Monday, March 19, 2012

Dimensions of Intelligence and Personhood

I recently listened to this podcast interview with George Dvorsky. He made some excellent points about extending the definition of personhood, one of which involved the concept of intelligence or sentience as a "spectrum". He cited the recent PETA court case which attempted to extend the definition of "person" to orcas, to include them in the class of individuals which should not be subjected to slavery. (See PETA Sues SeaWorld for Violating Orcas' Constitutional Rights).


Dvorsky suggested that by assessing certain attributes of various species' intelligence and physiology, humans could assess the possible inclusion of those species in the class of "persons". He explained that the intelligence of a species can be viewed as a spectrum: a nearly brain-dead human is the low end, a developmentally impaired human is further along, and a normal human is in the middle. By extension, a human genius would be at the far end of the spectrum.

Dvorsky also compared the intelligence of certain animals to that of humans, specifically the idea that an African Gray parrot is considered about equal to a (presumably normal) 3- or 4-year old human. Over the years there have been many such comparisons: dogs and cats and horses and pigs have all been compared to humans of various ages. He opined that the first species to be granted personhood will be dolphins, of which orcas are a subspecies. 


While I agree with his Dvorsky's opinions, I think the use of a linear spectrum will be too limiting, and will cause humans to misjudge, overlook, deliberately ignore, or denigrate aspects of intelligence which are different from human intelligence, but are no less valid or important. A traditional IQ score, for example, ignores intelligence and skills in artistic and social areas. I suggest an approach to this assessment based on multiple dimensions of intelligence, while keeping it simple enough to be represented in two dimensions and visualizable by average humans. I hope that this approach can be used to help in the process of opening minds to different ways of perceiving intelligence, and lead to greater acceptance of the personhood of various species.  


Here is an example of what I mean. This is a modified version of a "radar chart" which presents hypothetical scores on a dozen aspects of intelligence. This is just an example - I'm sure we could argue for years over which attributes deserve to be included. I have deliberately included some that are stronger in certain non-human species. 


















I suggest that the total score is what's important, not the scores in certain sectors which favor humans. (This particular kind of graph distorts the scores by showing more area in the outer rings, but you get the idea.) If we were to present these scores in a double-blind fashion, not knowing which species they represent, humans just might have to agree to set an objective threshold which indicates sufficient intelligence to qualify for personhood. Obviously there is variation between individuals of a species, sometimes very significantly, but a lack in one area or another does not disqualify a human from being a person. For example, a blind person would score poorly in Visual Pattern Recognition, but because of compensation by other senses, might outscore sighted humans in Auditory Pattern Recognition.


If we were to plot the scores for a variety of species, we would find that some clearly beat out humans in some aspects. Dvorsky casually mention in the interview that humans are the most intelligent species on Earth, but all things considered, that may not be true. I have not undertaken such a detailed, quantitative, objective study, but over several years I have begun to consider that the race may be very close indeed if we take a broad view of the definition of intelligence.


I agree with Dvorsky that cetaceans will be the first to be granted personhood. I think it may be a toss-up between dolphins/orcas and rorquals, for two reasons: dolphins/orcas are better known to most humans, and the intelligence they show more closely matches the "sectors" that humans currently appreciate and value. I'm basing my impressions on two excellent and deeply moving books; admittedly these impressions are anecdotal and emotional. 


In Listening to Whales, Alexandra Morton presents years of study of dolphins and orcas. Some sectors which will resonate with humans:
  • Clearly they have a language. We do not understand it, but we can recognize different dialects.
  • Clearly they are highly social, with complex, long-lasting relationships.
  • Their auditory pattern recognition probably exceeds our visual pattern recognition. Read the stories of how orcas can use their sonar to not only "see" surfaces, but also see through us and each other like sonograms, and how they can detect heart rates and breathing rates.
  • They mourn and in some cases bury their dead.
  • They highly value synchronicity.
  • They have exhibited compassion for humans.
  • They are easily bored, and like to make up games for their own amusement, and to perplex their captors.
  • They apparently have treaties between different populations to respect each others' food sources.
(From some other source I cannot remember, I learned that dolphins have names for each other. In muddy water, they communicate frequently to keep in touch, and they identify themselves with unique names.)

