Monday, May 25, 2009

Good news

Dr Binayak Sen has been granted bail. About time.

The story is not over yet. I'm sure the state government will do what it can to get him behind bars.

Saturday, May 16, 2009

The (relatively) good guys won

A recent US president said of the opposition: "They misunderestimated me." Manmohan Singh could well have said the same, only in correct English.

In 1991, this soft-spoken gentleman, who had been both an academic and a bureaucrat, was invited by Prime Minister P V Narasimha Rao to be finance minister in his cabinet. He embarked on a programme of economic liberalisation that sparked a boom, which continued through several administrations until the global meltdown of the past year. Nonetheless, liberalisation was not politically popular; Rao lost the next election, in 1996.

In 2004, he was sprung on the public at the last minute, after Sonia Gandhi declined prime ministership despite having led the Congress-led coalition to an unexpected victory. She did not want to court controversy over her citizenship; she believed Singh to be a capable leader; and, very likely, she thought he would be the better prime minister.

In 2009, he was promoted from the start of the Congress campaign as their candidate for prime minister. The public was asked, effectively, to choose between him and L. K. Advani. This is no longer 1996. The public chose Singh, overwhelmingly.

Because of his soft-spoken nature he has attracted charges of being weak, a puppet of Sonia Gandhi, and a temporary substitute for Rahul Gandhi. L K Advani attacked him viciously during the 2009 campaign, but it seems to have backfired. In fact, during his tenure, Singh pushed through the Right to Information Act, which he saw as one of the most important ways to improve transparency; he broke ranks with the Left to push through the nuclear cooperation deal with the US, and personally intervened with other parties (including some quite unsavoury politicians) to ensure that his government continued to have the support to last its full term; and, meanwhile, he let the Gandhis, Sonia and Rahul, focus on rebuilding the Congress party from the grassroots.

Here's a recent article by his former media advisor Sanjaya Baru, who reveals how often Manmohan Singh quietly got his way and how rarely he took credit for it: even in the case of the nuclear deal, he was willing to let the Left take credit for amendments if they supported it! As things turned out, Prakash Karat was obdurate and Singh (with the party's support) decided that the Left's support was dispensable.

The strategy of letting Singh govern the country and Sonia and Rahul Gandhi govern the party has paid off handsomely. Rahul's go-it-alone strategy has revitalised the Congress party in the north; troublesome allies and opponents have been marginalised, the Left is in tatters, and the BJP is on the back foot and unsure of what ideology to present the public. A new youthful generation of Congress leaders seems to be emerging to take over from the next election. And the public has marked their approval of the Congress's governance over the past five years.

Now that the ruling coalition has earned this "political capital", as the Americans call it, they must spend it wisely. Much needs to be reformed, in our economy, our bureaucracy, our educational system, and especially our judiciary. I don't expect miracles but I do expect an effort, especially now that the obstructionist Left are sidelined. And the rewards, for the Congress, could well be an even greater mandate in 2014.

Friday, May 15, 2009

Worrying about vaccines

Being the parent of a toddler means, among other things, giving the kid immunisation shots against various diseases at regular intervals. Understandably, this is not something kids enjoy. Moreover, some vaccines cause temporary reactions, that in rare cases can be quite severe. And while some vaccines are universally administered, others are very much at the discretion of paediatrician (and, in theory, the parents). So one tries to keep well informed.

So I read with interest a recent blog post where Dan Summers, a paediatrician, takes apart the anti-vaccine crowd, and in particular, this article by Jim Carrey.

The anti-vaccine movement is primarily driven by claims that vaccines are linked with childhood autism-spectrum disorders. There are at least two versions of this claim: first, that the MMR (measles-mumps-rubella) combined vaccine is linked with autism (a claim that caused particular scare in the UK); second, that thimerosal, a now-discontinued mercury-based preservative, is linked to autism (a continuing source of scare in the US). Both these are supported, at best, by anecdotal evidence. There are also claims that combined vaccines increase the shock to an immature immune system, leading to possibly undesirable side-effects.

