Friday, August 12, 2016

Complexity and Silicon Valley

The path towards understanding and appreciating conservatism in Silicon Valley is through complexity. Here, let's start at the beginning. Complexity as it is understood here starts with MIT Professor Edward Lorenz and his discovery of the error amplification effects of feedback-based systems. While attending complexity conferences, one hears the team "complexity" repeated too frequently, almost like a mantra. For Lorenz though, the complex combination of nonlinearity, stock-flow (integrative), and feedback effects in his 1960s-era different equation-based models led to unexpected amplification of small initial differences.

 At approximately the same time, in 1965, Friedrich Hayek applied new notions of complexity to politics and economics. Those who have been paying attention know Hayek from two career highlights: (1) The Road to Serfdom; and (2) his becoming a Nobel laureate in Economics in 1974. Rather than running though this work, let us instead focus on the matter at hand, how does complexity impact and inform conservatism? Complexity generally, and the complexity of socio-economic systems specifically, provides hard limits to what is knowable about these systems. So as political entrepreneurs and purported innovators make claims in the  marketplace of ideas, complexity places limits on what can be said.

The problem comes in that these same entrepreneurs and innovators don't want to be constrained by what they can say. Hayek explains at length how such claims, especially as they relate to and regard socialism, have demonstrated themselves to be false. Your correspondent agrees with Hayek's conclusion but will leave that argument for a later time.

What is especially informative is how these political entrepreneurs respond to complexities  limitations: in the words of Tom Wolfe, they scream like weenies roasting over a fire. The best example I've found is Herman Finer's The Road to Reaction. It's a remarkable example of what I call an "affect storm," or what others call "argument by outrage." While purporting to be logical and rational, Finer's book is a collection of half-thought-through accusations, what Hayek called, ""a specimen of abuse and invective which is probably unique in contemporary academic discussion." It's instructive to consider the incentives behind Finer's comments, and here we need to distinguish between good reasons and real reasons. Finer, being a Fabian socialist, portrays a concern for society's poorest. Here your correspondent draws insight from Tom Wolfe who credits his prescience to a focus on status. So in your correspondent's opinion, Finer's affect-laden response to Hayek is driven by the implications and consequences of complexity: that is, he can no longer claim special insight into socio-economic systems, which would limits his status and power. This in Finer's view, is clearly unacceptable.

 But much of the politics since the 1940s has been driven by just this tension between what is knowable and what is not as limited by complexity, which is a conservative concept. And that is what we see amplified in 21st century politics: this analytic tension between what can be known, and what is not knowable. The computer revolution, and represented by Silicon Valley, has something useful and important to say about this complexity by pushing the boundaries of what is knowable. However, this can only happen if senior policy makers and those interested in policy acknowledge these limitations, which is controversial and problematic.

What's more conservative, EE or CS?



As a Silicon Valley (SV) conservative, I was thinking to myself, “Which is more conservative, electrical engineering (EE) or computer science (CS)?” Let me allow the reader – whoever that might be – to consider the question briefly before continuing. (Jeopardy theme)

Before diving into the answer, let’s first establish the conceptual context. The foundation of conservatism -- and through inverse causality, liberalism as well – is the fact-value distinction, with conservatism associated with the fact side and liberalism with values.

So how do EE and CS compare in the fact-value distinction. The answer should be pretty obvious to anybody who’s studied both, EE is more fact oriented because it is a branch of and grounded in physics, while CS is more virtual, conceptual, and abstract. With EE, if you’re unsure about the ohms (units of resistance), farads (capacitance), or henrys (inductance) of a device, one can measure it. With CS, if you need more of whatever it is – memory, objects, or agents – one just defines more. Now there are limits to this argument, but CS is related more to math than physics – and then it doesn’t even need to be that related to math.

These observation might explain why SV is so overwhelming liberal, which is to say Democrat. EE is a discipline of design within constraint, which CS and the products associated with it, are comparatively less constrained by physics, or reality for that matter.

Monday, August 8, 2016

What does it mean to be an SV conservative?

