Skip to content

Our World Turned Upside Down

2011 August 28

Lately, the world has felt out of kilter. Things don’t conform to the rules we’re used to, but spin out of control, seemingly at random.

We’ve witnessed the Arab Spring, where entire cultures turned on a dime, rose up and overthrew regimes that had ruled for generations.  Financial crises start with obscure acronyms and reverberate throughout our society.  We appear to be in uncharted territory.

But is that really so or are we just becoming aware of realities that until now we’ve been blissfully ignoring?   Before the digital age, we were able to sequester our organizations behind legal and organizational barriers, staving off the tides of evolutionary change. That’s becoming less and less tenable.  Our challenge now is to face those forces head on.

The Old World

On New Years Day, 1801, Giuseppe Piazzi discovered the dwarf planet Ceres in the winter sky.  It was only a fleeting glance and soon disappeared from sight.  However, a young math whiz named Carl Friedrich Gauss was able to calculate where it would show up next.

A short time later, Ceres reappeared in the sky exactly where the 24 year-old phenom predicted.  This miracle was made possible by Gauss’s method of least squares, which fits messy data into a smooth pattern.  At the heart of the method is the idea that errors in observation are random and therefore are normally distributed like this:

This concept, encapsulated in the Gauss-Markov theorem, forms the basics of modern statistical analysis.  We use it to predict everything from defects in factories to defaults on loans.  Mathematical finance, which drives the models that run Wall Street, owes a lot to that little planet.


Nearly a century passed when, in 1906, the first flaw was found in the gem of mathematical perfection that Gauss had spawned.  It was in that year that Vilfredo Pareto published his famous study on wealth distributions and declared that 20% of societies tend to possess about 80% of the property.   

This principle, named after its discoverer, follows a power law distribution that looks like this:

Another half century went by before a young researcher at IBM labs named Benoit Mandelbrot found very similar distributions in signal distortions in communication lines. He didn’t stop there though. Before long, he was finding power laws everywhere, from the flooding of the Nile river to word distributions in texts to financial markets.

More recently, Chris Anderson of Wired magazine noticed that digital technology was bringing power laws to he fore.  He argued that improved market mechanisms were making the low volume products that make up 80% of commerce newly profitable.  He dubbed this phenomenon The Long Tail.

The Inexplicability of Interactivity

So why do things like communication lines and markets and word distributions act so different than the celestial bodies that Gauss applied his famous formula to?

For that matter, why is that when we look at people in terms of height or weight or strength or speed they look so much like Gauss’ bell curve, but when we look at how much money people have or how they spend it we get an 80/20 rule?

The answer is feedback.  While inert objects simply bounce off of each other, entities in living systems influence one another. The rich really do get richer. While nothing succeeds like success, failure often begets more failure.  As Malcolm Gladwell pointed out in his book Outliers, a small advantage can snowball into dominance over time.

It’s important, crucial in fact, to make the distinction between inert systems and ones that feed back on themselves.  A key factor in the latest financial crises was that, out of mathematical expediency, the wizards of Wall Street were applying Gaussian models where they clearly didn’t belong.  They underestimated volatility and we all paid a price.


In the past, we didn’t need to worry about power laws much.  The frictions inherent in our old, analog way of doing things acted as safety valves. We were able to adjust to changes before things got out of hand.  However, the speed of our new digital reality only tells part of the story.  We not only have more interactivity, but more things interacting.

The World Bank estimates that, between 1981 and 2001, 1.5 billion people rose out of poverty.  Since the World Trade Organization was established in 1995, 153 countries have joined?  We have more consumers and more firms in the global economy who all get votes on what gets produced, who makes it an where.

So what happens when you have lots of people making choices which effect everyone else? Well, as Thomas Schelling explained in his 1978 classic, Micromotives and Macrobehavior, something totally unexpected.  One person making a choice is decidedly different than 1 million people making the same choice.

If I decide to sell stock it’s no big deal, but if enough people have the same idea the market crashes, layoffs ensue, commerce sputters and so on.  When Justin Bieber’s mother liked his singing, it was a family affair, but when enough people on YouTube agreed, others heard about it and joined the party.  In a riot, even the righteous have been known to break windows.

The Right Side Up World

All of this begs the question:  Has the world really turned upside down or is if finally going right side up?  Through millions of years of natural history, organisms have come to dominate only to be then driven to extinction by forces beyond their control.  Why should we be any different?  Has the world changed, or just our perception of it.

I would argue the latter.  Mandelbrot pointed out the probelems in the models of mathematical finance as early as 1964.  Phillip Tetlock, in his 20 year study of political experts, concluded that their analyses were scarcely better than flipping a coin.  Look at just about any activity and the vast majority of people consider themselves above average.

The inescapable truth is that in our digitized, networked age we can no longer ignore uncomfortable realities.  We can’t take solace in the idea that things will even out or follow the pretty, clean models of statistical analysis.  Rather, they can blow out of all proportion or be laid asunder in the blink of an eye.  No one is safe.

Our world is messy, as it should be.  Anything else would be not only boring, but dishonest.

– Greg

15 Responses leave one →
  1. Larry Bridle permalink
    August 28, 2011

    Love your writings – the diverse, fundamental sources used, and your right-on applications to modern realities.

