pakistani army is often engaged on the front at the frontier, and Pakistan has become the center of much unpleasant international attention. America has sunk a lot of money and prestige in its Afghanistan operation, and would dearly like to proclaim it a success. There is also a significant if quite minor security angle to the American operation. India has become more integrated into the global economy. Spectacular attacks like those on international hotels in Bombay are likely to jeopardize India’s status as a business and investment destination much more than the mundane attacks on commuter trains and movie houses.

Yet it may be more than arguable that the unstable equilibrium is preferable to the unknown consequences of any action that would disrupt business as usual. It cannot hurt to remind ourselves of the different responses of India and America to comparable attacks. India has largely taken them in its stride without making a big fuss. As a consequence, it has continued to maintain an 8% growth rate even as several hundred people are killed every year in bombings, and a few thousand in Kashmir every year (the latter are primarily Kashmiris unsympathetic to the Indian cause). A severe disruption of the status quo could have the highest cost for India since it shares a long and at places an unmanageable border with a nuclear Pakistan (think dirty bombs), not to mention that its important cities are within range of Pakistani missiles.. Gaurav has suggested that India can reduce the adverse effects of spectacular attacks on India’s international image as an investment destination by a few careful media regulations. Starving western (and probably Indian) media of video feeds would turn these from “spectacular” attacks to mundane deaths and killings worth only perfunctory coverage.

America has already paid the price for disturbing a stable if unpalatable equilibrium in Iraq. Indeed the primary victory of the 9/11 attackers was in knocking America off-balance psychologically. What should have been no more than a $200 billion loss has turned into a multitrillion loss (the worth of the lives lost in the 9/11 attacks was only about $20 billion assuming the rather generous $6 million per life used by the government in its usual cost-benefit analysis). America could settle for maintaining a nominal, internationally recognized client regime in Kabul. America could check Jihadi forces by retaining the right to bomb targets of interest all over Afghanistan. Under such an arrangement, even if the Taliban held parts of the country (non-hostile warlords would hold most of the rest), they would not be much more of a threat to American security than they are now. In the alternative, America could actually try to win hearts and minds by helping the people of the area. No one in the region save the Jihadi forces regard America as an existential nemesis: indeed, America has been a friend of several Afghan groups in the past and present, and has been Pakistan’s most loyal ally for decades. Unfortunately, this is not a practical solution since America will need to spend significant money in helping the region’s peoples. The American people and foreign establishment have shown themselves far more willing to spend $100 billion on killing and occupying a people and their lands (it is called “supporting the troops”) than spend $1 billion on helping those people.

Pakistan has little to lose from the unstable equilibrium. It can revel in its national identity as the lands of the subcontinent’s Muslims by nurturing extreme Islamic groups, and by continuing to be a thorn in India’s side by supporting a partially Jihadi insurgency in Kashmir, and terrorist attacks by Jihadi groups in India. Pakistan will not need to acknowledge its role in the slaughter of millions in Bangladesh, fulfill its duties as a nuclear-armed nation, or face the consequences of exporting terrorism to other countries.

But what if the unstable equilibrium ceases to be an equilibrium, and becomes a slippery slope where the Jihadi forces have a real shot at getting power in Islamabad? Though still very improbable, this scenario is not as outlandish as it was even a couple of years ago. If Pakistan were not nuclear armed, the world might have allowed the struggle between fundamentalist and moderate groups to play itself out over the customary few decades. However, its nuclear status ensures that the world will not leave Pakistan alone if there is a threat of a fundamentalist takeover. Such is the price to be paid for becoming a significant power! With this in view, I will discuss some of the measures that can be taken by outside players (America and the international community) to forestall this outcome. It should be understood that these measures should be considered only when the equilibrium is seriously threatened, or if there is a very high chance of the measure succeeding. The equilibrium persists because Islamic fundamentalists have never garnered significant support in the key provinces of Punjab and Sindh, which contain close to 80% of Pakistan’s population. It will not be seriously threatened unless extremist parties gain significant support in Punjab. Currently, they hold around 1% of the seats in Punjab, and received just over 2% of the nationwide vote. However, we must remember that the government of NWFP surrendered Swat to the Taliban despite the extremists winning just over 10% of the seats and votes in NWFP. Thus, “significant” support which should trigger the softer option means anything approaching a double-digit share of the vote, and “substantial” support which should trigger the hard-line option means a vote share well short of the 29% that gave the PMLN a near majority in the Punjab.

