Architecture and mental health crisis: two case studies

Recognizing how urgent the need is to shelter and support people in mental health crisis, do we have time for architecture?

Can't we just make a box and put help (or people) in it?

← Back To IA Theory

Big cities have higher concentrations of everything, including humans in crisis.

A mundane crisis we’ve all experienced when visiting a big city is having to go pee. And yet, public infrastructure to deal with this inevitable aspect of our humanity is mostly nonexistent. So too, with mental health crises: the public infrastructure to support what’s inevitable with us humans (especially in big cities) is mostly missing. It’s largely private charities that build and operate what’s available in big cities in the US, and consequently what’s available will always lag behind what’s needed.

Case #1 - Physical Shelter For People In Mental Health Crisis

In a 1989 documentary film about the work of the architect Christopher Alexander, the filmmakers read us in to the story of one such private charity in San Jose, California, that sought to build a shelter for homeless people with mental health problems.

One of the main threads of the story concerns the difficulties this charity experienced working with Alexander. 21 minutes in, we hear the executive director of the organization tell the filmmakers, with palpable regret, that many of the people his organization was founded to serve had remained unhoused during an exceptionally cold winter season due mostly–it would seem–to Alexander’s inability to compromise on matters of design, construction, budget etc. He wonders aloud if it wouldn’t have been better to just “build a box” to put these poor people in.

The filmmakers do a good job of contrasting Alexander’s testy contention that the process must take as long as it takes, and cost what it costs, with the charity director’s skepticsim that all this architecture mumbo jumbo is ultimately “worth it.”

But alas, the architect’s value proposition around all that mumbo jumbo is never proven out, because the film ends before the building was completed.

And, my gosh, what audacious claims about value Alexander makes in this film! And in the many thousands of pages in books where he elaborates his theories. Pages and books and theories that I have come to love dearly after having the opportunity to see for myself, and to hear from some the people involved, about how the mumbo jumbo was and is, indeed, worth it.

In the case of the Julian Street Inn, it’s only possible to see for one’s self, and to talk with some of the people involved, after submitting to a criminal background check. And even while policy prohibited me from photographing or interviewing the residents, I did chat quite a bit with the woman who runs the place.

She told me that the 90-day intensive program they offer at Julian Street is offered in other buildings around the city, too, and that the outcomes for people who are treated at Julian Street are significantly better.

Alexander’s myriad (and often counter-intuitive) decisions about the situating of material in space at Julian Street, so hotly contested with the client, resulted in a building of exceptional beauty and value.

A building that plays an active role in the healing of the people it shelters.

For a non-architects who want to apply Alexander’s theories in other fields, and to make living structures in other kinds of space, the path forward from reading his books and admiring his buildings has typically been limited to analogy.

But not today.

I believe that the two case studies I’m sharing with you today can and should be evaluated on the basis of the same criteria: on the basis of how decisions about the situatedness of elements in space contribute to human wellbeing. They both concern matters of life and death for people suffering from the effects of mental health crisis, and my contention in this article is that the difference in quality and value between what Chris Alexander fought so hard to build his way in San Jose, and what otherwise and more typically would have been built there, is precisely the difference between the information structures below.

Case #2 - Cognitive Shelter For People In Mental Health Crisis

Patient Safety Plan - Specimen 1

If you’ve never seen a patient safety plan, that’s probably a good thing. Although I am certain there are people you love and interact with every day who benefit from having easy access to a tool like this (and the professional help that usually accompanies it).

The rate of death by suicide for people in the United States has increased 28% over the last reporting period. Clearly there is a self-harm crisis in this country, and from what I’ve read online, use of a form like the ones under scrutiny here is considered to be a best practice when people are being released from the custody of a mental health care facility in the wake of an episode of self harm.

I’ve been looking at the Google Images results for searches on “patient safety plan.” Most of what comes back looks just like the template shown here. We’ll call this version Specimen 1.

It’s been said that 99% of the built environment around us was made without the participation of an architect. I suspect that 99% of the information structures people encounter today in the health care system are made the same way. Surely, Specimen 1 was made this way: to be something that looks like everything else on the computer, something that’s functional in a check-list kind of way, and “usable.”

This might be less of a problem, or maybe no problem at all, if information space was fundamentally different than physical space, and thereby exempt from the rules and forces that govern what works in real life, in the real world. After all, this is just some shit you fill in on a meaningless piece of paper after you have a freak-out, right?


My contention in what follows is that every kind of space is the same, fundamentally. Cyberspace. Information space. Document space. Physical space. It’s just space. And, consequently: that the factors behind the facts of comparatively higher performance in a Chris Alexander building are equally “in play” when we make information structures to support mental health using documents and conversations.

Here in the world of documents, as in the world of bricks and mortar, the question is identical: couldn’t we just make a box?