One of the questions I cannot quite shake is this: Why is Silicon Valley so gosh-darned liberal? I grew up there and became a conservative of sorts. I grew up reading the San Francisco Chronicle and wondered at the insane decisions of so-called liberal adults whose politics and policies were ever so easy to criticize. But increasingly, instead of merely criticizing liberals, I increasingly wonder what it means to be a conservative?

There are a couple of insights that illuminate for me what being a conservative is. The first comes from the Gospels according to Matthew and Luke in which Jesus says that, "a tree is known by its fruits." My interpretation, in the "here" and "now", that names and reputations are based on the long-term consequences of a person, policy, or population. Note that this is very different from our modern-day media-driven democratic world of reputation based on short-term group opinion.

Second, Tom Wolfe's "The Great Relearning," provides multiple examples of the progressive ideas of intellectuals being implemented only to end in disaster. Wolfe's primary example is the sexual revolution, which was implemented with such fanfare but generated a of unanticipated, and sometimes anticipated, consequences.

Third, this focus on consequences leads, in turn, to a focus on complexity generally and causal complexity specifically. That is, what makes consequences "unintended" is that human cognition is not very good at anticipating the consequences of large policy changes because of the complexity of the social systems involved. However, power computers, the kinds created and used by Silicon Valley engineers, can be used to address and account for that complexity.

However, there is a division in Silicon Valley between analysis used for accurate prediction, such as that used for engineering design and corporate operations, and that used for sales and marketing. That is, sales and marketing rely more on affect and emotion rather than cognition and accurate analysis. And here is the key division between standard liberal and much rarer conservative beliefs in Silicon Valley: the former generates sales and votes based on affect, while the latter generates more accurate analysis based on accurate cognition and accounting for complexity.

As the costs of affect and profit-driven policy analysis mount and become ever more apparent, the need to engage in accurate, engineering design-like policy analysis will increase, but given the tenor of America's 2016 presidential election, that time is still some ways off.

Saturday, July 16, 2016

Hail Caesar!

Philosophy, region, and the philosophy of religion can appear in the most unexpected places. Why just the other week, I was watching the movie Hail Caesar!, when a insight hit me. The movie features both priests discussing the Trinity and communists, led by Herbert Marcuse, discussing the dialectic, which is itself unusual for a movie. The latter made me nostalgic for graduate school. However, the nature of both concepts -- the Trinity and Hegelian dialectic -- are complex, which means "hard to explain." In both cases people -- both Christians and communists -- argue hilariously. The sum total of both debates can be summed up by the Rabbi: "These men are screwballs!" 

But the key insight is that, viewed from a certain perspective, neither the Christians nor the communists are screwballs. Instead, both are using the ill-suited but until recently the only tools available -- propositional logic and prose -- are inadequate for the task. Instead there are what are called feedback relationships that pervade, comprise, and control social systems including political systems. These relationships, being complex and essentially mathematical (actually differential equations), are pervasive in electrical engineering but not so much in religion, politics, philosophy, or economics (PPE).

The question then becomes, how best to bring this insight into PPE? There are several insights that bear mentioning. First, the temporal and causal perspective is increased beyond what science traditionally can handle. Successful science experiments usually are rigorously controlled so that their causal factors and results can be clearly identified. While a useful enterprise when skillfully executed, such experiments are of limited utility for complex social systems that evolve over extended, multi-generation time periods.  Recent results on wicked problems have begun to characterize how to think about complex, temporally extended policy problems. Narrative techniques are especially helpful to think about social structuring and policy problems, as with the Bible, but more about that later. 

 

Sunday, July 3, 2016

Dogmatic Complexity

I've been reading Francis Hall's 10-part Dogmatic Theology -- though certainly not all of it -- and Ian Barbour's Religion and Science. As I read these two books, I find them, in some sense, fundamentally unsatisfying because they're primarily descriptive. That is, they decompose or split the problem into lots of separate parts and offer lots of definitions, but while useful, it doesn't really address the problem. It doesn't get at the underlying pattern, connections, or reality that underlies religious thought. These different pieces must instead be lumped together. Now critics will say it's impossible, and I get that, but I have an intuition that we know enough about complexity, complex system, and complex social systems that we can say something innovative and useful about both religion and policy.