Lots of peer-reviewed studies have, supposedly, refuted links between either MMR or thimerosal, and autism. This article reviews the situation. And only a few fringe loonies would argue that we would be better off without vaccines. It is wonderful to be in a world without smallpox, and to look forward to one without polio. Various once-common childhood killer diseases have now become rarities. Dr Summers says he has never even seen a single case of measles, still a common (if rarely fatal) disease in India. What's not to like about vaccines?

Dr Summers also exposes the bogus nature of Jim Carrey's "supporting" links. Like most conspiracy theorists, Carrey does not link to actual scientific literature, only to paranoid-sounding sites like "Generation Rescue".

Here's the problem though: I find Dr Summers unconvincing not for the things he refutes and exposes, but for the things he ignores in Carrey's article. And if I, a strong skeptic on most "alternative medicine" views, am unconvinced, I am sure millions of others will be even less persuaded.

First of all, Dr Summers says, "Indeed, I have never seen a single case of smallpox, polio, tetanus, measles or a handful of other once-common childhood illnesses..." But Carrey entirely agrees with the need to vaccinate against "the most serious threats including measles and polio"; I'm sure he would agree that tetanus is more serious than measles, and nobody vaccinates against smallpox anymore. So this is a strawman at the outset.

Dr Summers says "the advent of vaccinations has been as close to an unalloyed good as is likely ever to occur." If he merely refers to the eradication of smallpox, the near-eradication of polio, and the success against several other serious diseases, few should argue. But his article suggests something stronger: every approved vaccine is close to an unalloyed good. I think few doctors would agree with that (and he himself does not quite say it, only implies it).

Carrey says that there are 36 vaccines in the current US immunisation schedule; the version I saw only included 15 (as I discuss below), but perhaps he includes optional approved vaccines. He continues that there have been "no tests for cumulative effect or vaccine interaction" of all these vaccines, let alone the 100 or so in development. If there is such a test, Dr Summers doesn't supply a link to one.

Carrey claims that "American children are being given twice as many vaccines on average, compared to the top 30 first world countries". I don't know the true numbers, but Dr Summers doesn't refute this. I will return to this topic below.

Carrey says that "Paul Offit, the vaccine advocate and profiteer, who helped invent a Rotavirus vaccine is said to have paved the way for his own multi-million dollar windfall while serving on the very council that eventually voted his Rotavirus vaccine onto our children's schedule", and quotes strong words on this topic from a Congressional committee. Dr Summers does not address this.

Carrey says that "Veterinarians found out years ago that in many cases they were over-immunizing our pets, a syndrome they call Vaccinosis. It overwhelmed the immune system of the animals, causing myriad physical and neurological disorders." He also says veterinarians removed thimerosal from animal vaccines in 1992, and wonders why human vaccines lagged behind. Summers does not address this (or refute these claims).

Carrey admits that vaccines are good and bad, but says "One thing is certain. We don't know enough to announce that all vaccines are safe!" Summers, and the CDC in the US, seem to be announcing precisely that.

Now, here's the main bothersome issue that Summers entirely ignores: if vaccines were such an "unalloyed good", why do US children reportedly receive twice as many vaccines as children in other developed countries?

In my developing country, India, the National Immunisation Schedule lists the following vaccines up until 6 years: BCG (tuberculosis), OPV (polio), DTP (diphtheria, tetanus, pertussis), Hepatitis B, Hib (Haemophilus influenzae B), Measles, MMR (measles, mumps, rubella), and various boosters for these. That's seven vaccines for ten diseases. (Source: IAP Guidebook on Immunisation, 2009, available here.)

The US schedule that I can find for children below 6, here, lists vaccines for a total of 15 diseases: in addition to the above, there are rotavirus, pneumococcal, influenza, varicella, hepatitis A, and meningococcal (while BCG is missing). The UK schedule lacks several of the US vaccines: namely, rotavirus, influenza, varicella, hepatitis A and B are missing.