Some Comparisons Are More Equal Than Others

Comparison may be the thief of joy, but it provides one of the clearest paths to understanding. Especially, as I’ve learned in my work on the biography of Richard Saul Wurman, when the scale of the comparison is equal, or can be equalized. To my way of seeing, the two specimens of a patient safety plan I’ve provided here for your contemplation are indeed ripe for fruitful comparison, because they’re rendered at the same “scale.”

We can think about scale here in at least two ways. First, and simplest: in both instances, what we’re looking at is a single sheet of 8.5 x 11” paper.

The second way we can think about the scale of the comparison as being equal here is in terms of the context of use. Both versions of this single sheet of 8.5 x 11” paper are used in the same way: primarily in a mental health care delivery environment, with the participation of a caregiver who’s working with a patient.

And thus, with the benefit of this equalized scale for the comparison, it’s even clearer (to my way of seeing) that the 2nd specimen is better than the 1st specimen. It’s the only one of the two that’s been architected.

I will indulge in some preliminary analysis of the differences here in terms of information architecture, but really what excites me about the comparison is how we can use the same theory that explains Chris Alexander’s work on buildings to explain what’s going on with these two works of information architecture.

The difference has to do with space, and with choices about situating material in space.

Patient Safety Plan - Specimen 2

The primary difference in what’s happening here in the 2nd specimen, and what’s happening with most of the other versions of the safety plan, is spatial.

Architecture, like sculpture, is a spatial practice. Working in terms of architecture means that choices about situating things in space are presumed to be both meaningful and meaning-generating.

Since the earliest days of the field of information architecture, Richard Saul Wurman insisted (and still does) that choices about arranging information in space are decisions about architecture, and that these choices directly affect what things mean, and what’s possible to understand.

When we compare Specimen 1 with Specimen 2, one of the most striking points of divergence has to do with the choice each document creator made with regard to sequencing the information modules.

4 of the 6 steps delineated in Specimen 1 appear in Specimen 2. But unlike Specimen 1, the content modules in Specimen 2 are not numbered, and they’re not called steps. And only two of the modules in Specimen 2 are the same size. We could say that most of the things in Specimen 1 are “sames,” while with Specimen 2 there’s a lot of difference. The word Alexander would use is differentiation.

As an axiom, Alexander’s colleagues have found that living structures exhibit more differentiation and articulation when compared with assembled structures. Living structures, they say, tend to have a higher degree of complexity, and a more elegant (as physicists use the word) way of evolving themselves, through emergence.

In my observation of a social worker’s use of Specimen 2, the first thing that was said to the patient was that they could fill the form out at their own pace, following any sequence they’d like.

And to my way of seeing, this opening act of facilitation by the caregiver—this generous way of inviting the person in crisis to exercise agency in how they relate to the piece of paper—was predicted in and afforded by the document’s structural design. Richard Saul Wurman’s definition of architecture is “the thoughtful making of space and place.” The spaces and places in Specimen 2 thoughtfully anticipate what a good social worker would want to see emerge while guiding a patient through this process.

Although I have no direct experience with Specimen 1 in a real context of use, when I imagine the way both the caregiver and the patient might relate to each other through the structure of the piece of paper when that piece of paper is designed like Specimen 1, as compared with what I observed with Specimen 2, I think it’s not unfair to predict that Specimen 2 would afford more emergent information and behavior on both sides of the clincian’s desk, when compared with Specimen 1.

This Is Not About What It Looks Like

There are myriad reasons, from the point of view of design, to dislike both of these specimens of the patient safety plan. Happily, we can set most of those reasons aside for the duration of this discussion about architecture. While Specimen 2 does employ graphical devices like drop-shadows, bevels, and arrows, in the present discussion, these elements are taken into consideration only insofar as they’re employed as means of shaping space. For our purposes here, the precise line weights, type faces, arrow styles etc. of any design element are presumed to be fungible, and that the styling of these elements is of low or no consequence to the analysis.

What’s not fungible, and of high consequence to anything that calls itself architecture, are the geometric properties and spatial relationships among the elements that are employed to make the structure be what it is.

Order And Space

To me and Andrea Resmini’s way of thinking, these non-fungible, high-consequence variables around how elements get situated in space are encapsulated in the concept of topology. We agree with Wurman’s axiom that the creative organization of information creates new information, only in our preferred parlance, we’d say that changing the situatedness of things in space (topology) changes what they mean (ontology). And that the consequence of changes to ontology and topology are only really available to us as phenomena to study in the choreographies that people enact in the environment.

With Specimen 1, it’s clear that one is meant to start at the top, and then fill out each step in the order it appears in the vertical sequence on the page from top to bottom. One, two, three, etc. The eye-path starts at the start, and then follows the left margin all the way down. And when you’re done, you’re done.