They key to this project is lumping instead of splitting, but how does one do that? One must first identify a "way in," and that's provided through philosophy. There are a number of words that for me provide a way in though:
  • Teleology, which has been criticized as confusing causes and effects
  • Trinitarianism, the concept of Father, Son, and Holy Ghost which postdates the Bible
  • Dialecticalism, Hegel's concept of thesis, antithesis, and synthesis
The way through to think clearly about these abstruse terms is through complexity, which is in turn defined as nonlinear, stock-flow (integrative), and feedback causal relationships as tell as network relationships. Complexity is used dogmatically here in that the underlying mathematics and physical patterns associated with complexity are assumed to be true and applicable. The modern thinkers who for me underpin these thoughts are Jay Forrester (system dynamics), Norbert Wiener (cybernetics), Friedrich Von Hayek (economic control), and Herbert Simon (bounded rationality) though there are many who have contributed to this literature.

Negative, balancing feedback relationships result from an odd number of negative, which means change in the opposite direction, causal relationships. For example, the more people, the more deaths, but the more deaths, the fewer people. Marxists tend to say such relationships are "internally inconsistent," though a proper understanding of feedback relationships shows that they are both ubiquitous and confusing. Hence the lack of clarity that surrounds religion, policy, and the social sciences. These fields are all doubtlessly complex, but only recently, with the advent of powerful computing, are the tools available to address and account for this complexity. One limitation is that those who study social science usually have little expertise in the computational and mathematical techniques that underlie complexity. How far the complexity sciences can uncover, reveal, and inform social systems remains to be seen, but it is an endeavor that will certainly be worth the effort.

Saturday, June 25, 2016

Complex Christianity

The desire to reconcile science and religion has been a goal since at least the Renaissance. Machiavelli's modernism broke with Europe's religious past by explicitly distinction between positive and normative -- doing what works as opposed to what the Church says one should do.  Swedenborg sought to reconcile explicitly reconcile science and religion. Today others undertake the project to reconcile science and religion from a literary perspective, while others, such as Sam Harris, adopt a scientistic -- that is, an exaggerated and inappropriately applied -- perspective to attack faith and religion.

The problem, as Sam Harris should understand, is that science as it is practiced is generally inapplicable to the social sciences, philosophy, and the humanities. Science is based on the experimental method, which Richard Feynman pithily defined as first guessing a new law, second performing an experiment, and third, comparing the guess to the experiment's results. If the guess doesn't square with experiment, then it's wrong, and that is the key to science.

What's great for particle physics though may not be appropriate for public policy due to the complexity of the latter. That is, with longer time-frames and more moving parts than a physics experiment, public policy in the real world is a different animal altogether. Christianity understands and accounts for this when Jesus noted that a tree is known by its fruits in both the Gospel according to Luke 6:43-45 and Matthew 7:15-20. These two passages recognize that there are short- and long-term consequences that can work in opposition to each other, and it is the long-term consequences by which policies and actions should be judged. This are confounded and made unclear by the inherent complexity of social systems.

There are multiple examples that could be developed the demonstrate the importance and unpredictability of long-term consequences:
In each case there are a clear set of benefits that are promised, and yet the New Class scientific policy planners, socialists, and decision makers don't deliver what they promised. And yet, Sam Harris does not examine these failures, which are legion. Christianity, in contrast, is the tree for the fruits for countries in which people want to live. Understanding Christianity in terms of complex social systems seems to be entirely possible given the technical state of the art, but it is certainly not of interest to and will not be rewarded by Silicon Valley, the elite academy, or DC. Nevertheless, it seems to be a worthwhile goal to pursue.

Friday, June 24, 2016

Engineering as Policy Dogma

One of the enduring questions that I have is this: Why are intellectuals generally and Silicon Valley specifically so darned liberal. I don't have any real good answers to this, but I do have some thoughts regarding why engineers should be conservative if you consider that to mean being on the fact side of the fact-value distinction.

The foundation for this argument rests in dogma, a set of beliefs that are not questioned. For me, as an engineer, my dogma is engineering. I believe what I learned as an engineer because I've proved to myself that these rules work. Specifically, I refer to control theory, information theory, and computational theory, all of which impact policy and influence policy resistance through complexity. Now I've had arguments with philosophy professors who take as their dogma democracy theory, social justice theory, or critical legal theory, but none of these can come close to the trust or reliability offered by engineering. If the engineering 3 are incompatible with the philosophy 3, then that says more about the latter than the former. After all, the engineering 3 are all based on logic and rules, so why should they be incompatible with philosophy?

