SENS and Tractability

On Friday, June 9, Technology Review published the results of three attempts to critique Dr. Aubrey de Grey's SENS proposal. These critiques are curiously geared more toward demonstrating that SENS is "unworthy of learned debate" than toward what might actually be considered "learned debate" and true scientific scrutiny.

But -- this emphasis is part and parcel of the intent of what is termed "The SENS Challenge". Anyone who looks up "life extension" on the Internet is not likely to get very far without encountering some mention of SENS, de Grey, or the Challenge. I will not bother spelling out the terms of the challenge here in detail; that would be needlessly redundant.

If you are interested, read the terms of the challenge over at Technology Review. I will suffice to say here that Dr. de Grey has been trying for quite a while now to have his SENS proposal subject to rigorous peer review, and the fact that three other biologists have at least taken a cursory look at SENS (albeit with the goal of deconstructing the venture as overly optimistic at best, and pseudoscientific nonsense at worst), is a formidable step in the right direction.

Read the Challenge submissions, Dr. de Grey's rebuttals to these submissions, and counter-arguments here.

This sort of thing really makes me wonder if I ought to run back to school and get a biology degree -- no matter how much self-study I engage in, there is always a nagging feeling that I'm probably missing the sort of context that full-time biologists are privy to, simply by virtue of being able to spend most of their waking hours thinking about biology.

However, I do think there is plenty of value in commentary provided even by people who are not biologists. The SENS challenge, and even SENS itself to a certain extent, is not so esoteric in form and content that one would require years of experience as a biologist just to be able to evaluate whether it constitutes a worthwhile experiment. There are certain properties inherent in good (that is, actual) science and engineering proposals that will be readily recognizable to anyone with decent critical thinking skills.

For instance, I am neither an expert on zoology or the geographical features of Scotland, and yet I am reasonably convinced that claims of a Loch Ness monster do not stand up to evaluation through the scientific method. I am also reasonably convinced that evolutionary theory is a robust and valid means of describing the development of life on Earth (and much more tenable than so-called "Intelligent Design"), though again, I am not a biologist. Some things are more scientifically valid than others, and you don't need to have a degree in a particular field in order to evaluate claims and deconstruct arguments.

Why am I going on about this? Well, mainly because I have already seem quite a bit of commentary online in reaction to this recent development in the SENS saga. I imagine that even more reactions will emerge from the life extension / scientific community in the coming weeks preceding the July 11 judgement of the three SENS challenge submissions. The Internet is, in some ways, more of an intellectual meritocracy than any other forum presently available to people for the exchange and scrutiny of ideas. I would urge everyone coming upon this blog to read the following, in addition to the challenge submissions linked to by Technology Review:

Reason's invitation to judge the SENS challenge for yourself.

George Dvorsky's insightful take on the matter

Eric Boyd's detailed impression from the Digital Crusader angle

The forum discussion at ImmInst

Betterhumans' members discuss what SENS might require to actually move forward

I would also urge anyone with interest in hastening the goal of healthy life extension to think carefully about SENS, the challenge, and work on brainstorming ways to assure that even "publicity" exercises like the SENS challenge lead to real progress toward the achievement of long, healthy lives for all individuals who desire them. Certainly, "learned debate" is a good start, but as much as I am absolutely convinced that ethics are inseparable from good science (especially when the science concerns people's bodies and their very lives), I also realize that action needs to take place. In laboratories, in research facilities, in people's homes, in hospitals.

Those of us with the means can donate to the Mprize, and I would like to particularly emphasize the Rejuvenation prize. Those of us with Net access can continue to emphasize the fact that there is no difference, from a moral standpoint, between wanting to live another day when you are 99 and wanting to live another day when you are 9. Doctors can start to examine their own attitudes toward health and aging, to assure that the lives of elderly patients are given as much respect as those of younger patients (that is, stop describing things like memory loss and various aches and pains as "things everyone should expect when they get old").

What I've read and managed to process so far of the challenge submissions, rebuttals, and counter-arguments seems to indicate that even in the scientific community (except for a noted few), the very notion of drastically extending human lives is considered to be so unfathomable as to be unworthy of true scientific consideration. One thing that struck me about the challenge submissions was the space taken up by offering definitions of pseudoscience and quackery -- definitions that are most likely understood quite readily by the average skeptically-minded armchair philosopher, and that any serious scientist would most likely have to understand in order to become and work as a scientist in the first place.

Certainly, caution is in order so as to make sure we do not edit evidence, discarding that which would seem to indicate our goal is more difficult to reach while focusing on those results and ideas which seem more attractive. However, caution and dismissal are two very different things. If life extension truly is to be achieved, anyone who is serious about helping to hasten its development can not afford to fall prey to wishful-thinking traps. Wishful thinking will not keep anyone alive -- the only thing that has any chance of doing this is real scientific progress, and real scientific progress necessarily entails constant objective examination of results.

As noted earlier in this entry, the Net is abuzz with commentary on this new development in the scrutiny of SENS. Some commenters may take a generalized view, some may go into great detail about what they see as the main points of the current available information, and others may look at the challenge, the rebuttals, the counter-arguments, and the already-existing commentary and attempt to focus their own commentary on something which they feel qualified to remark on.

As an engineer, I think I might actually be somewhat qualified to comment on the "tractability" issue (emphasized by Weinstein in his paper). And of course, anyone reading this should absolutely feel free to criticize the analogy I'll attempt to make -- my aim is to present a useful means for thinking about engineering approaches to aging interventions, not to massage my ego or anything similarly ridiculous.