Many children in India receive more vaccines than the immunisation schedule recommends, and the same may be true in the developed world. The Indian Academy of Paediatrics recommends typhoid in addition to the ones in the immunisation schedule, and also suggests that pneumococcal, rotaviral, Hepatitis-A and Varicella "can be given after discussion with parents". Needless to say, such discussion rarely occurs.

But why does no country include all 36 vaccines in its schedule, and why are other countries omitting so many vaccines included in the US schedule? Are other countries being perverse and regressive in ignoring the "unalloyed good" provided by all those US vaccines? Or are there other factors to consider?

What are the possible drawbacks of vaccines? One is obvious: cost. This is relevant not only in India but very much also in the US, whose health coverage leaves much to be desired. It is the reason DTaP (the newer form of DPT, with an acellular pertussis vaccine that has milder side-effects) is not yet officially favoured in India. (Vaccines in India's National Immunisation Schedule are delivered free of cost via the public health system.)

Another is side-effects. Most vaccines have side effects; for the smallpox vaccine, one in a thousand have severe rashes, and about one in a hundred thousand had more severe, life-threatening complications. The risk was acceptable compared to the dangers of contracting smallpox when it was widespread. But as smallpox got effectively eradicated, the risk was judged unacceptable and the vaccine is no longer administered.

A third is an unknown and controversial factor: the effect of administering many vaccines simultaneously. More and more frequently, vaccines are "bundled" with five or more often administered in a single shot. Skeptics argue that the effects of this sort of "shock" on an immature and developing immune system are at best questionable. Defenders think there is little danger and an infant, of whatever age, is exposed to many more antigens and possible pathogens every day than what are administered in a vaccine. But few doctors or immunologists, I think would assert that there is no danger whatever.

Unfortunately medical science is not an exact science: despite the most exhaustive clinical trials, certain side-effects, interactions and allergic reactions from drugs can take years or decades to discover.

Unfortunately, also, patient mistrust of doctors and medicine seems to be rising all over the world, and not only are "alternative medicine" therapies of various levels of dubiousness rising, but proven medical therapies and treatments are being ignored by patients who read scare-mongering half-truths. Vaccines are an example: today's young parents have never encountered smallpox, polio, diphtheria, tetanus or many other dangerous diseases, but do encounter the reactions to the vaccines and read frightening but unsubstantiated stories on the internet; therefore many reject vaccines.

But how to tackle this? I think Dr Summers' approach is not the answer.

I know many committed, sincere doctors who are concerned with nothing except the welfare of the patient. I believe this is true of the majority of doctors, around the world (though perhaps not the majority of high-profile, headline-grabbing doctors.) They are aware of the limitations of medical science, are willing to prescribe alternative methods (ayurveda, yoga, acupuncture) when appropriate, but prescribe powerful drugs when required. It is a pity when patients do not trust such advice, and assume that all antibiotics must be bad regardless of circumstance -- or that vaccines for certain diseases are not required because you don't run into those diseases any more.

But patient mistrust is not limited to the doctors: it extends to the pharmaceutical companies. And, here, I think the mistrust is merited.

Carrey cited the example of the rotavirus vaccine, included -- allegedly under the pressure of the developer -- in the US schedule but not in most other countries. But this is only one of their many crimes.

Pharmaceutical firms are routinely accused of focussing their efforts on "lifestyle drugs" that are directly marketed to the public, with minimal consideration of side-effects. See, for example, this article in Nature Medicine.

Infant formula makers used to claim that their products are superior to breastfeeding; when such advertising was banned in most countries, they resorted to more indirect claims that planted the same impression in the minds of the public. (See this article by George Monbiot, for example.)