    Curious if you have looked at works by Thomas Homer-Dixon,, specifically his books ” The Ingenuity Gap” 2001, and The Upside of Down, 2006. I think it fits your genre of thinking and writing.

    So I’d be very interested to hear your take on how his ideas fit/ hopefully add to helping us all get thru..

    Thanks for enlightening blogs. I do find them useful.

    Cheers, Larry

  2. August 28, 2011

    Thanks for the tip! I’ll check it out.

    – Greg

  3. August 28, 2011

    Hi Greg,

    How are you?

    Yet another blog. The scope of this and similar posts are very much in the domain of my interests as well.

    We all have to “blame” the obtuse Louis Bachelier who thesis supervisor, the famous mathematician Poincare, found it rather average. His random-walk was forgotten and ignored.

    “Unluckily” for us, Paul Samuelson not only rediscovered it but developed and spurred whats is now orthodox financial theory with all corresponding tools by financial “geniuses” such as Fama, Cowles, etc.

    It is just that the normal distribution is only good for so far whereas wall street kids were applying it as panacea to all their number problems.

    Anyhow, we have seen the dark side of it now. Hopefully we will learn.

    P.S. btw, check my experimental blog – a sort of a cross-disciplinary roundup posts in 101 or less words. lemme know what u think of it.

  4. August 28, 2011

    Thanks! I love that “The Crazy Ones” text. Have you seen the video?


  5. August 28, 2011

    I don’t think it is true to say that the world flips between normal and inverse-exponential modes of behaviour. From my recollections of systems theory from thirty years ago, there are many modes but Gaussian-normal isn’t one of them. You touch on feedback: I think what you are referring to is runaway positive feedback.

    Not all rich get richer: many of them get suddenly poorer, so it is not feedback that makes this happen – it is success. I think inequity of wealth distribution has more to do with social structures for redistribution and tax enforcement than runaway feedback.

    All systems are a mix of positive and negative feedback (along with other elements such as hysterisis and others i forget). And no system remains linear – in the real world it always hits limiting constraints.

    Your basic premise may be true: that the world is inherently more unstable in the digital age because feedback happens faster and positive feedback is amplified. But the answers lie in more complex science than the model presented here: we need more damping (negative feedback) in our systems, i.e. controls.

    The modern world has indeed lifted billions out of poverty so clearly it works and shouldn’t be abandoned. What we need to do it tune it to reduce the number of unstable states it has – mathematical states that is, not counties 🙂

    sadly there is only so much we can do with the systems: human groups are inherently irrational and unstable. Justin Bieber’s success being a case in point. Bull runs being another: every now and then the whole crowd scatters for the turnstiles, no matter what controls you put in place. people are panicky by nature. But we can damp the effect with controls. Right now we don’t. As you say, we let the Wall Street quants take them off.

  6. August 28, 2011

    Yep yep!

    I have the link to it on YouTube (not the same as yours but the same video)! I just quoted it as I thought not only video but the words were great!!

    The timing of the ad aired was also quite interesting. It was aired in 1997 upon return of the prodigal “son,” Steve Jobs, in his bid to rebuild the brand Apple!!


  7. August 29, 2011

    Interesting ideas. Thanks for sharing.


  8. August 29, 2011

    I didn’t know that. One of my very favorites just got better!


  9. August 30, 2011

    Wrong. The theorists did not just use the wrong curve leading to the banking failures.

    The banks themselves are using the wrong methodology, a methodology that conflicts with the natural laws already are largely already known but I have had to restate and enlarge on my blog and in my workshops.

    is expected to correct this error along with my workshops, my new software for lending, ( a donor or an investor is needed for funding), and my forthcoming book.

  10. August 30, 2011

    IT Skeptic has the right idea.

    He will find more along those lines on my

    This is where we need leadership and clear vision. It is just there waiting for us to work together on it. May I show the lead here?

    My book will be dedicated to


    But honestly Greg, YOU ARE DOING A FANTASTIC JOB.

    I love your writing and the researches that go into them. How do you do it? I couldn’t!

  11. August 30, 2011

    Thanks for the tip.


  12. August 30, 2011

    Thanks for the kudos, Edward. Good luck with your ventures.


  13. August 30, 2011

    Thanks Greg.

    I have added a reference to this site on my blog and given credit to IT Skeptic where credit is due.

    My site is busy recruiting minds like his/hers so guys if any of you feel the urge, welcome aboard.

    Let’s move it, not just write about it!

  14. August 31, 2011

    hm.. interesting.

    I agree with IT Skeptic re need to differentiate negative and positive feedback. And what you say about rich getting richer and poor getting poorer is in popular folklore known as Mathew effect (from gospel of Mathew).

    As for a system to damp it, there is no system which can damp everything.. its like a lim() in mathematics – it can try to get closer and closer but will never do. Kurt Goedel is i think to be credited here for incompletness theorem (showing for first time how hitherto considered complete euclidean system was indeed not complete).

  15. August 31, 2011

    Good points. Thx.

    One small clarification – Godel ‘s incompleteness theorems were applied to the Principia Mathematica developed by Russell and Whitehead. It would apply equally to Euclid (or any axiomatic system), but as Non-Euclidean geometry was developed almost 100 years before by people like Gauss and Riemann, it would have been very much beside the point.

    – Greg

Leave a Reply

Note: You can use basic XHTML in your comments. Your email address will never be published.

Subscribe to this comment feed via RSS