The softer option is to exert extreme economic pressure on Pakistan to achieve definite outcomes. The economic conditions must include generous aid on the one hand, and extreme sanctions including extreme trade sanctions and denial of practical access to international financial institutions on the other. The outcome sought should amount to no less than a revamping of Pakistan’s identity. This would include among others: taking strong action against all terrorist groups, surrendering all listed government officials to an international court to face charges of supporting terrorism (the names of a few ISI officials submitted by America to the UN can be the starting point for such a list), a settlement of all outstanding issues with India by a date certain (de facto acceptance of the status quo), and a revamping of the education system to exclude not only extreme Islamism, but also extreme anti-Indian sentiment. Pakistan has a recent record of changing the course of governments through popular pressure. A sufficiently tight squeeze on all sectors of society couple with generous incentives and an effective public campaign could well deliver even such a radical shakeup. These would doubtless be seen by Pakistan as humiliating conditions, and an assault on its national identity. But one must also remember that the right to support and orchestrate terrorist attacks in India has been seen by Pakistan as a nonnegotiable part of its national identity. A very mild version of this option is seen in the Obama administration making significant aid to Pakistan conditional on Pakistan stopping its support of groups, which carry out terrorist attacks in India. Even this version, which only asks Pakistan to stop supporting terrorist groups (as distinct from taking action against them), has become quite controversial.

The more hard-line option involves a redrawing of the regional map, and a dissolution of Pakistan. The redrawing would create ethnic nations in the area. Thus, Punjab, Sindh, and most of Baluchistan would become their own nations. Northeast Baluchistan, NWFP, and FATA would be joined with southern and eastern Afghanistan to form a Pashtunistan. The rest of Afghanistan would form its own nation, or choose to be absorbed into neighboring nations like Tajikistan, Uzbekistan or Iran, based on regional demographics. There would have to be a solution for the Pakistani regions of AK and FANA, which will not ally them with Punjab. The military action would need to be NATO based, and exclude India to avoid a nuclear incident. Naturally, this solution should be implemented only when the extremists gain substantial support in Punjab.

Though this is a high risk strategy, there are a couple of reasons why it is likely to deliver the best long-term outcome for all the peoples involved. The two most prominent recent examples of failed states falling to Islamic extremists are Afghanistan and Somalia. Both were riven by ethnic or clan factions. Since the primary role of traditional governments is to reallocate resources, a factionalized nation is much more likely to be unstable due to competition among the factions for a larger share. Islamic governments hold out not only the prospect of religion as a unifying force amid the actions; they proclaim that their primary role is the establishment of an Islamic way of life. This is what made the Islamic fundamentalist regimes (the one in Somalia was not particularly intolerant) remarkably stable until they were overthrown by foreign invasions. Pakistan also has strong regional factions, which would find an Islamic fundamentalist regime attractive in difficult circumstances, and find self-determination attractive if it can be easily achieved. As mentioned before, Islamic fundamentalism has a greater draw for Pakistan, which sees itself as the lands of the subcontinent’s Muslims, and which sees largely Hindu India as its existential nemesis.

I have mentioned before how the visceral anti-Indian sentiment among Pakistani people gives any groups that attack India and Indians a shortcut to legitimation. This visceral anti-Indian sentiment is unlikely to survive a dissolution of Pakistan. The largest post-dissolution nation would be Punjab with a population around 8% of the Indian population, and an area barely a quarter of the current Pakistan. The fiction that this nation comprises the lands of the subcontinent’s Muslims would be even more difficult to maintain. Any credible rivalry with India will be equally unrealistic. The new nations would likely settle into a pattern of peaceful coexistence similar to Bangladesh. Like Bangladesh they will derive their national identity from internal cultural history rather than by positing an external nemesis. Western hopes about Pakistan have always rested on the hypothesis that the moderate peoples of the Punjab and Sindh can be leveraged to defeat the Jihadi forces in the frontier regions. However, given the virulent anti-Indian (and increasingly anti-American) sentiments, the Jihadi forces have found it much easier to manipulate the sentiments and resources of Punjab and Sindh. There is no chance that this situation will change organically in the next few decades. Dissolution of Pakistan will starve the Jihadi forces of the logistical, human and technological resources of Punjab and Sindh. They will therefore become much less dangerous, and their smaller inconsequential state will be easier to manage. The moderateness of the people of Punjab and Sindh, when freed from the polarizing influence of anti-Indian sentiment in a post-dissolution situation, will likely create modern secular nations concerned with economic progress.

The three broad options I have laid out here: the status quo, the softer option of squeezing the Pakistani people to deliver an overhaul of their institutions and identity, and the hard-line option of the redrawing of the regional map are not necessarily preferable in a descending order. Indeed, the probable outcomes would make them preferable in an ascending order. It is primarily my risk-averseness and cost-benefit analysis approach to civilian casualties that makes me rank them as I do. It is important to understand that if Pakistan starts sliding down a slippery slope, the mild half-hearted policies that have marked the international approach to Pakistan hitherto will be dangerously ineffective. The options I have outlined here are among the more credible approaches to avoid the serious consequences of a nuclear Pakistan turning into a failed state. Policy makers must have clear thresholds for deploying these options, based on the extent of support for extremists in Punjab. Until then we will continue to see the half-hearted pressure for superficial results, which will not address the need for the overhaul of the Pakistani institutions and identity that make it susceptible to Jihadist forces.