So how this argument manifest itself? First, Hayek in his 1967 essay "Complexity" talked about how advanced math and computation impacted policy and analysis. Second, Jay Forrester's system dynamics explicitly applied engineering to policy problems by accounting for aspects of causal complexity such as feedback relationships, stock-flow (integrative) relationships, and nonlinear relationships. Each of these causal relationships can confuse the human mind, but all three combined make even fairly simple policy problems hard to predict.

But you see politicians, professors, and policy professionals regularly acting like they know it all when they don't. This lesson was brought home to me when I had coded up a fairly large software system--over 5000 lines--and I didn't know how it would react when I changed it. So I made changes very carefully and always ensured I could get back to a knowable state when making changes--that is, I took pains to conserve the system. I took these precautions even though I knew more about that code base than anybody else on the planet and the code system was simple compared to social systems. And yet so-called experts and elite leaders regularly recommend massive, huge, and irreversible changes to systems about which they know very little. Any disinterested analysis would reveal that such changes are good for them but probably not so good for the system itself.

Thursday, June 23, 2016

Brexit, Conservatism, and Complexity

In 1995, Gerald Schneider and Patricia A. Weitsman edited a volume entitled, Towards a New Europe: Stops and Starts in Regional Integration. There's a model in that volume that seems particularly germane in the present day showing that the European Union (EU) was started based on the short-term benefits of integration, while the long-term costs get pushed off and forgotten. Well, those costs are now coming due, and the fact that they're long-term doesn't make them any less real.

I was also thinking about my time at SEU and how I expressed and pursued my conservativeness. I didn't memorize and repeat speeches by William F. Buckley, Ronald Reagan, or Margaret Thatcher. I know who Edmund Burke is, but I haven't read him extensively. I read Leo Strauss in school, but that was later. No, my conservativism was rooted in engineering -- that is, doing what works and what has been shown to work. It seems to me that a policy that impacts hundreds of millions of people should be crafted with the same care that an engineering company makes its products, say Apple's iPhone for instance. However for politics this is not the case as was made vividly clear by Ben Rhodes' explanation of how he helped Obama pass his Iran policies. Very little care was given to the costs and consequences of these policies, but much care was given to the arguments, talking points, and passage strategies. This is a problem.

At the core of this issue is system complexity, which manifests itself as policy resistance. That is, policies often do not achieve their intended benefits and the consequences can be very different from those originally envisioned. The unintended consequences of these complex social systems being influenced are at the heart of conservatism, which involves the interaction between senior decision makers, their analysis teams, and an external system. This interaction is so fundamental, but it's not well understood or entrenched in the policy community.

Hayek wrote memorably about such matters in the Road to Serfdom after World War II in 1945. Without going into that particular argument, there are two consequences that I do want to highlight. First, there is the response to Hayek by Herman Finer, Road to Reaction, which was also published in 1945 and is remarkable for its over-the-top affect and almost complete lack of argument. It's amazing because Finer, a political scientist and Fabian socialist who taught at Harvard, basically calls Hayek every name in the book for opposing socialism and it centralized planning to which Finer might have aspired. Perhaps more amazing is that nobody really seems to have had a problem with Finer's argument, or lack thereof, because his colleagues all agreed with him. Second is Hayek's essay "Complexity" that was published in 1967, which is amazing for its prescience and lucidity. He works through many of the policy details and implications that were later used by Herbert Simon and Elinor Ostrom (note that all three of them won the Nobel in Economics in 1974, 1978, and 2009 respectively.

In rereading Simon's 1983 Reason in Human Affairs, he notes that being math and empiricism to these topics is always controversial, which strikes me as odd because it should be fundamental, but it's not, which means it's a topic worthy of further consideration. Perhaps most controversial to successful and intelligent academics is that their cognition or rationality is bounded in relation to the systems their trying to impact and whose behavior they claim to be able to predict. The model that compared and constrasted the short-term integration benefits that helped form the EU and the long-term costs that now threaten it provide just one example of the limited capability of policy makers to impact complex social systems in a beneficial and long-term manner.

