Fair Warning: Do not embark on working to learn this process unless you're willing to rewire your brain about pretty much everything you were taught in architecture school. It took me a few years of pattern-hunting to rewire my brain a quarter-century ago, and I can attest that it's really hard work because what you're about to read runs counter to all conventional wisdom… especially what's in the last half of this post. Since then, I've worked with some of the most brilliant minds in architecture, and we've all faced the same difficulty to rewire our brains. But if you succeed, this process will pay many dividends for years to come.

There are countless types of patterns; some are patterns of human behavior, others are patterns of behavior of other creatures, others are patterns of behavior driven by natural laws, like the law of gravity reliably pulling everything unattached to fall to the earth. But this post focuses only on patterns of the built environment in a particular place, from the scale of a neighborhood to the scale of a region.

A pattern is something that occurs repeatedly. But when you begin a pattern-hunt, you don't know what happens repeatedly; everything you see when you step out onto the street is just an instance, not a pattern, the first time you see it. It is only after seeing enough instances of the same thing that you can no longer count them on one hand that you should say to yourself "I think this might be a pattern." At that point, sharpen your eyes to see more of them, including types and sub-types of what you believe to be a pattern. Once you can no longer count the pattern and its types and sub-types on two hands, go ahead and consider it a pattern. This first step is by far the easiest step.

The follow-up step is to get better acquainted with your newfound pattern because the ultimate goal is to get well-enough acquainted that you can tell stories about it. So as you're getting better acquainted, keep asking yourself "why might the old folks have done this repeatedly?" And by "old folks," I'm mostly talking about dead folks because most patterns we see today are industrial, like "hey, there's another black Kia Sportage." And industrial patterns are positively boring compared to the old folks' patterns because the industrial patterns are so completely identical that they can be identified by their make & model or by their barcode, whereas patterns produced by living traditions of humans have a great variety in a narrow range, which has two powerful benefits:

The narrow range makes clear the problem or problems the pattern was meant to solve. Patterns employed by living traditions aren't just superficial style; they're there to get useful stuff done.

The great variety is a sign of life. Nature does it best, like leaves on trees of a certain species, or fish of a certain species in a pond. But humans produce patterns as well, and the great variety gives us abundant hints as to the multitude of means all those old folks employed with this pattern to solve whatever the problem was that spawned it to begin with.

These two benefits combine to answer an old question first posed by Christopher Alexander a half-century ago: what is this quality identified with almost all of the world's greatest built environments and their surroundings? He never proposed a name, calling it instead the Quality Without A Name, or the QWAN to Pattern Language insiders. And I'm not proposing "great variety in a narrow range" as the long-lost name. Instead, it is the combination of conditions that creates the QWAN. That's why pattern-hunting is so important; if we want to learn how best to build great places for this people in this place, we really need to hunt up the useful patterns with which to begin.

So let's get back to the story-telling. The first step once you've gotten better-acquainted with your pattern is to ask "why did the old folks do this?" As just noted, look to the narrow range of the pattern as the best evidence of its reason for being. Once that's established, look to the great variety as noted above for the wealth of stuff the old folks tried to fulfill the pattern's mission.

Given the purpose and its wealth of approaches (the great variety) for achieving that purpose, ask yourself "does the purpose of this pattern hold meaning for those alive today? If the answer is yes, move on to the next step. If the answer is no, this algorithm has another branch: Is there another purpose (or purposes) with which this pattern might have meaning for those alive today? If the answer is yes, move on to the next step noted above, and do the work noted below. If the answer is no, however, then let it go.

Some words in a spoken language fall out of use over time, so it should come as no surprise that patterns in a pattern language fall out of use as well. If your goal of pattern-hunting is to write a pattern language book, then you must stay firmly and deeply committed to writing a book for the living, not a book of the dead. This is the hardest thing all those brilliant minds I've worked with have failed (so far) to overcome; they're all drawn back to thinking like historians describing historical things rather than a story-teller telling stories to the living of things meaningful today.

So at this point your job, should you decide to accept it, is to craft a story rational enough, compelling enough, and inspiring enough that those alive today make better choices as a result of your story. So be a story-teller, not a historian. Here's a good example of why and how to do so:

Consider window lite patterns. If you ask a historian about them, they'll likely launch into a monologue on glass-making technology in 1820 versus 1860 and how that affected lite patterns. Other historians would eat that up, but for everyone else, their eyes would glaze over. But if you instead explain how the muntins that divide the window lites diffuse light that comes into the room, and how that diffused light makes everything in the room more beautiful (including the humans) then you've given them a real benefit to those living today, their eyes open wide instead of glazing over. If you then explain how each muntin added to a window increases the cost of the window, you have now given them both sides of the cost-benefit equation, so they can make a rational decision on how many muntins they want.

So that's it. Pretty much. But don't discount the immense mental inertia of our training to think like historians instead of story-tellers, because it is strong. Just remember; these are stories for the living, not stories for the dead… because the dead aren't listening. But the living are building; let's do our best to help them build better.