The Jury is Out: Correctness of Democratic Decisions

18 Mar

Rousseau saw majority preferences as an expression of the general will (The Social Contract). With Condorcet, the general will was also imbued with the notion of correctness. As contradictory evidence is ample, it is time to remove the false comfort of any epistemic benefit accruing from such aggregation.

Condorcet Jury Theorem, originally postulated by Marquis de Condorcet, formalized by Duncan Black, runs as follows:

If,

  1. The jury has to decide between two options using simple majority as the decision rule
  2. If each juror’s probability of being correct is greater than half (~ competence)
  3. Each juror has an equal probability of being correct (~ homogeneity)
  4. Each juror votes independently (~ independence)

Then,

  1. Any jury of more than one juror is more likely to arrive at the correct answer than any single juror
  2. As n increases, the probability of arriving at the correct answer approaches 1

There have been multiple attempts at generalizing Condorcet, mostly by showing that violations to one or more of the assumptions don’t automatically doom the possibility of achieving an epistemically superior outcome. 

One way to summarize the theorem is that math works to the extent the assumptions hold. And the assumptions often do not hold.

Applying Condorcet’s Jury theorem to Electoral Democracy

To apply CJT to democracy, we must assume citizenry to be a jury and the decision task in front of it as choosing the “right” party or candidate.

The word jury is saddled with association with courts in the American context, and it is important to disambiguate how the citizenry differs from a jury of citizens summoned by the court. The disambiguation will allow us to cover key issues that affect the epistemic utility of any “aggregations” of human beings.

In the court system, a jury is randomly (~ within certain guidelines) selected from the community. It is generally subject to a battery of voir dire questions so as to assess their independence, lack of conflict of interest, biases, etc. It is sworn to render a “rational” and “impartial” verdict. The jury is instructed in the applicable law, including evidentiary law. And members of the jury are asked not to learn about the case from any other source other than what is presented within the court, which itself is subjected to reasonably stringent evidentiary guidelines. The jury is also guarded against undue influence, for example, bribes by interested parties. The jury is also made to at least sit through extensive presentations from ‘both sides’ and their rebuttals and generally asked to deliberate the evidence among what is generally a ‘diverse’ pool before reaching a verdict.

On the other hand, citizens that come out to vote are a self-selected sample (roughly half of the total body), highly and admissibly ‘non-independent in how they look at the evidence, generally sworn to ‘parties,’ unconstrained by law on what evidence to look at, and how to look at it, generally extensively manipulated by interested ‘parties,’ rarely informed about the ‘basis,’ rarely arriving at a decision after learning about arguments by ‘both sides,’ and rarely ever deliberating.

The comparison provides a rough template for the argument against positive comparisons between the epistemic competence of juries and that of the citizenry. However, Condorcet’s argument is a bit different—though many of the above lessons apply—and hinges on the enormous n in a democracy. The only other assumption that one then needs is each juror having more than 1/2 chance of having it right, or some variation thereof. The key claims that can be made against Condorcet can come from two sources: 1) theorization of the sources and extent of violation of the assumptions, for example, independence and competence, and 2) inapplicability due to incongruence, etc. The various contentions emerging from the two sources are covered below in no particular order.

Rational Voting, Sincere Voting

While it is one of the weaker cases against applying Condorcet – mostly because the counterargument imagines a ‘rational’ voter – the argument deserves some attention – mostly because of its salience in the political science literature. One of the axioms of political science, since Downs, has been that information acquisition is costly. Hence it follows that as the decision-making body becomes larger, and as the chance to be a ‘pivotal voter’ goes down, the incentives to shirk (free-ride) increase.

Austen-Smith and Banks, among others, have shown that ‘sincere voting’, voting the best choice based on the information signal, is not equilibrium behavior as rational voter votes not only based on the signal but also on the chance of being pivotal. (1998, APSR), taking the claim (perils of strategic voting) to its logical extreme and applying it to ‘unanimity rule’ (not majority rule, though similar less stark contentions apply, which they note) show that as jury size increases, the probability of convicting the innocent increases.

Extreme Non-independence

Given p > half is a ‘reasonably high’ threshold – jurors performing better than random – especially in circumstances of misinformation, problems can arise quickly.

In the current state, about 90% of the voters exhibit high forms of non-independence emerging from apathy and partisanship. It also reasons that reduction in either one will lead to a higher probability of citizenry choosing the ‘better choice’ on offer and arguably better choices on offer. Partisanship also means that people have different utilities that they intend to maximize. The other 10% err on the side of manipulation.

What’s on the Menu?

To the extent there are two inferior choices to choose from, one can imagine that in the best case, the polity will choose the slightly better one among the two inferior choices. Condorcet offers no comfort for what kind of choices are on offer – perhaps the central and pivotal role of any normative conception of democracy. In fact, it is likely that the quality of choices on offer (‘correctness of choices’) is likely to be a function of probability with which a body politic knows about the ‘optimal correct choice’, and the probability that it chooses the ‘optimally correct’ choice (which is likely to be collinear with odds of picking the ‘better choice’).