Dr Summers talks about the importance of peer-reviewed research, and challenges skeptics on vaccines to publish their work in peer-reviewed journals. Unfortunately, pharmaceutical companies have done their best to pervert the scientific peer-review process too. It is mandatory for scientists to disclose sources of funding and potential conflicts of interest in their manuscripts, but this is not always kept above board. In a recent spectacular case of fraud, an anaesthesiologist who had built a significant international reputation turned out to have faked his clinical trials for over 10 years; his research was funded by pharmaceutical companies who produce many of the drugs that his papers promoted. And even more egregiously, Elsevier Science -- publisher of many leading journals in every discipline -- was found to have published fake journals, at the behest of pharmaceutical companies (in particular, Merck), that had every appearance of containing peer-reviewed research but in fact contained in-house research from the pharma giant presented in academic form.

I think this sort of thing hurts medicine and public health much more than a few conspiracy theorists ever do, because it undermines confidence in the entire medical system, and indeed, the scientific peer-review system.

Dr Summers writes that "Carrey wrote his post and appears as a spokesperson for Generation Rescue while affecting the posture of an informed and enlightened ambassador for truth." But when large swathes of peer-reviewed research is guilty of precisely the same crime, why should Carrey seem less credible? How are we to trust published research when it is so easily perverted by big money?

I tend to believe the published "basic research" that I read, because it is unlikely that the authors have hidden agendas. But in order to verify the claims that there is no statistical link between autism and MMR or thimerosal, I will not only have to read those papers, but verify that none of those authors had a conflict of interest in writing those papers. This is rather hard to do sitting at my desktop on the other side of the world. (I may also have to verify the statistics used: see this article on hypothesis-testing in the medical literature.)

So while I do not believe vaccines are bad, I am suspicious of mercury-based preservatives in vaccines, unwilling to believe disclaimers from the CDC, and unwilling to accept that every vaccine included in the US schedule (or being pushed on my son by a corporate hospital in Chennai despite its absence from the Indian schedule) is being considered on its merits alone. Take the hepatitis A vaccine: the WHO's recommendations say that large-scale childhood vaccinations should be "considered" only in regions of intermediate endemicity; but it is included in the US schedule (a low-endemicity country) and was recommended by our paediatrician in India (a high-endemicity country).

And I would like to see a lot more research on vaccine interactions, before administering five or more bundled vaccines in one shot to an infant.

And industry-funded research is unlikely to persuade me that my worries are unfounded.

Tuesday, May 05, 2009


Dr N S Murali, a well-known surgeon, a key figure in the Voluntary Health Services hospital in Chennai, the founder of the Seethapathy Clinic and Hospital in Chennai, and my uncle, died suddenly while on a break in Coventry today. He was 73. His wife had died, equally unexpectedly, a month ago.

The phrase "a towering figure" is over-used, but he was one. The shock is palpable. But his work will survive: both directly (his son and daughter both work in his hospital) and indirectly (he influenced many, many people in the city, and probably outside too). RIP.

[UPDATE 06/05/09: Obituary in The Hindu.]

More Indus thoughts and links

[UPDATE 17 Sep 2010: The story continues here.]

I am an expert neither on archaeology and history, nor on computational linguistics (though some of my interests come close to the latter). My previous post attracted 56 comments, and I eventually closed comments because it seemed that nothing productive was going to occur, and meanwhile certain individuals seemed to be using the space to redo arguments that had already been hashed out elsewhere years ago.

I would like to view the problem as a Bayesian one: given two or more hypotheses, each with prior probabilities, and a set of data, calculate the posterior probabilities of the hypotheses. This is essentially what we all do in "learning from experience". By "prior probability" is meant how likely we consider the hypothesis in the absence of data. By "posterior probability" is meant how likely the hypothesis should seem after we have seen the data. These posteriors will become priors when we see a new set of data. When we try to answer questions such as "Given that it is cloudy and humid, will it rain?" we base our answers on years of accumulated experience.

