Can stuff be good?
And who deserves the blame if it is bad?
Re-Assembling Reality #33b, by Mike Brownnutt and David A. Palmer
In the previous essay, we established that technology is not neutral. It changes you, for both good and bad. In this essay we are going to play a blame game: when technology changes the world for the worse, whose fault is it?
Are you just using it wrong? The nature of design
There is an idea that if you use technology in a way that is detrimental, you are using it wrong. The previous essay raised an early flag that things may not be so simple: all technology is necessarily ambivalent. This ambivalence means that technology necessarily has detrimental effects. One advantage of binoculars, for example, is that they allow you to distance yourself from your subject; you don’t have to get up close to observe nature. One disadvantage, however is that they allow you to distance yourself from your subject; you don’t have to get up close to observe nature. This is not because you are using it wrong, or using it in a way that was not intended. It happens when you use it right, in exactly the way that was intended.
You can use binoculars to look at birds, or burn ants, or prop doors open. None of these uses is neutral, but all of them seem to rest with free choices made by the end user. The designer, or manufacturer, or seller has no say in things once the binoculars are in the hand of the end-user. This idea is worth discussing, because, if it is correct, it undermines the entire raison d’etre of design theory.
There are three things that may be noted about the nature of design:
— The design of an object constrains both what can be done with an object, and what cannot be done with it.
— The design of an object communicates to you how the object should be used.
— A designed object carries with it an implicit endorsement of the ideals embodied by its design.
Let us consider each of these in turn.
Firstly, the design of an object constrains both what can be done with an object, and what cannot be done with it. A door’s hinge constrains the door so that it can be opened by pushing and cannot be opened by pulling. A saloon-door hinge constrains a door’s use so that it can be opened both by pushing or pulling, but it cannot resist being opened by either pushing or pulling. Right-handed scissors cut well when held with your right hand, and badly when held with your left. A landmine with a hair-trigger cannot safely be used as a paperweight. This is not mere social convention of how a person chooses to use an object. It is the very nature and purpose of design that an object can do some particular things and cannot do others.
To the second point: the design of an object communicates to you how the object should be used.
One universal constant throughout all cultures is that people never read the instructions. It is for this reason that most designed objects do not come with instructions. Consider this: Where do you keep your driver’s license? Where do you keep your ID card? Or your travel card? Most likely you keep them in your wallet, along with your credit card, your store loyalty card, and numerous other pieces of plastic. No one in government told you to keep your driver’s license in your wallet. They just made it the same size as your credit card, and let you do whatever you wanted. You are, certainly, free to clip your driver’s license to your belt, or put in in your rucksack, or stuff it down the side of your shoe. But you don’t. If the government had wanted you to keep it in your shoes, they would have sewn the information onto a sock. But they didn’t. Without written instructions, designers go to great lengths to convey to you how to use something by the object’s form. Ideally, the message communicated implicitly by the design about how an object should be used aligns with the actual nature of how the object can be used. And if it does not, even instructions will not help.
Written instructions, as a rule, are much less persuasive than the message conveyed by the object’s form. By way of example, a door-handle permits you to pull a door, while a smooth plate means you are physically unable to do anything other than push it. It is a poor designer who puts a handle you can pull on a door which can only be opened by pushing. It is a forlorn hope that by adding a small sign saying “Pull,” the explicit written instruction will somehow override the actions implicitly communicated by the (bad) design.[1]
Finally, a designed object carries with it an implicit endorsement of the ideals embodied by its design. The mere existence of a pair of left-handed scissors tells you that at least one person (whoever designed and made the scissors) thinks that it is worth taking some time to make a left-handed person’s life a little easier. This is an implicit endorsement of the importance of left-handed people’s comfort. By contrast, if no one ever bothered to design or make or stock left-handed scissors, this implicitly sends the message that manufacturing efficiency, or standardisation, or keeping prices low, or keeping profit margins high is more important than left-handed people’s comfort.
Putting all of this together, all designed objects — whether well designed or poorly designed — are constrained to do and not do certain things, to communicate how they should be used, and to embody an implicit endorsement of how they are used. A plastic fork is designed to be used once and thrown away. By its flimsy nature, it may be necessary to throw it away after a single use, as the handle bends and the tines snap. Its very existence communicates to the world: throw-away culture is OK. It cannot be argued that disposable cutlery is neutral, or that anyone who harms the environment by its use must be “using it in the wrong way”. If you use a disposable fork in a profligate way, you are using it right!