Policy Choices

Policy choices are an array of infinite counterfactuals. To choose the ‘most correct,’ one would need a population informed enough to disinter the right choice with a higher probability than any other wrong choice. Given infinite choices, the bar set for each citizen is very high. The chances of current citizenry crossing that bar—non-existent.

Three or More Choices

The manipulability of the system offering more than two choices is well documented and filed alternately as Condorcet’s Paradox and Arrow’s impossibility theorem. Much work has been done to show that propensity of cycles in a democracy is not great. (For example, Gerry Mackie, ‘Democracy Defended’) One contention, however, remains unanswered for the binary choice version -American Democracy often reduces larger sets into two options. One can imagine that the preference order for citizens will depend on unoffered choices. Depending on how multiple choices are reduced to two choices, one can think of ways ‘cycling’ can work even in the offered binary choices. (David Austen Smith) More succinctly – all binary decisions in democratic politics can be thought to come from larger option sets, and the threat of cycling hence is omnipresent.

Correct Decisions

CJT, a trivial result from probability, when applied to voting with two choices, is just that an electorate is most likely to arrive at the more likely choice of each of its members. The probability of achieving that comes close to 1 as n increases.

If we assume that electoral democracy is a competition between interests, then we just get majoritarian opinions, not ‘correct’ answers. As in there is no common utility function but a different set of utilities for different groups – so people look at a common information signal and split based on their group interests. In that case, the ‘correctness of the decision really reduces to the ‘winning’ decision.

Median Voter: Condorcet in Reverse

In many ways, applying Condorcet to democracy is applying things in reverse. We know that politicians create policies that appeal to the ‘median voter’ (as opposed to the median citizen). Politicians work to cobble together a ‘majority’ such that the probability of the majority picking them is the greatest. Significantly, policy preferences that can be sold to the majority have no similar claims as made by CJT. Another important conclusion that can be drawn from the above is that since the options on offer can manipulate the population, it is likely that the errors are not at random.

Democratic Errors Don’t Cancel

In The Rational Public, Benjamin Page and Bob Shapiro argue that one of the benefits of aggregation is that the errors cancel out. Errors may cancel if they are random, but if they are heteroskedastic and strongly predicted by sociodemographics, they are likely to have political consequences. For example, such errors very plausibly reduce the likelihood of certain people making a demand or from coalescing to make demands in line with their interests.

Formation of Preferences, Aggregation of Preferences

Applying CJT to democracy, we can roughly proxy what preferences will emerge from available data. Assuming people have a perfect lens to the hazy data, the “probability that the correct alternative will win under majority voting converges to the probability that the body of evidence is not misleading.” (Franz Dietrich and Christian List in ‘A Model of Jury Decisions Where All Jurors Have the Same Evidence’)

While even the probability calculated thence is optimistic— we know that evidence isn’t the same for all jurors, and the lens of most jurors is foggy—it is a good start to thinking about what data is available to the jurors, how it is used by the jurors (citizens), what the consequences are of different information and ‘analytic lens’ distributions.

Letting the Experts Speak

If our interest is limited to getting the ‘correct outcome,’ then we ought to do better (in terms of likelihood of arriving at the correct decision) by polling people with higher probabilities of getting it right. We will also save on resources. Another version of the idea would be to do a weighted poll, with weights proportional to the probability of being correct. The optimal strategy is to have weights proportional to log p(correct)/p(incorrect) (Nitzan and Paroush, 1982; Shapley and Grofman, 1984).

It isn’t as much a contention as a prelude to the following conclusion – Any serious engagement with epistemic worthiness as a prime motive in governance will probably mean serious adjustments to the shape and nature of democracy and, in all likelihood, abandonment of mass democracy.

60% is Different from 51%

The key consideration in CJT is choosing the ‘right’ option from the two on offer. Under this system, 51% doesn’t quite differ from 60% or 90% for all yield the same ‘right choice.’ Politics works differently. Presidents tout their ‘mandates’ and base their policy agendas on them. Congress and Senate have a slew of procedural and legislative rules that buckle under larger numbers. Thinking about Congress and Senate brings new complications, and here’s why – while the election of each member may be justified by CJT, the benefit produced by elected representatives needs another round of aggregation – without some of the large n benefits of mass democracy. Here again, we may note – as McCarty and Poole have shown – that the ‘jury’ is extremely ‘non-independent,’ prone to systematic biases, etc. In addition, no longer is choice limited to two – though each choice task can be broken down into a series of Boolean decisions (arriving at the ‘right decision’ in this kind of linear aggregation over choice spectrum will follow a complex function of p(correct choice) for each binary decision.

Summary

Conjectures about the epistemic utility of electoral democracy are rife with problems when seen through the lens of Condorcet. This isn’t to say that no such benefits exist but that alternate frameworks are needed to understand those benefits.