It’s quite a contrast, when I imagine myself being asked to fill out Specimen 2, where the spatial relationships among the elements lend themselves to a cyclical, clockwise eye-path and information interaction. Although it’s not hard to imagine working the form counter-clockwise, too. The structure invites interpretation: the absence of step-numbering creates an inviting ambiguity.

Enough about Specimen 1’s comparative shortcomings. What makes Specimen 2 good?

It’s the claim I made at the outset: that it’s the same reasons that make a Chris Alexander building good.

In the three decades of work and writing that came after the publication of A Pattern Language, Alexander flipped the question of cause and effect when it comes to the difference between good and bad buildings, to ask and potentially answer far deeper questions regarding the nature of order itself. The order of anything: because everything is placed in space.

In a 4-volume magnum opus titled The Nature of Order, Alexander identifies 15 geometric/spatial properties that seem to show up and to characterize the ways a thing has been made when somebody makes something good.

Not unlike mindfulness and mediation, these 15 properties can’t be used to fix anything after the fact.

And by the same token, you can’t implement them using a checklist.

Nevertheless, Alexander’s research has shown that the 15 properties can be used to explain and predict the quality and value of the things others make, in spite of one’s having no particular expertise in the maker’s field or craft.

For example, Alexander found that he was able to develop an immensely valuable collection of antique carpets by evaluating and choosing them using the same 15 properties he’d developed in The Nature of Order to evaluate the quality of buildings.

As you might expect, the properties are usually interconnected with each other, and Alexander’s theory states that an artifact (in this case, we’ll swap-in a document for a building) can be said to be more profound, more beautiful, whole, and alive as a function of there being a denser and more elegant network of the properties among its parts and wholes.

In the same way that the properties never appear in isolation, according to Alexander, they always operate at multiple scales. Even so, with close observation it is often possible to point to a specific expression of one or more of the properties, as I’ve tried to do in the adjacent image.

I made a half-hearted attempt to apply the same analysis of the properties to Specimen 1, but that version of the patient safety plan was made in such a way that when I look for them, the properties are very weak. It’s a stretch to say that they’re present in any meaningful way, at all. To my way of seeing it, Specimen 1 was made in a manner that presumes content and space are neutral, and from that way of seeing, it’s not wrong to just define a container and then fill it in an orderly (how about sequential steps!) fashion.

4 of the 15 properties don’t lend themselves to being mapped onto a region of the document, in colored overlays, the way the others do. These 4 properties either cover both the container and the contained, or they don’t.

I think as a quartet, they provide an interesting macro-level lens through which the relative inadequacies of Specimen 1 as a whole come clearly into focus.

  • Non-Separateness: is the degree of connectedness an element has with all that is around it. A thing which has this quality feels completely at peace, because it is so deeply interconnected with its world.

  • Good Shape: creates positive space around it, is very strongly distinct from what surrounds it, and has a feeling of being closed and complete.

  • Inner Calm: is a special simplicity or economy developed over time in which all things unnecessary, or not supporting the whole, are removed.

  • Strong Centers: can only occur when other centers are intensifying it. A true center is defined not by its internal cohesion, but by its relation to context.

If your loved one had the option of using one or the other of these patient safety forms, would any of these four properties help explain why you’d prefer them to use Specimen 2 instead of Specimen 1? Or vice a versa?

So What? What Now?

Unlike the 15 properties, whose emergence is made less likely the more you try to manifest them, you can use information architecture thinking in your practice right now, in an explicit way, to improve the clarity and quality of your work.

I encourage you to ask yourself, explicitly, during your acts of making, how and why your decisions about situating information in space (topology) will change what things mean (ontology), and how those changes affect what’s possible for people to understand (choreography).

In an ideal information architecture development process, you’d then get to test a bunch of different potential decisions about ontology, topology and choreography with different kinds of people. I know of no better way to ensure the outcomes the work we do in so-called information space is meant to guarantee than to work at it in this way.

And then, when you complete an iteration of the thing you’re making, and want to predict the amount of wholeness, beauty and life in what you’re building, the 15 properties can help you understand the degree to which what you’re making will fit in with everything else in the environment.

P.S. If you’d like a baseline test of the 15 properties to then use as a means of calibrating your appreciation of them in your work, and in the work other people do, here’s a great way to get that done. Step 1: go sit in your favorite place. Step 2: look around. I’ve found that whenever I am in the presence of something beautiful, whole, and alive, these properties are schmeared all over the place.

P.P.S. If you’d like to lean more about the ideas of Christopher Alexander, and to talk about applying them in digital practice, be sure to join me on September 15 for Architecting Information, part of a live conversation series that’s taking place on a Sunday morning each month, starting next Sunday.