Bayes' Theorem states, basically, that given a set of mutually exclusive hypotheses Hi, with prior probabilities P(Hi), and given some previously unknown data that pertains to these hypotheses, the "posterior probabilities" of the hypothesis P(Hi|D) are proportional to P(Hi)P(D|Hi) where P(D|Hi) is the "likelihood" of the data given the hypothesis. That is, the posterior probability is proportional to both the prior probability of the hypothesis, and the probability of seeing the data given the hypothesis. To ensure that the posteriors sum to 1, there is also a "normalisation factor" that is the sum of all posteriors.

An example may make it clearer: Suppose a patient is being tested for a particular kind of cancer. In people of that age group, this particular cancer occurs in 0.1% of the population (one in a thousand people). The test correctly reports cancer 99% of the time in patients with cancer (it gives a "false negative" 1% of the time). However, in patients without cancer, the test incorrectly reports cancer 5% of the time. In this case, the test is positive. What is the probability that the patient has cancer?

There are two hypotheses: H1 = the patient has cancer, H2 = the patient does not have cancer. Their prior probabilities are, respectively, 0.001 and 0.999. If the patient has cancer, the probability of seeing the data (the positive test) is P(D|H1) = 0.99. If the patient does not have cancer, the probability of seeing the data is P(D|H2) = 0.05. Then Bayes' theorem tells us that the posterior probabilities of H1 and H2 are, respectively, proportional to 0.001 times 0.99 and 0.999 times 0.05, or respectively, 0.00099 and 0.04995. After normalising, the probabilities are roughly 0.02 for "the patient has cancer" and 0.98 for "the patient does not have cancer" -- the information given by the tests is insufficient to overcome our prior information about the likeliness of cancer.

Bayes' Theorem can be readily proved in the "frequentist" interpretation of probability theory, which until recently was the only widely accepted interpretation. In this interpretation, a probability of an outcome is the fraction of times, in a large number of "trials", that the outcome can be observed. If one has N identical situations, and P of them yield positive outcomes, the probability of a positive outcome is P/N. (Think of coin tosses: if you toss an unbiased coin a thousand times, you will get heads in roughly 500 of them.) In the cancer example, the rate of occurrence of the cancer in a general population and the success rate of the test can be quantified via frequentist methods.

Where Bayesian methods were controversial is when the frequentist picture does not apply -- one does not have access to a large number of trials. For example (to get ahead of our story): "What is the probability that the Indus script represents a written language?" A frequentist would call the question meaningless, unless there is a large number N of Indus-like civilisations, each with similar scripts, and it was known that for P of those civilisations the script represented a language: then P/N is the prior probability of the language hypothesis. But in our case, N=1 and P is not known. Similarly, in the above medical example, if it were known that the cancer is a genetic condition and the patient has a family history of it, that would affect the prior probabilities, but it would be hard to calculate the correct priors. A good doctor, however, would certainly take the information into account in some way, and not dismiss the question as meaningless. And a Bayesian would say that a "gut feeling" assignment of the prior probabilities is better than no assignment at all.

If any useful observation came out of my previous post and the comments therein, it is this: we need to calculate P(D|H), that is, the probability of seeing the data shown by Rao et al. given the language hypothesis (HL) and given the non-language hypothesis (HNL); and the prior probabilities for each of those hypotheses. If we could actually do these, then we could assign a fairly confident posterior probability for each hypothesis.

Given the data for other languages in the Rao et al. paper, I would estimate P(D|HL) to be close to 1. That is, if the Indus script is a language, I would think it very likely that conditional entropies would closely resemble the data that Rao et al. show in their figure 1A. In comments to my previous post, I estimated P(D|HNL) as about 0.1: that is, if the script is not a language, the chance that it looks so much like a language is about 0.1. This was based on a crude argument: a generic sequence-generating process could lie anywhere between the "type 1" (fully random) and "type 2" (fully correlated) lines in figure 1A. Languages occupy a very narrow band in this region, that accounts for about 10% of the area (or 10% of the height at any given number of tokens). The probability of hitting that narrow band by chance is then about 10%. Of course, one can quibble with this: perhaps there is a large class of non-linguistic sequence-generating algorithms that will give conditional entropies in this band, but I think the burden is on those who protest to demonstrate that such classes of algorithms (a) exist and (b) are likely to have been used.