In Among Whales, Roger Payne summarizes years of study of right whales, fin whales, blue whales and others. He expounds in great detail how the larger ones have developed extreme signal-processing capabilities which may explain why their brains are so huge. Lesser-known skills which, if proven, are clearly significant factors of intelligence, but are not well known among humans:
  • Fins and Blues can sort out signals (vocalizations) from noise even across the entire North Pacific basin.
  • Using this near-global communication, which predates humans' electronic communication networks by millions of years, they can coordinate their travels to arrange meetings at certain times and places.
  • They can use sound shadows to locate islands or land masses from hundreds of miles away.
  • They are creative - the complex music humpback whales produce is quite well known among many humans.
  • They also have complex and long-lasting social relationships.
  • It seems that they may think on a longer, more cyclical time scale than humans, which makes interpreting their behavior and intelligence difficult.
From other sources, it is clear that some gray whales now consider humans to be non-threatening, and actively encourage their young to approach us. I'm anthropomorphizing a bit here, but it may be that as a group they have forgiven humans for the genocide we perpetrated against them. For the same reasons we take our kids on field trips to the Holocaust Museum, gray whales now take their calves to meet humans, their former enemy, in their little rubber boats. If so, and if we add Compassion to our chart, whales clearly exceed humans in that aspect of intelligence.  

I'm sure I'm not the first to think of a multidimensional representation of comparative intelligence. I'm just now learning about this field; I'm sure the bioethecists and the Non-Human Rights Project people have thought of this approach. 


So why are transhumanists and singularitarians interested in the personhood of cetaceans? Because eventually humans are going to have to come to grips with the rights of "persons" which are extensions of humans, or former humans, or electronic clones of humans. Acknowledging the personhood of cetaceans, great apes, and elephants will come much sooner, and will provide humans some good practice in broadening their definition of "person".

Sunday, January 29, 2012

Cetacean species I have seen in the wild

Gray whale
Bottlenose dolphin
Common dolphin
Humpback whale
Fin whale
Minke whale
Blue whale
Orca
Risso's dolphin
Pacific white-sided dolphin

Saturday, January 21, 2012

Books on or related to the Singularity

These are all books and major I have read in the last couple of years that are about or deal with some aspect of the singularity:

  • The Singularity is Near: When Humans Transcend Biology by Ray Kurzweil
  • The Intelligent Universe: AI, ET, and the Emerging Mind of the Cosmos by James N. Gardner
  • The Age of Spiritual Machines: When Computers Exceed Human Intelligence by Ray Kurzweil
  • Whole Brain Emulation: A Roadmap by Anders Sandberg and Nick Bostrom
  • The coming Technological Singularity: How to Survive in the Post-Human Era by Vernor Vinge
  • Artificial Intelligence as a Positive and Negative Factor in Global Risk by Eliezer Yudkowsky

Fiction:

  • Singularity Sky by Charles Stross
  • Marooned in Realtime (Peace War) by Vernor Vinge
  • Accelerando (Singularity) by Charles Stross


These are on similar topics of science and cosmology but not explicitly about the singularity:

  • Transcend: Nine Steps to Living Well Forever by Ray Kurzweil and Terry Grossman M.D.
Fiction:

  • Rule 34 (Halting State) by Charles Stross



Friday, January 20, 2012

Learning Applescript

I'm a programmer, always have been, always will be. My current job doesn't involve much programming, so I enjoy doing projects on the side. About a year ago I got a MacBook Pro and have been looking forward to doing some programming there, but have been too busy with home improvement projects etc. I  wrote a program to manage backups from my server to a hard drive, but that was just a bit of Perl, nothing fancy.

After upgrading to Lion and buying an e-book about it, I learned that AppleScript is included, so I also picked up an e-book "Learn AppleScript: The Comprehensive Guide to Scripting and Automation on Mac OS X" and also Apple's language guide. I'm only about a third of the way through the Learn book, and I've done variations on several of the exercises.

Wow.

Wow.

I think AppleScript will be my new favorite language. The English-like syntax is OK - I don't really need that level of simplicity - but the power of what it can do ia amazing! Controlling scriptable applications is so powerful. I can envision doing lots of things with mail, Office programs... If one tried to do this in any of the C-based languages, it would take a lot of work. AS makes it simple, at least that's how it works so far.

With just what I've learned so far, I've pulled data on songs from iTunes, linked that to Finder to get data about the song titles, done simple searches... it almost lets you treat iTunes like a database.

Today's exercise involved events and setting up a simple client-server system. It turns out that AS itself is a scriptable app. I created a server script that runs in the background, executes a routine task every 10 seconds, and also responds to commands from a client. From a GUI dialog box in a client app, I can start, stop, and request services from the server app. The client is 11 lines of code. The server (which just speaks "initializing", speaks the current time on request, and speaks "quitting") is 22 lines of code.

And all this fits neatly into the Mac GUI... scripts can be dragged to the dock like any other apps.

Wow.

I can't wait to see what's in the remaining two thirds of the book!