The Story of Coke
Bad doors are annoying, but the ability to nudge someone’s action from pulling when they should push to pushing when they should push seems less than world-changing. Let us therefore instead consider the story of Coke.
Coke had a problem: they wanted to sell people more Coke. That in itself was not a problem, but everyone drank Coke already, and they drank it all the time. People would get a Coke, and their friends would get a Coke. They would drink it with lunch when they were out and they would drink it at home with dinner. If your success as a company is measured by growth of sales, total market saturation is a problem. Short of getting people to sit down and drink two servings of Coke with each meal, you simply cannot sell more Coke. And it is surely not easy to get people to drink two servings of Coke at every sitting. Right?
In 1955, Coke introduced the King Size bottle to the US market. At 10 oz (285 ml) it contained two servings of Coke, but the bottles were not resealable. Now when Coke drinkers opened their bottle, they had four options available to them:
1) they could drink half of it, and throw the other half away,
2) they could drink half of it, and carry the rest around to drink later when it was warm and flat,
3) they could ask a friend if they wanted to share their Coke, or
4) they could drink two servings of Coke.
Where the story goes from here depends on where you live. In many countries, people now drink two servings of Coke as standard. In others, they drink three. Some have five servings as standard. This is not because Coke provides a set of instructions about how much of their drink you should consume. Rather, they give you a 32 oz. cup, and let you decide what to do with it. You are free to choose to share your 32 oz. cup with four other people. But most people do not choose to exercise that freedom.
It is clear that soft-drinks containers — be they cans, cups, or bottles — are designed with particular purposes in mind. These purposes are not communicated in a set of instructions about how it should be used, but imprinted in the form of the container itself. And they are not neutral. One purpose of a soft-drinks container is to contain soft-drinks. You could drink Coke by cupping your hands under the soda fountain. This would lead to a very different Coke-drinking experience and, in many respects, the technology of bottles, cans, or cups offers advantages over the hold-it-in-your-hands technique. However, another purpose of modern soft-drinks containers is to get people to buy additional servings that they neither want nor need.
This is the nature of design. The person who drinks five servings of coke in a single sitting is not using their cup wrong. They are using it in exactly the way that is intended.
Diversity in non-neutrality
In these examples, the non-neutral nature of technology takes on a wide variety of kinds. In its weakest form, non-neutrality of technology might simply mean that the technology changes the situation. Drinking water from a cup and drinking water from your hands is different. I am not going to say that either of them are wrong, or even that one of them is better; they are just different.
Some kinds of technology might cause changes which are arguably good or bad, but not particularly bad in a moral sense. Drinking coke from a cup is much less sticky than drinking it from your hands. To say that the cup is an improvement is not to say that sticky hands are in any way morally problematic. It is a purely practical improvement.
But some technological changes do have a moral aspect. A technology which facilitates and implicitly endorses drinking five servings of a sugary drink is morally dubious. A technology which increases profits for the few, while stoking an obesity epidemic for the many is, to put it politely, morally problematic.
Are bad designs made by bad people?
The example of soft-drinks cans seems fairly simple: someone made the decision to put profits before people. The technology’s morally problematic aspects sprung from a person’s morally problematic priorities. The problem seems not to lie with technology, but with people. We might therefore hope that if we can ensure that people are good, they will create good technology. There is a certain truth to this framing, but it risks too-quickly backgrounding technology, and shifting our focus instead to people. In so doing, it risks making us oblivious to very real issues besetting technology.
By way of illustration, consider an architect who designed a house of the kind shown in Figure 1a. In this house, the life of the home was in the kitchen. A lot had to be done in the kitchen, so everything was done in the kitchen: preparing food, washing up, entertaining guests, looking after the children. You cannot get from the front door to the living room without going through the kitchen. A person working in the kitchen is in the centre of the action.
