Sunday, February 25, 2007

Mercy and Email

Cruelty has a human heart
And jealousy a human face,
Terror the human form divine,
And secrecy the human dress.
William Blake, A Divine Image

George Will's column today is celebratory of Clint Eastwood's "Letters from Iwo Jima" making it to The Academy Awards because it serves as an example of how much our perspective has changed about enemies in wartime. He contrasts the portrayal to the historical backdrop of anti-Japanese sentiment during World War II, quoting Admiral Bull Halsey's billboard at a base that excoriated troops to "Kill Japs, Kill Japs, Kill More Japs."

The topic of moral drift in America, Europe, Australia and many other countries is fascinating to me because there does seem to be a movement towards increasing sensitivity to human suffering, increasing belief in the universality of human rights, and an impressive improvement in the very language we use to talk about other people. Even among the poorly educated with whom I occasionally have contact, outright racism or homophobia is hedged very carefully into neutered statements about "they don't respect life like we do" or "I just can't stand the thought of two men kissing." I haven't heard the "N word" used in 20 years.

I mentioned previously that increased education and communications likely play a part in this effect, but I would like to refine that suggestion a bit based on research about how email flame wars erupt. Flame wars often arise, it seems, due to the lack of verbal and facial cues that we use in phone and face-to-face conversations. The rate of misinterpretation for email is as high as 50% according to some research. Not surprising, since communication via textual symbols is a side-effect of general communications capabilities that we assume helped with face-to-face information exchange and social interaction.

I think a similar phenomena is at work in terms of the general moral Zeitgeist. Even if evolution has tuned us to believe out-group should be less worthy of rights and opportunities than in-group, I think the rise of mass communications and visual presentations of other cultures that show our common humanity essentially converts xenophobia into acceptance by providing direct facial evidence of the out-group's similarity to us. In 1945, when incendiary bombing of Tokyo began, the amount of knowledge and understanding of Japanese culture in America was virtually nil. The Middle East and Islam had the same quality prior to the first Gulf War. But, as our understanding grows about the common humanity, we find it harder and harder to dehumanize. And faces may be the key.

Here's the parallel verse from Blake:

For Mercy has a human heart,
Pity a human face,
And Love, the human form divine,
And Peace, the human dress.
William Blake, The Divine Image

Friday, February 23, 2007

Chimps and Spears

From Washington Post via SF Chron:

Chimps make wood spears, kill smaller animals for food

Females described as problem-solvers in primate culture

Friday, February 23, 2007

Chimpanzees living in the West African savannah have been observed fashioning spears from sticks and using the handcrafted tools to hunt small mammals -- the first routine production of deadly weapons ever seen in nonhuman animals.

The multistep spearmaking practice, documented by researchers in Senegal who spent years gaining the chimpanzees' trust, adds credence to the idea that human forebears fashioned similar tools millions of years ago.

The landmark observation also supports the long-debated proposition that females -- the main makers and users of spears among the Senegalese chimps -- tend to be the innovators and creative problem-solvers in primate culture.

Using their hands and teeth, the chimpanzees were repeatedly seen tearing the side branches off long, straight sticks, peeling back the bark and sharpening one end, the researchers report in Thursday's online issue of the journal Current Biology. Then, grasping the weapon in a "power grip," they jabbed into tree-branch hollows where bush babies -- small monkeylike mammals -- sleep during the day.

After stabbing their prey repeatedly, they removed the injured or dead animal and ate it.

"It was really alarming how forceful it was," said lead researcher Jill Pruetz of Iowa State University.

The observations are "stunning," said Craig Stanford, a primatologist and USC anthropology professor. "Really fashioning a weapon to get food -- I'd say that's a first for any nonhuman animal."

Scientists have documented tool use among chimpanzees for several decades, but the tools have been simple and used to extract food rather than to kill for it.

Some chimpanzees slide thin sticks or leaf blades into termite mounds, for example, to fish for the tasty, crawling morsels. Others crumple leaves and use them like sponges to sop drinking water from tree hollows.

But while a few chimpanzees have been observed throwing rocks -- perhaps with the goal of knocking prey unconscious, but perhaps simply as expressions of excitement -- and a few others have been known to swing simple clubs, only people have been known to craft tools expressly to hunt prey.

Pruetz and co-worker Paco Bertolani of the University of Cambridge made the observations near Kedougou in southeastern Senegal. Unlike other chimpanzee sites under study, which are forested, this site is mostly open savannah. That environment is very much like the one in which early humans evolved and is different enough from other sites to expect differences in chimpanzee behaviors.