Estimating the prior probabilities is a whole other problem. Someone who professes complete ignorance would assign a prior of 0.5 to each hypothesis. With my favoured likelihoods of the data, above, this yields a posteior probability of about 0.91 for the language hypothesis, and 0.09 for the non-language hypothesis.

But we are not completely ignorant: we do know quite a lot about the Indus civilisation. So how do we assign a prior to the two hypotheses?

For Steve Farmer, Richard Sproat and Michael Witzel, the answer is: there is zero probability that the Indus civilization was literate. Their arguments are in this 39-page paper, but Farmer summarises it here in one sentence:

"Not one ancient literate civilization is known — including those that wrote routinely on perishable materials — that didn't also leave long texts behind on durable materials."

To this we can add one more claim from their longer paper: the statistical distribution of Indus symbols, including the large number of "singletons", that is, signs that occur only once, is proof that it could not be a language. (The word "proof" actually occurs twice in their paper, and the title is " The Collapse of the Indus-Script Thesis: The Myth of a Literate Harappan Civilization". In other words, there is not much room for doubt -- at least, according to these scholars.

But the one-sentence summary of Farmer is easily refuted: only three other equally ancient advanced civilisations are known (Babylon, Sumer, Indus), and the Indus was by far the largest and most advanced of the these. Farmer's sentence loses its impact somewhat when one realises that "Not one ancient literate civilisation is known..." means "Not one of the three that are over 4000 years old".

Iravatham Mahadevan writes an excellent article in the Hindu demolishing the Farmer thesis on archaeological and historical grounds. I think his arguments can be summarised, Farmer-style, in the following sentence:
Not one civilisation is known, at any time in history, that was mainly urban, lived in planned cities with water supply and sanitation, had extensive trade networks, accurate measurement systems, and occupied an area of over a million square kilometres, but were illiterate.

Here we are not comparing with just three ancient civilisations, but with hundreds more between that time and ours. Many would say that we don't need to compare: the absurdity of the hypothesis that such a civilisation would be illiterate is self-evident.

As for the statistical arguments and the singleton count, Ronojoy Adhikari points me to this page containing data from Bryan Wells showing sign distributions from Proto-Sumerian, Proto-Elamite, and Uruk. Perhaps Farmer et al. will now argue that these were not scripts either.

Here is my feeling on what has happened here: Before 2004, the Rao et al. paper would not have gathered any attention. (Of course the Indus system is a language script! Why are you discussing it?) But that year, Steve Farmer managed to persuade two others -- one of whom, Michael Witzel, is a well-respected authority in the field -- to add their names to his thesis that it is not a language. The resulting manuscript was absurdly and unprofessionally bombastic in its language, while containing essentially nothing convincing. Regardless of the work of Rao et al, their hypothesis would have died a natural death -- but Rao et al do have Farmer et al to thank for enabling them to publish their work, with its obvious conclusions, in a prestigious journal like Science. Farmer et al are so rattled that they promptly post an incoherent, shrill, content-free, ad hominem rant on Farmer's website. Sproat even shows up on my previous post, leaving a chain of comments that reveal that he has neither understood, nor cares to understand, the argument. All those who dissent from their 2004 paper are Dravidian nationalists.

So that leaves the question: how do we assign prior probabilities for the two hypotheses? I think the opinions of Farmer, Sproat and Witzel can be discounted. If we instead asked: "Given that every other urban civilization with water supply and sanitation was literate, how likely is it that the Indus civilisation was illiterate?" I think the answer would be: "Extremely unlikely."

If we assign a prior of 0.9 for language (based on the above, I'd put it higher) and 0.1 for non-language, and retained my likelihoods as above, the posteriors are: 0.99 language, 0.01 non-language.

I expect it to get more convincing. Some actual (non-rhetorical) evidence to the contrary would, however, be very interesting.