‘Disproving’ God

24 Jan

A tribute to Charles Darwin on his 200th birth anniversary

Some say that science should be absent from discussions about God is because while science concerns itself with the material, God concerns himself with the ‘spiritual’ ‘nonmaterial’ realm. But there are a number of theories of God that explicitly deal with the material realm. To that extent science has standing.

One prominent theory of the divine made with ungodly frequency is that God has a direct impact on the material well-being of humans. Here’s why the claim falls short:

Given systematic temporal and geographic variance exists in poverty, life-expectancy, etc. and given we have been able to attribute a majority of the causes for such to human action, it appears God plays only a peripheral part in the destiny of man, albeit a larger role in the destiny of women (mostly through the hands of believers). Mortality rates differ by geography. For example, the US versus Africa, Northeast US versus Southern US (maybe god likes ‘godless NE liberals’). And mortality rates vary by time—we live longer today than we did 200 years ago. The variance in mortality and life-expectancy also seems to respond to human intervention—discovery of new technologies and medicines, war, etc. So unless we believe God systematically dislikes Africans or liked people born 200 years ago less, when arguably people were more moral on some variables favored by the current fundamentalists, we have little reason to believe that God is a large force in determining life-expectancy, or mortality.

Let’s assume for devil’s sake that God is a confounding variable, which is true in at least one way—he is alleged to work in mysterious ways, i.e. God determines both temporal, geographic, racial and other kinds of variance in distribution of poverty, the human action to which it is causally attributed, and life-expectancy. But that version of God conflicts with our theories about human action (say greed) and our theories about God, who ought not to reward people motivated by such things as greed. But then punishment can come in the afterlife – via hell, where an ever larger number of people are being systematically tortured through a great expense of energy, and in a manner that will leave the Bush administration officials chagrined. Even if we imagine that theories of after-life action are true, their impact on the material world is limited to the extent people believe in the threat of punishment. To that extent, God is an instrumental identity for achieving some version of morality.

Another challenge to the presence of God comes from the probabilistic nature of our causal models. Theories about God ought to be perfect (explain about 100% of the variance) while theories of social action can be probabilistic. To ascribe probabilistic thinking and action to God would significantly conflict with theories of God, though one can imagine that he sets the mean, and will cause the error term. A starker version of the same would be that God allows free will, and to that degree that he allows for it and the world is shaped by free will, and God is immaterial to bettering social condition. Another reason to discount challenge to God theory can be the following – we just don’t know the generating mechanism (or life/death) and probabilistic conditioning seems to come from fitting known world models onto data generated by God model, which is by the way synergistic enough with the world model (more poverty = earlier death) to be disturbing.

One way to look at the argument presented here is that God may exist, but s/he/it isn’t particularly strong. And if strength/omnipotence is taken to be a fundamental descriptive attitude of the object (God), it is likely that the object doesn’t exist as well. The counterargument to the above would perhaps need to factor in differing conceptions about the object and its power. For example, one may say that s/he/it is doing all it can to reduce evil to its lowest form – and that is indeed the present condition. Perhaps then more minimally – since he is already doing all he can and rest depends on us – we can argue that God isn’t a particularly useful intervention for changing one’s situation in the world.

Understanding Delhi and the Delhi Walla

8 Jan

The article was written for The Delhi Walla

The Delhi Walla is a journalist’s blog, albeit without the drama and urgency with which journalism and journalists are often associated with today. The writing on the blog represents that prior tradition among journalists which was about subtle observation, gentle humor, as evinced in journalists’ travelogues, and in shows like BBC’s “From our own correspondent.”

The blog is a significant achievement. More so because reporting on cities is generally skillfully and purposefully bankrupt, formulaic and inane, an orgy of crummy descriptions of pointless people, and events, and soulless corporate jingles about places to eat, and entertain, infested almost always with a touch too colorful poorly shot photos.

With an eclectic choice of topics, a choice that is many a times dictated by the city rather than by an urge to puppeteer description in grips of pincers of prejudice, with gentle and subtle humor, Mayank shines a weak but almost always pleasant humanistic light on the myriad facets of Delhi, and the occupations, preoccupations, habits, of its residents. The wonderful aspect of the blog is that it catalogs ‘real life,’ an all too absent commodity in newspapers, be it then a story about the need to find a second home in a city with cramped homes that provide all too little privacy, the rather oddly structured stories on colonies (as they are called in Delhi), or the succession of charming articles on bookstores, and their proprietors. Perhaps seen hence, it is a writer’s blog. And that is probably a more accurate description of the sensibility of the blog, and the author, and explains the void comparisons to newspapers that I make above.