Pruetz recalled the first time she saw a member of the 35-member troop trimming leaves and side-branches off a branch it had broken off a tree.

"I just knew right away that she was making a tool," Pruetz said, adding that she suspected what it was for, as well.

But in that instance, Pruetz was not able to follow the chimpanzee to see what she did with it.

Eventually, she and Bertolani documented 22 instances of spearmaking and use, two-thirds of them involving females.

In a typical sequence, the animal would discover a deep hollow suitable for bush babies, which are nocturnal and weigh about half a pound. Then the chimp would break off a nearby branch -- on average about 2 feet long, but up to twice that length -- trim it, sharpen it with its teeth, and poke it repeatedly into the hollow at a rate of about one or two jabs per second.

After every few jabs, the chimpanzee would sniff or lick the tip, as though testing to see if it had "caught" anything.

In only one of 22 observations did a chimp get a bush baby. But that is reasonably efficient, Pruetz said, compared to standard chimpanzee hunting practice, which involves chasing a monkey or other prey, grabbing it by the tail and then slamming its head against the ground.

Chimpanzee behavior is widely believed to offer a window on early human behavior, and many researchers have hoped that the animals -- which are humans' closest genetic cousins -- might reveal something about the earliest use of wooden tools.

Many suspect that wooden tools far predate the use of stone tools -- remnants of which have been found going back 21/2 million years. But because wood does not preserve well, the oldest wooden spears ever found are only about 400,000 years old, leaving open the question of when they first came into use.

The discovery that some chimps make wooden weapons today supports the idea that early humans did, too -- perhaps as early as 5 million years ago, Stanford said.

Adrienne Zihlman, a UC Santa Cruz anthropologist, said the work supports other evidence that female chimps are more likely to use tools than males, are more proficient tool users, and are crucial to passing that cultural knowledge to others.

"Females are the teachers," Zihlman said, noting that juvenile chimps in Senegal were repeatedly seen watching their mothers make and hunt with spears.

"They are efficient and innovative, they are problem-solvers, they are curious," Zihlman said of females.

And that makes sense, she said.

"They are pregnant or lactating or carrying a kid for most of their life," she said. "And they're supposed to be running around in the trees chasing prey?"

Wednesday, February 21, 2007

ABCs and Magical Thinking

The most effective form of psychotherapy is based on assessing and eliminating irrational thoughts and beliefs. That's right. Cognitive Behavioral Therapy--originally called Rational Therapy or Rational Emotive Behavioral Therapy--is effective against depression, obsessive compulsive disorder (OCD), anxiety disorders and bipolar disorder. In some cases, like for OCD, the effectiveness can be stunning with 80% of patients reporting an improvement via cognitive methods.

So how does it work? There is a handy acronym, ABC, that stands for Activating Event, Beliefs, and Consequence, and lays out the basics. At core is the idea that a patient's cognitive model is irrational. Their irrational beliefs lead them to interpret an Activating Event in an irrational manner, leading in turn to negative consequences like cycles of depression or obsessive thoughts. We have all experienced these sorts of problems in our lives. Ever obsess over a boy or girl? But for major depression or OCD or anxiety disorders, the irrational beliefs are fixed and persistent in leading to negative consequences. Cognitive Behavioral methods focus on getting a new map in place by undermining the irrational beliefs to convert the A into a positive outcome for C. Simple as that. No need to discuss your relationship to your parents or anything so elaborately Freudian (unless there are specific beliefs related to them: "I feel like Mother is constantly looking over my shoulder").

Now, what I want to get at here is that the effectiveness of cognitive behavioral methods in undermining irrational beliefs also highlights the general issue of magical thinking--non-scientific attributions of causality.

During grad school I would quite often come up against New Agers because I circulated in an arts community and New Age beliefs are very common in arts communities. Occasionally I would challenge some of the claims, but always in a very gentle and (I thought) kind manner. I remember one example where a friend suggested I should buy these mushrooms that you soak in water. You then drink the gray water they soak in on a daily basis and you get fantastic health, vigor and vim. I had never heard of this mushroom and had no idea whatsoever whether her claims were true or not. I asked her how the mushroom worked and she described a very complicated metaphorical landscape in which the immune system is a fortress and the mushroom helps to build moats around the fort, or something very similar. I mentioned that I tended to think of the immune system as a more fluid battlefield where the enemy is constantly being identified, surrounded and killed. That was all I said, but the reaction was angry, attacking and spiteful. In fact, I think she hung up on me. This happened a few more times during those years, and in each case the person in question was a college-educated artist or humanities major. Most were in grad programs or were in late undergraduate.