Monday, June 13, 2016

The November Plan

Was just reading an article on the upcoming Olympics in Brazil and noting, yet again, how socialist promises are going unfulfilled and have instead resulted in tragedy. A simple comparison of the promises made by the European Union (EU), a favorite topic of Harvard's Center for European Studies and whose out-of-control immigration is threatening Europe, Brexit, in which Britain is collapsing, as well as Venezuela, where people are rooting though garbage looking for food -- and these are just the most recent examples of the long-term consequences of socialism. So why is socialism so popular when its track record is so horrendous? In fact, I would say that the relationship between the short-term appeal of socialism and its eventual long-term failure is the most predictable relationship in politics, even more so than the much beloved democratic peace theory.

The question then becomes, why is it that this highly predictive and consequential regularity is so unstudied? I would even go further to say that why is it that professional politics are actively hostile against questioning socialism? The answer is that political elites benefit disproportionately from socialism. Historically Hayek's critique of centralized planning, The Road to Serfdom and The Intellectuals and Socialism, is an instructive starting point. But Hayek's thoughts on complexity point the way toward a more modern critique of socialism and understanding of conservatism. But a review of Venezuela's socialist experiment shows that the Chavez family is fabulously wealthy as their country starves. And recent research shows that so-called political experts seldom deliver what they say they will and are rarely called to account for it. This highly predictive failure of socialism -- including its metastasized variants progressivism, communism, leftism, Fabianism, globalism, and extreme democracy -- is why I call it "The November Plan," because after it passes through, nothing lives. Making the case why this is the case and having people understand the consequences of socialism is the most pressing issue not only for intellectuals but, more importantly, for the citizens of the countries who implement these inevitably doomed-to-fail November Plans.

Sunday, April 24, 2016

Straussianism and Philsophers

Studying with Prof. Paul, a Straussian, was very different from studying with Prof. Irving, but understanding and articulating the differences makes for an interesting exercise, which I'll undertake initially through the fact / value distinction. Often in graduate school, we would discuss the distinction between "positive and normative," that is, "the way things are rather than the way things should be." This continuum manifests itself a tension between facts and values, with facts being inarguable and values, in contrast, being open to distinction. Both are unarguably part of the human experience, but the key difference comes in reconciling this tension, a process that is as old as philosophy itself. However I view the fact-value distinction as a tension between complexity and simplicity, which from an engineering perspective is obvious, but that perspective is not quite as universally understood as it might seem to an engineer. That is, reality is complex, while the human mind can only comprehend a small portion of that reality, which in turn drives thphilosophical dynamic. 

 There are a number of examples of this tension between simplicity and reality, between clarity and complexity. One of the earliest and most clear is that between Plato's Republic and Laws. The Republic was written early in Plato's career and is clear, straightforward, and direct. He makes sharp distinctions between topics, which makes his arguments easy to understand and remember. However Plato's Laws was written later in his career and was probably more true but less clear. Plato said, "Sometimes it's this way, but other times it's that way..." yawn. In contrast to Plato was his student, Aristotle, who was more a natural scientist than his teacher. Even though Aristotle is remembered for all he got incorrect, it was his perspective on and interest in the natural world of cosmos that is of particular interest for this author. Most significant is his acknowledgement of the complexity of the natural world, which is reflected in the political in terms of excellence or virtue. The world does not present people with well structured, closed-form solutions. The world is more complex and thus different.

This division between the human and natural, between the simple and complex, between values and facts manifests itself as a tension throughout the history of political philosophy. Modern philosophers feature an exemplary pair in Kant and Hegel. Although it's been a few years since I read either, the key difference is between the simplicity, values, and rationality of Kant and the complexity, facts, and empiricism of Hegel. Kant performed thought experiments, wrote much, and seldom strayed from his university. Hegel, in contrast, tried to explain the complexity of the world he saw around him. His most famous attempt to do this is the dialectic composed of thesis, antithesis, and synthesis. Of course the communists took Hegel's ideas and perverted them as communists tend to do, but I argue that Hegel was getting at something fundamental about the complexity of the real world. His work on historicity and categorization structure support this conclusion. The real insight is that Hegel did his best with the tools, concepts, and ideas that were available to him. Computer-based tools provide undreamt of capabilities today, so it seems that we should at least try to apply them to these important philosophical and policy questions. 