Understanding
One can try to understand things of interest by disinterring things, breaking them apart skillfully, through analysis and connecting those parts into an explanation or simply description conjoined by some connective tissue. It is a bit like looking at white light through a prism, with colored rainbow being the distillate. Of course, more often we just describe a part of one color, and the rest is at best in the penumbra. Analysis is generally purposive and demands specificity. It struggles to contain, and cast, and organize, and too often the aim is to achieve that aha! moment. For all these reasons, the enterprise is often fraught with problems of myopia, and of force.

Another feature of the analytical method is the method of writing – it is writing through contestation. For example, the account that I provide here is often times a ‘negative’ account — describing what this blog isn’t, rather than simply focusing on what it is. The method may be insightful if the analysis has legs, but it is seldom enjoyable.

The Delhi Walla chooses differently; he observes, describes, narrates, engages in reverie, and gently analyzes. He does it with great modesty and some charm. His method of understanding isn’t analytic introspection, but subtle observation that produces that warm flush of vague but liberalist accepting, even embracing, empathy, and exultation in the shared existence. It is akin to the understanding and exultation one feels while standing on the roof of the house on a pleasant summer evening, and looking over the gullis and Mohalla.

Delhi
Delhi is an easy city to caricature. It is bleak, dirty, loud, and crowded. And it is certainly all that. But the reality is simultaneously substantially more mundane and textured. Likewise, people sometimes mistakenly make the inferential leap from bleak surroundings to bleak lives; all too often bleak surroundings are peripheral to the fuller psychological lives lived among acquaintances, friends, relations, and more.

Delhi is a city that carries the hopes and aspirations of people living in it, the location of deaths, marriages, jobs, cars, monuments, history, politics, money, and more. One can take respite, if so is needed, in the beauty of some of its monuments, sometimes in just its familiarity, in its traditions and landmarks, even in its oppressive heat, as Mayank occasionally does, food, conversation, and intimacy of friends and family, among other things.

The Delhi Walla
The Delhi Walla is an eclectic account of Delhi. It is an ode to the passions of Delhi Walla — the Muslim heritage of Delhi, books, Arundhati Roy, and gay life in the city. It is an account of his questions, and more interestingly a live account of an unfailingly interesting life.

On Representation in a Democracy

3 Dec

“Representation means the making present of something that is nevertheless not literally present.” (Pitkin, 1967) In a representative democracy, which, minimally construed, involves a mediating assembly for political decision making, representation, implies an attempt to find the people in assembly’s political decision making.

But what do we mean by that? Should we look for people’s values, thoughts, current (or not so current), strong (or also weak) opinions (“phantom,” as Philip Converse puts it, though they may be?) in governance? And where exactly should we look? Should it be policy, or institutional design, or process, or in the race and gender of representatives? Not to say that all of this rests upon the idea that things like opinions can be coherently expressed by people (in the aggregate), and are cognizable in institutions, policy outcomes, etc.

Other than these seemingly intractable questions of measurement, we also have substantive questions. Who is represented, and to what degree, and why? And we must struggle with some normative questions that lie adjacent to the empirically posed question above. Who/what should be represented, and to what degree?

Origins of our thinking about representation

There are two intersecting facets of how representation is understood, and perhaps should be. One is highlighting the cultural construction of concern with representation, and the other is historical understanding of representation.

Partly, idealized notions of representation are built against the inequalities manifest in the economic processes. The need for political equality/representation is a necessary counterpart to the society that has salient economic inequalities built on the mythology that anything is possible for everyone.

Some of our understanding of representatives and control by people, constrained as they are by social norms of the Congress or bargaining between President and Congress, ought to be shaped by historical and normative conception. Historical foundations of the current form of representation in the US can be traced to at least Madison. As is commonly surmised, elite deliberation as a model for representation was developed against the fear of the mob. True, but there was much positive thought guiding Madison’s idea of a modern democracy, and representation. It wasn’t just that unconstrained mass democracy is unsuitable, or the larger logistics based argument that mass democracies are untenable, Madison’s claim was that the desired effect of (elite) political representation is “to refine and enlarge the public views, by passing them through the medium of a chosen body of citizens, whose wisdom may best discern the true interest of their country.” Of course to what degree he succeeded in that ideal is open to conjecture, if not open derision.

Partly we also get an understanding of what is to be represented by the prominent instruments that key institutions provide people to express or control their representatives. If representation implies the extent to which political leader acts in accordance with wants and need and demands on the public, then we ought to look into how the public can express its needs, and how those are funneled in the political process. Let’s take, for example, vote. We know for a fact that vote itself is a poor instrument for expressing multifaceted preferences. A vote is binary, or at best trichotomous. So typically the role of a citizen was conceived to be relatively minimal, at least on a per capita basis.

But by constraining ourselves to a discussion about voting, arguably the single most potent symbol of democracy, we fail to fully understand the ability and opportunity provided by democratic governance systems.