I have to add, here, that these kinds of reactions have been paralleled only by the angry responses I have received occasionally from Christians to my self-professed atheism. Some of those Christians have been people very close to me and their vehemence and anger was hurtful. Just like the mushroom incident, however, the responses caught me by surprise. I have rarely challenged anyone's beliefs unless asked directly for my opinion (well, except in newspaper editorials of which I've written some fairly nasty ones on the intelligent design debate and church-state separation) and the topic of religion hardly ever comes up in daily conversation. I certainly never said anyone was stupid or did anything that should provoke an ad hominem attack, yet even gentle philosophizing does seem to have that effect on some.

My 8-year-old son encounters similar problems in school. We cultivate tolerance and understanding in him and he is a remarkably nice kid. We also tell him that he has lots of time to figure out what he wants to believe about religion and try to answer his questions as best we can. He lives in a time of wars and unrest partly partly caused by religious hatred and intolerance, though, and he has a generally negative opinion at the moment. Maybe I shouldn't let him watch the news? But most disturbing is that he is routinely told that he is going to go to hell by kids at school. Now I don't recall that sort of thing coming up when I was in grade school. I just don't think we talked about the topic of religion or that religion was considered more of a private matter. I was always an atheist (well, some New Age thinking in late high school and college crept in) but the topic never came up until much later in life; no one bothered me about it and I didn't bother anyone else. But now it is a regular and hurtful affront to the sensibilities of a charming young mind (I'm biased, I know, but all evidence points to us doing a pretty good job as parents).

So the question that I pose here is where are the separating hyperplanes between magical thinking and dangerous irrationality? Neither really meets clinical psychological criterion unless they result in self-destruction, unlawful behavior, or the holder of those beliefs asks for help. Moreover, much of the time religious belief results in positives for the individual, including strengthened community ties and improved sense of self-worth. It may even boost the immune system. But the barrier is more porous for die hard believers for whom literalism irrationally empowers unlawful or simply mean behavior. Is there a cognitive behavioral therapy that can be applied on a larger scale to help draw down the negative impacts?

Richard Dawkins has recently been proclaiming that teaching children religion from a young age is effectively child abuse because it essentially robs them of their human right to make an informed decision about faith, reason and belief. The perspective is extreme, but it points to the question of how to reduce the negative consequences of extreme irrationality in our societies. One idea that I see applied at my son's elementary school is values and character education that is aimed at reducing bullying and promoting tolerance. There are a series of ideas that are articulated through posters published in the multi-purpose room (nee cafeteria) and library that promote a kind of ABC-like process for assessing the consequences of one's acts on others. It is largely an empathy modeling system, but does have a component of self-assessment to determine why the student thought it was alright to terrorize other students or treat others with disrespect. The approach is not applied to the religion issue, however, and instead the policy appears to be to inform the students that they are in school to learn and have fun, and that religion should be discussed elsewhere. I say "appears to be" because I have only heard it from my own kid's mouth and think he is doing just fine in not responding in kind to the comments of his classmates.

Are these methods enough to inoculate the kids against extremism? Likely not, but developing empathy models early on will likely serve us all in the future.

Saturday, February 17, 2007

Speech and The Semi-Sentient Insult

I picked up a new Windows Vista machine for testing purposes related to my startup recently and found myself trying-out the automatic speech recognition (ASR) engine. Under my the terms of my seed funding, a subcontractor of mine is investigating using ASR to transcribe scientific talks. The transcriptions would then be indexed to make the content searchable in a knowledge management service. I thought I would contribute a bit to their work by getting some qualitative and quantitative understanding of the quality of ASR today. I've also been encountering ASR in more and more places: my car integrates with my Blue Tooth phone and supports voice dialing; my son has an R2D2 toy that has crude ASR.

The quality of the Vista ASR for dictation started off very poor, but I was pleasantly surprised after a small amount of training that I could dictate at around 95% accuracy if I use a very careful "anchorman"-style enunciating patter. That 95% figure was over some challenging technical terminology, too, like "algorithmic information theory" and "computational linguistics". Not bad, but at a normal pace with "ums" and asides thrown in, the accuracy dropped to around 40%, which is essentially unusable. Even worse, my subcontractor reports that trying speaker-independent untrained systems from Dragon/Nuance goes to around 10% accuracy and that the out-of-the-box claims just border on outright lies for Naturally Speaking.