These tensions manifest themselves in the present day as well. John Rawls of Harvard wrote A Theory of Justice, in which there's an "original position" in which you don't know if you're rich or poor. This rational, thought-based style is called what makes Rawls a "neo-Kantian." In contrast to Rawls I offer Leo Strauss who places more emphasis on facts, history, and tradition, making him more of an Aristotelian -- that is, he recognizes that his writings will not capture the great complexity of reality, and such efforts will come up short, but it's worthwhile and beneficial to try. 

Extending from the 20th to the 21st century, the great cognitive prosthetic power of the modern computer can be applied to philosophical and policy questions as a way to manage the complexity inherent in them. However, doing so requires bridging the cultural gap between philosophers and engineers, between social scientists and computer scientists. As an engineer and computer scientist moving into the worlds of philosophers and social scientists I expected the divide would not be that large, but I was wrong. 

    Saturday, April 16, 2016

    Straussianism and Conservatism


    I decided not to get too worked up about Prof. Irving. This wasn't so much a choice as a realization  that a masters student fighting a professor with tenure a SEU was a no-win proposition. Better to spend one's efforts elsewhere, which is a trick I learned from the board game go: if you are in a bad position on the board, stop playing there and come back to it later when the situation may have improved. Another variation on this theme is, "If you're in a hole, stop digging." I decided to focus instead on the upcoming spring semester's classes. They call this "shopping for classes," and as I was doing so I ran into a colleague. Cathy was from Eastern Europe, an ardent socialist, and a PhD student who was on scholarship and could have gone to another elite PhD program. As a baby masters student with a couple of engineering degrees, I looked up to her. Cathy was a little too serious, self-important, and argumentative, but this was graduate school, and she certainly wasn't the only one.

    Cathy was going to take a class from a Straussian philosophy professor, Prof. Pauls. I said yes for one and only one reason, because I knew Prof. Irving hated Prof. Pauls with a white hot passion, so that was good enough for me: whatever he was selling, I was buying; whatever he was laying down, I was picking up. There was only one problem: I didn't really know what a Straussian was. As with most subject in philosophy, and here we're talking about political philosophy, there are multiple levels to uncover and examine. First, Straussians are people who follow in the footsteps of Leo Strauss, a professor at the University of Chicago, one of the key super-elite universities in the world. He wrote the book Natural Right and History, which I must have read as a graduate student but I remember almost nothing about. There's also Cropsey and Strauss's History of Political Philosophy, which we called "the purple Bible" and provides synopses of the west's great philosophers. But it's not what Strauss wrote so much as the way he read. You'll often hear the phrase, "close-reading Straussians," which highlights the perspective of a reader trying to understand what the writer was trying to say at that point in history rather than what the reader's modern interpretation might be. As an engineer, I interpreted Strauss as acknowledging an objective truth and trying to understand it however difficult rather than holding that the material world is up for debate, interpretation, and innovation. Straussians are often associated with Aristotle, who as an ancient tried to understand the natural world and got almost everything wrong, but it was his perspective and seriousness that left it's mark. But unlike an engineer, these Straussians tried to understand the world with words rather than numbers, an attitude that I appreciated and a method that left its mark on me. My engineering degree taught me math, my computer science degree logic, and my time in political philosophy, English. 

    But the real question is this: What is the relationship between Straussianism and conservatism? Conservatism is something beyond just "close reading," but it's that Christian perspective captured by the Gloria Patri doxology, "As it was in the beginning, is now, and ever shall be: world without end," which gets at the universality and reality of physics and engineering. If reality is that which persists after one stops believing in it, then physics and engineering capture that Straussian, conservative perspective: that reality is something to be learned, understood, and shaped within limits rather than something to be battled, dictated to, and shaped according to the desires of people. This is a working definition of the relationship among Straussianism, conservatism, and realism as embodied by physics and engineering. It's a topic that will be addressed again.