My particular line of work is concerned with the issue of electromagnetic interference. This is an area of engineering that, apparently, few EEs are eager to approach. In fact, one of the things that made me curious about this field was the fact that many of my co-workers referred to EMI principles and practices as "black magic". As someone who has long engaged in skepticism-as-a-hobby and who takes great pleasure in debunking ridiculous claims and in demonstrating that what looks like "magic" is invariably just an aspect of nature that is insufficiently understood, I approached EMI theory from the mindset that it couldn't possibly be magic and that there were ways to systematically approach and address EMI problems.

EMI makes a lot of EEs leery, and prompts the application of descriptors like "black magic" to itself, partly due to the superficial appearance of EMI equipment (weird-looking antennas, spectrum analyzers, rolls of alumimum foil, copper tape) and partly because of the nature of identifying and responding appropriately to EMI problems. In recent years, standards for EMI compliance have become more and more stringent, and there is something of a push to design devices with regard to EMI compliance right from the outset. To make a comparison to aging interventions here, this "push" to design devices which are less vulnerable to EMI effects might be somewhat analogous to early-life (including pre-birth genetic treatments) efforts to create a more robust organism. This is the preferred approach, one that is considered an investment and one that can avoid having to "patch up" the device later on.

However, the nature of engineering practice means that we do not always have the luxury of creating new devices from scratch. Sometimes, the challenge is to take a device which already exists (and which is subject to certain restrictions on how much can be changed, in the structural sense) and make it more robust -- in a sense, make it more capable of surviving in environments that are presently causing it to fail.

The "symptoms" of an EMI problem can be extremely bizarre (again, adding to the "black magic" mystique) -- monitor screens wavering or blinking, sudden and random resets, shutdown when the unit is facing one way but proper operation when it is facing another way, etc. Sometimes, when troubleshooting an EMI problem, this troubleshooting entails things like wrapping devices in aluminum foil, shifting them subtly from one orientation to another, and even just waving one's hands in front of the misbehaving device!

All of this, regardless of how strange it looks, can lead to information about the problem exhibited by the device. And sometimes the "weird looking" troubleshooting exercises can lead to the application of solutions that look even weirder, or that seem counterintuitive. In addition to this, sometimes the solutions to EMI problems on an older device -- one that cannot be redesigned from the ground up -- end up being stunningly simple.

Many EMI problems can seem overwhelmingly complex -- even intractable -- at first. Because the behavior "could be anything", because it's difficult to repeat, because different environments seem to have drastic effects on functioning -- some EMI problems have been known to make even very experienced engineers throw up their hands in utter frustration. The devices I've worked on tend to contain upwards of several hundred components -- each essential to the overall functioning of the device. And while this is certainly far less complex than, say, a biological organism, the number of combinations of parts that could be interacting with each other, or responding to environmental influence, is huge. I've heard people describe EMI problems as seemingly "hopeless", because there is "just no way" we can isolate the exact cause of the problem and fix it right at the source.

But -- the trap in this reasoning is that often in engineering, you aren't looking for a perfect or even an elegant solution. You are, instead, looking for a solution that passes the test. In the case of EMI, there are numerous possible standards a device could potentially need to meet -- and being aware from the outset of a design or troubleshooting challenge as to the conditions the device needs to meet can drastically speed up the design or troubleshooting process.

Making the analogy to aging interventions again, a "perfect" solution is not necessary, and there is no reason that SENS can be claimed untenable on the basis that it fails to fully address the complexity of an organism. The "test" that an aging intervention needs to pass is not that it entails a complete understanding of the mind-bogglingly complex nature of an organism, as well as everything that could possibly go wrong in every possible situation -- but that it allows an organism to maintain optimal functioning long after evolutionary destiny would predict.

It is very, very important to distinguish between these two goals. You don't even need to know what a decibel is to figure out that a piece of cleverly-placed foil can make a horrifically misbehaving device suddenly start working again, and sometimes, the reason for why such approaches work is determined after the fact that they do indeed work has been established. However, this does not mean that no understanding of the principles of physics is needed in order to fix an EMI issue, just as approaches like SENS cannot be undertaken without a good understanding of biological processes. The key is to develop a "good enough" understanding.

An engineering approach to the problem of senescence is almost certainly the quickest means to a potential working solution -- because, as de Grey points out, technology allows us to "sidestep ignorance". If I had to solve every EMI problem through performing detailed analysis on every circuit in a system and charting all the possible things that could ever go wrong, and listing all the possible reasons why a given symptom might be observed -- before actually doing anything about the problem -- I don't think I'd have a job for very much longer.

There is another useful parallel between EMI design and the notion that SENS attempts to address areas of aging that may or may not be proven deleterious in the long run. It is quite possible to build a functioning device that has carelessly laid ground traces, or poorly configured internal shielding, that still functions.

It is possible that maybe the device doesn't need all those decoupling capacitors. But if a particular feature has been identified that in some ways distinguishes devices that pass EMI tests from those that do not pass, and this feature is cheap and easily applied (or comes as a "package deal" with known-useful interventions), then there seems to be no good reason to avoid putting in that feature as a "just in case" measure. It can't hurt, and it might help.

The same could be said for SENS as a whole -- it doesn't seem that funding this research could hurt (and if anyone makes the argument here that it would represent an inappropriate misallocation of resources, I'd encourage them to direct this argument to those responsible for putting on large-scale sporting events, producing designer shoes, developing sports-utility vehicles, producing tabloid magazines and newspapers, and numerous other examples of "unnecessary" human endeavors).

At worst, SENS will be revealed, through rigorous peer review and possibly experimental practice, to be a failure. At best, it will prove to be the means for many individuals to experience longer, healthier lives. I honestly don't see what any one of us has to lose by supporting approaches such as this (while at the same time acknowledging that more tenable and valid approaches may emerge over time).