Still, 95% is an impressive technological achievement and the methods used to achieve these results are important enough that I want to mention them and describe a bit how I believe they relate to the topic of this blog.

A language model is a statistical distribution of how phonemes, words, characters, phrases, parts of speech, or any other feature of language occurs in a corpus of text or speech. ASR uses several interacting language models to try to predict what you are saying at any given time. A speaker independent engine has a large, generic language model that is supposed to be flexible enough to accommodate the vagaries of accents and intonation, prosody and pace. A trained system has a generic model with some of the probabilities adjusted by hearing you read some material (hopefully not too much material). When recognition takes place, the system has to take the sound bit after some filtering and try to match it against the most likely word in the language model based not just on the word itself but also on the other words that came before it. So the language model is not just a library of word probabilities, but also of the contexts of word probabilities. And here is where the systems fail: the context analysis is currently too shallow and lacks higher-order corrections from semantic and syntactic sources. Well, that is a bit too egregious of me, since there are some higher-order corrections implicit in the sequences of the temporal language models. It's just that language is so sparse that there is not enough context to encode all of the complex interacting variables that go into a perfect model. It works that way for us, too, since we can easily be made confused by very rare verbal utterances ("furious the fever fled forth from me..."; "what, say again?").

The point here is that with increased computing power and enhanced training there is no reason that ASR can't get to 98% or even 99.9% speaker-independent. I would guess that we need another 10 years to close that gap, but there are no theoretical challenges for most mainstream speaking transcription tasks. And when that happens, there will be a shift in the way that we regard computing machines. Children will grow up expecting voice control over toys, tvs, dolls, video games. They will expect conversational capabilities in their entertainment, and will develop a two-tiered mental model for how to interact with people versus the merely verbally reactive. I am guessing it will become a common way to insult someone by conversing with them like they are among the semi-sentient.

Friday, February 16, 2007

Ancestry and Art

Does artistry recapitulate phylogeny?

My wife and I got DNA ancestry analysis done for our holiday presents late in 2006. We used based on some reviews we found combined with price considerations. The reports we got back confirmed a couple of speculations we harbored about our pasts and also revealed some unexpected outcomes.

The method of analysis is based on Polymerase Chain Reaction (PCR) to amplify specific sequences that are used for forensic DNA analysis. That's right, the 13 specific loci are also used by the FBI and CSI units as a genetic fingerprint for solving crimes. At each loci there are different possible patterns of base pairs, with each pattern that occurs in the wild called an allele. The alleles have numeric codes in the literature and population studies have been done to establish the distributions of allele patterns across different ethnic groups. In fact, you can even download a huge Excel spreadsheet application that contains many of the distributions of allele patterns based on FBI studies and other sources in the forensic DNA literature.

For me, the analysis confirmed a Black Irish model with Tunisian, Spanish, Portuguese and Italian arriving through my Scots-Irish lineage (established through my family tree), and Russian, Norwegian, Flemish and Austrian presumably through my Dutch/Scandinavian side. Note that the comparisons are with "deep ancestry" in the sense that the distribution tables reflect populations that are known to have been somewhat genetically stable, so if your ancestors landed for a generation or two in Copenhagen (like mine) before moving on, you don't get to see the Dutch.

My wife had some more interesting results, with Sub-Saharan African (Equatorial Guinea) coming remarkably high on the list, just under her Polish and Russian. OK, very interesting, and it might be due her long line of Southern slave owners on one side visiting the slave quarters now and again, and then bringing the children up as their own. Or maybe it is just a false positive. Omnipop (the NIST Excel application) also confirmed that her allele pattern looked much like diaspora African American populations, but her mother had her analysis done and showed no signs of the African origin. Since her father is second generation Polish-Lithuanian, it is hard to imagine how African made it into his bloodline, so we are curious about the possibility of a false positive on the African link. Bummer, if so, but it made for some great fun over the holidays.

The picture above is a detail from an artwork I put together back in January. The artwork is an example of generative art relying on the actual short tandem repeating DNA sequences taken from my forensic profile to render the form and structure of a tree-like representation. In addition to the "language of life," I superimposed poetry over the sequence branches derived from Spain and Norway (specifically a poem about Yggdrasil, the World Tree, for the latter).

As I write this, I am awaiting the arrival of the final, framed version. At 60" by 60", the printing and framing process has run to around 4 weeks, now, much longer than the time needed to develop the software, tune it, and finalize the color and poetry contributions. On the negative, I just missed being able to enter it into a juried show, but I will find a venue soon enough.