More simply, not all representation is via representatives. Democratic governance systems provide multiple ways to shape the public’s agenda, shape public opinion on the agenda, and how it is fed into governance. It provides multiple modes (lobbying, media, etc.), multiple institutional entry points (courts, legislature, public hearings of executive branches etc.), multiple temporal entry points (at the crafting of law, or as its failings are exposed, or in restricting its application “prerogative of the executive branch etc.), through communication of dissent, and consent, and hence allows for representation in many different ways within the many different institutional frameworks.

Measurement: Who is represented?

One way to assess who is represented is to merely track the economic well being of various groups over time. Another would be to correlate opinion/policy of representatives with that of the opinion of the constituency. Given politicians often actively shape opinion, and the problems with using correlation (as highlighted by Christopher Achen), the measure is largely doomed. In addition, any such measure ought to incorporate the agendas of people (problematic to measure), and their opinion on those agenda items. In other words, we ought to measure two things: are issues considered important by people/constituents considered similarly important by the representatives and the ‘correlation’ in opinion on those issues. In absence of similar agenda priorities, the question about agenda would be hard to measure. And certainly, concerns about strategic/manipulative agenda setting by politicians (Page and Shapiro recently came out with a book “Politicians Don’t Pander” that gives this worry some legs) would be of import here as well.

What should be represented?

The answer to the question is murky. Clearly, multiple things need to be represented. For example, say a policy has a disproportionally negative impact on a small group of people—their concerns perhaps ought to be represented. It is inarguable that the representation structure somehow constrains what is to be represented, depending on how widespread the cognition of its impact is – and to what salience. Part of our answer to what ought to be represented depends on our conception of democracy. So if governance is at heart about the allocation of meager resources, and it is certainly at least about that, then does ‘representation’ of one’s interests (hard to define) at the bargaining table as interest groups (or mobilized segments of society) present their cases the ideal?

If we minimally understand people’s wants as interest in better outcomes and assume that better outcomes emerge from good information, we can perhaps then focus on the representation of (all) information, be it the differential impact of certain policies, or some innovative technique.

Role of a representative

Heinz Eulau et al. present two models of thinking about the role of a representative: 1) Who is being represented (district/state)—this needs to be further disinterred, and 2) how—trustee, delegate, politico or hybrid? These axes are a small but essential kernel of a theory of a representative. Yet it would do us a disservice if we think that representatives do one or the other, on any of the dimensions. To a very significant degree, the institutional mechanisms have evolved to dole out the pork (district) and deal with say national issues as well, and many a time getting the former accomplished seamlessly as part of the latter. Alternately phrased, it does us a disservice to think of district and state orientation as polar opposites of a continuum. A broad set of policies accommodate both. Similarly, trusteeship needn’t automatically contradict the role of a delegate. The theorization of the role of a representative ought to take into account the fact that given that over a large set of policy issues, the population has minimal (if not phantom) opinions, what is his/her role and responsibilities? Is it opinion leadership or manipulation (again the reference to Page and Shapiro Politicians don’t pander)? The Page and Shapiro version is considerably closer to the dystopic version outlined by Pitkin— mass democracy inevitably fades into fascist manipulation. The argument, differently expressed elsewhere, goes like this – representation in a democracy is best understood not in terms of accurate correspondence between pre-existing citizen preferences and subsequent government decision but rather as a constructive (if working ideally) process that shapes the very same preferences and perspectives that are represented.


Hanna Pitkin. The Concept of Representation. (1967)
Heinz Eulau et al. The Role of the Representative: Some Empirical Observations on the Theory of Edmund Burke. (1959)
Christopher Achen. Measuring Representation. (1978)

Did India’s Economic Miracle Begin in 1980? Why?

19 Nov

Two recent papers (and many previous ones)—From the Hindu Rate of Growth to the Hindu Rate of Reform and From ‘Hindu Growth’ to productivity surge: the mystery of Indian growth transition—present evidence that India’s growth accelerated starting 1979, and not, as is often noted, post-1991. The papers go on to conjecture about possible causes for the same including the green revolution, internal liberalization, etc.

Rodrik and Subramanian posit “that the trigger for India’s economic growth was an attitudinal shift on the part of the national government in 1980 in favor of private business. The rhetoric of the reigning Congress Party until that time had been all about socialism and pro-poor policies. When Indira Gandhi returned to power in 1980, she re-aligned herself politically with the organized private sector and dropped her previous rhetoric. The national government’s attitude towards business went from being outright hostile to supportive. Indira’s switch was further reinforced, in a more explicit manner, by Rajiv Gandhi following his rise to power in 1984.”

The same point is made, albeit in a different language, by Atul Kohli, Professor of Economics at Princeton.

Evidence of Growth in GDP in the 80s:

Average decennial growth rates across countries and regions
Average decennial growth rates across countries and regions

One can easily surmise from the graph that the growth rates in the 1990s (2.5) were twice that of what it was in the 80s. However, the growth in the 80s, compared to past 20 years was again significantly higher.