Thursday, February 15, 2007

The Clitoris and Civilization

The clitoris may be responsible for civilization.

Now hear me out, because I am semi-serious. The clitoris is a very odd aspect of female anatomy because it receives limited stimulation during intercourse, the critical act of procreation and proximal cause for all things evolutionary. In fact, 70% of women never have orgasms during intercourse.

Yet, there it is, capable of arousal and even capable of multiple orgasms. Moreover, there is some evidence that the chances of conception increase when an orgasm does accompany intercourse. There is even a video that shows the cervix moving down into a pool of semen during female orgasm, as if to better help get the semen to the destination.

One theory is that the clitoris is just vestigial. Doubtful, as we can see our cousins the Bonobos using them for all kinds of kinky bonding rituals. Instead, a more compelling idea to me is that the haphazard positioning of the clitoris and the time and effort needed to make intercourse and female orgasms compatible are related to female mate choice making. A male who is likely to listen to the woman's desires--and fulfill them--is also a male who will be more of a partner in terms of the long-term childrearing requirements of these big-brained babies.

If this is the case, then we should have seen evolutionary pressures towards greater degrees of compatibility and lower rates of male aggressiveness over time, and increased mid- to long-term partnering.

All because of the clitoris!

Popes and Reason

I recently reviewed the Wikipedia article on Pope Benedict's controversial statements on Islam. The Pope's topic was on the relationship between faith and reason, and was apparently mostly targeting the secularity (I use the less-loaded term instead of "secularism") of some societies. It is worth noting, however, that Pope Benedict appears to consider elements of Islam to consider reason as subordinate to the will of God, and that such a consideration is a metaphysical crisis point in terms of any urge to violence that arises out of extremist subpopulations of Islamic societies.

What makes this noteworthy to me is that The Pope seems to support the idea that Christianity and reason are fully compatible, or at least that they are more compatible than Islam and reason. Now I don't want to grab the easy fruit of historical criticism of this perspective, but instead am trying to envision how the Vatican builds compatibility between "reason" and some of their social perspectives like antagonism to contraception. Is it a reasonable position? Or is it one that emerges from a policy choice that is historically and, perhaps, liturgically consistent, but that is at odds with reasoning on public health and safety?

The compatibility of reason and faith must, at some level, involve cherry picking elements of each. The Pope accepts The Big Bang because it offers some metaphorical consistency with Biblical descriptions, but considers principles over public health on procreation-related matters. In these matters, I believe we are seeing the kinds of pragmatism that has made Christianity survive into modernity: syncretize traditional religious elements (from Black Madonnas to Pagan Holidays), become a partner with secular government, focus on social charity, become compatible with reason as necessary.

We can only hope that Pope Benedict is proved wrong by moderate Muslims.

An initial foray...

I saw a clip of Richard Dawkins on YouTube recently discussing how the Zeitgeist has shifted towards greater and greater respect for human rights in at least the Western World, if not the world as a whole. He led off with some remarkably racist quotes from Abraham Lincoln that reflected the prevailing views of the era, then moved on to describe how carefully the fate of civilians has been managed in the US invasion of Iraq when compared with mass civilian bombings of World War II in London, Dresden and Tokyo.

Things have changed. We have changed.

His point is that that shift happened due to a complex dialog within our societies that gradually has beaten down our prejudices to the point where it would take an extraordinary set of circumstances for us to collectively and unquestioningly engage in acts of appalling cruelty that were accepted even 50 years ago.

Part of the change is due to technology, I think. Global communications systems support the publication and awareness of the actions of governments, corporations and people. Part of the change is also due to the enhanced effectiveness of the press in Open Societies as they publicize events and humanize transgressions. Part of the change is due to increasing education levels and the ability of more people to understand concepts like rights and responsibilities, and to feel empowered by a sense of ownership of their own governments.

I liken this drift to an evolutionary process, but one that has a definite resistance to an overtly selfish genetic analysis because it does seem to be optimizing towards virtuous goals rather than simply supporting the selfish interests of individuals. Or does it?

This blog will be exploring some of the cultural, social, biological, philosophical, theological and personal issues that surround ethics, morals and virtue. I will point out up front that I am a materialist with a poetic side and am not above cheering for ideas that seem interesting but have minimal support in the broader literature. I will also point out that I am an engineer and scientist by training with a special interest in the evolution and simulation of intelligence, which will tend to color my discussion.