India's GDP between 1960 and 2007
India
India's GDP growth rates between 1960 and 2007
India
India's per capita GNI (PPP adjusted) between 1960 and 2007
India

The Educated Racial Voter

13 Nov
McCain share of white vote by state, and by region
McCain share of white vote by state, and by region; Source: Exit Poll Data by Edison Mitofsky
Obama's vote share among people with a postgraduate degree by state
Obama's vote share among people with a postgraduate degree by state

More:

“Here in Alabama, where Mr. McCain won 60.4 percent of the vote in his best Southern showing, he had the support of nearly 9 in 10 whites, according to exit polls, a figure comparable to other Southern states.”

“Mr. Obama won in only 44 counties in the Appalachian belt, a stretch of 410 counties that runs from New York to Mississippi.”

“Southern counties that voted more heavily Republican this year than in 2004 tended to be poorer, less educated and whiter, a statistical analysis by The New York Times shows.”

New York Times

Andrew Gelman’s take on the 2008 results

“As with previous Republican candidates, McCain did better among the rich than the poor,.. but the pattern has changed among the highest-income categories…”

The 2008 Presidential Campaign in Brief

13 Nov

November 10, 2007: One of the first scandals to break out during the campaign was about planted questions in Hillary’s town hall meetings. “They asked me if I would ask the senator a question. I said, ‘Sure, you know,'” Gallo-Chasanoff told CNN. “He showed me in his binder, he had a piece of paper that had typed out questions on it. And the top one was planned specifically for a college student. It said ‘college student.'” ‘A video on MSNBC shows Gallo-Chasanoff reading the question word for word, and then winking when she was done.’ ABC News

November 10, 2007: “I love my wife and my five sons and their five wives. Wait a second. Let me clarify that. They each have one.” Mitt Romney (Economist gave this quip the title – Best Freudian slip; ABCNews.com)

December 12, 2007: In kindergarten, Senator Obama wrote an essay titled ‘I Want to Become President.’ “Iis Darmawan, 63, Senator Obama’s kindergarten teacher, remembers him as an exceptionally tall and curly haired child who quickly picked up the local language and had sharp math skills. He wrote an essay titled, ‘I Want To Become President,’ the teacher said.”
From: Clinton campaign’s press release.

December 13, 2007: “It’ll be, ‘When was the last time? Did you ever give drugs to anyone? Did you sell them to anyone?'” Shaheen on Obama
Bill Shaheen (husband of NH Senator-elect Jeanne Shaheen; national co-chairman of Clinton’s campaign at that point)

February 24, 2008: Bill Clinton speaking about Hillary’s inability to win caucus states – “the caucuses aren’t good for her. They disproportionately favor upper-income voters who, who, don’t really need a president but feel like they need a change.” Audacity of Hopelessness by Frank Rich

March 8, 2008: “She is a monster, too – that is off the record – she is stooping to anything,” Samantha Power; Obama’s foreign policy adviser.

March 10, 2008: Hillary Clinton chief spokesman Howard Wolfson declared Monday that Clinton does not consider Obama qualified to be vice president.

March 11, 2008: “I will not be discriminated against because I’m white.” Geraldine Ferraro

“If we can’t trust Mitt Romney on Ronald Reagan, how can we trust him to lead America?”
From John McCain’s attack ad on Romney

“The Clintons will be there when they need you,” said a Carter friend. (Maureen Dowd, NY Times)

May 3, 2008: When asked, at the Republican presidential primary debate at Simi Valley, whether any of the candidates did not believe in evolution, three candidates – Tancredo, Brownback, and Huckabee – raised their hands.

May 9, 2008: “Senator Obama’s support among working, hard-working Americans, white Americans, is weakening again.” (Hillary Clinton, Interview with USA Today)

Google News - Clinton Accuses Obama
Google News Archives timeline graph of citations of 'Clinton Accuses Obama' between August 2007 and August 2008

August 21, 2008: “I think – I’ll have my staff get to you. It’s condominiums where – I’ll have them get to you.” (John McCain unsure about the number of houses he owns.)

A special tribute to Palin:

September 24, 2008: “As Putin rears his head and comes into the airspace of the United States of America, where– where do they go? It’s Alaska. It’s just right over the border. (Interview with Katie Couric, CBS News)

In defense of Palin, she never said that she could see Russia from her house. (Time)

September 25, 2008: Couric: And when it comes to establishing your worldview, I was curious, what newspapers and magazines did you regularly read before you were tapped for this to stay informed and to understand the world?
Palin: I’ve read most of them, again with a great appreciation for the press, for the media.
Couric: What, specifically?
Palin: Um, all of them, any of them that have been in front of me all these years.
Couric: Can you name a few?
Palin: I have a vast variety of sources where we get our news, too.
CBS News

October 1, 2008: “Well, let’s see. There’s — of course — in the great history of America rulings there have been rulings.” Sarah Palin (When asked by Couric to name a Supreme Court decision, other than Roe vs. Wade, that she disagreed with; CBS News)