Time moves on, there is a little more money, a little more space, and the architect designs another house, of the kind shown in Figure 1b. With the greater space, you can keep the “workings” of the household hidden away from the “presentable” side of the house. The kitchen gets its own (little) room, around the back, with a door on it that you can close. Guests can get from the front door to the lounge without ever needing to go through, or even see, the kitchen. The hostess cooks alone, as she must, for the kitchen is far too small to fit a second person in. The guests sit with their backs to the kitchen door, watching TV. The kids come home and do their homework on the table in the lounge. Maybe a tutor is hired to help them with it, because Mum cannot be expected to keep running in and out of the kitchen, or to shout instructions through the kitchen door. A person in the kitchen is excluded from the action.
In a culture where it is predominantly women who rule the kitchen, the latter house excludes women from social and family interactions. As such, it is morally problematic. One could even say that it is sexist. [2]
Was the architect who designed this house morally bankrupt? Probably not. Was the designer intent on making women’s lives difficult, straining family ties, and undermining social interactions? Probably not. More likely they just never thought their design through. If they had spent their career being told that their creations were neutral — amoral and value free; neither good nor bad; not discriminatory, racist, sexist, or political — then they would have no cause to think very deeply about where to put the kitchen.
From this, it is clearly possible to meaningfully make statements such as “I am not saying that the designer is sexist; I am saying that the design is sexist.” Certainly, one may insist that the designer is not without blame: the designer may have been thoughtless. But that is not the same as the designer being sexist. The house is able to embody moral values which are distinct from the moral values of the designer.
How can a thing embody values?
We have so far said that a house can be sexist. We have also said that the value systems the house embodies — such as valuing presentability over social interaction — need not align with the value system of the architect. What, then does it mean for a thing to have values? Does it plot world domination in its dark heart? Would it benefit from talking to a priest, or confessing its sins?
To get a handle on this, let us consider one of the most basic technologies: written words. When you write a word like LOVE on a page, it is just ink and paper; a pattern of light and dark. The ink is not good or bad, the photons transmitting the information from the page to your eye are not kind or cruel. They are simply what they are. They could not be otherwise.
Nonetheless, we speak of “good words” and “bad words”. We have no qualms with the idea of naughty words or offensive words; kind words or encouraging words. Words have meaning. They carry significance and convey intent. The person who writes a word imprints in the form of the ink an intention, and the person who reads it responds to what they read. If the author is an effective communicator, the reader’s understanding will align with authorial intent. The author may effectively communicate an uplifting message, or may effectively communicate a destructive message. Rare is the author who would claim that words are neutral artefacts, conveying no meaning, and incapable of conveying meaning. We recognise that the words “Drink more Coke” convey a meaning, and are intended to convey a meaning. And we recognise that the meaning conveyed is not neutral. The shape of a Coke can conveys this message no less than the shape of the words.
The word, however, is not the author. A word can exist, disseminating its message to the world, long after the author is gone. It also takes on a life of its own. It can even convey meanings unintended by the author. While I worked in (German speaking) Austria, a friend in (English speaking) Australia sent a birthday present by post. Wanting to ensure that it did not get delayed at customs, they wrote on the box clearly with large letters: GIFT. It is unfortunate that Gift is the German word for poison, and the box was consequently delayed even longer in customs than usual. The fact that a word can convey a meaning other than the author intended does not permit the author to write whatever they want. Rather it places an onus on the author to go to reasonable lengths to consider how the word could be interpreted in a variety of reasonable contexts. The flexibility of meaning also does not permit a person to keep on repeating a problematic term simply because the original intended meaning would have been appropriate. Rather it puts the onus on members of a language community at each and every stage of language use to consider the implications of the words they use in that context.
The same holds true for technology. The technology embodies certain values. It communicates certain things. The creator of the technology has a responsibility to think ahead to a range of reasonable contexts in which the technology might be used. As a responsible human being, they should take reasonable responsibility for what they create, and not simply say “how could I have known?” At the same time, and as with language, each individual user of technology has a responsibility to engage critically with the technology they use. They must consider the implications of the technology they use in the context in which they use it. And they must consider and reconsider its use as the context and culture around the technology develops.
Pointing fingers
Having promised at the start of this essay to point fingers, I suppose we should make good on that promise. The technologist who is genuinely evil bears a responsibility. The technologist who did not think it through bears a responsibility. The technologist who is in a position to make it better bears responsibility. The end user who is corrupting their own humanity or the humanity of others bears a responsibility.
The astute reader will notice that I have placed moral responsibility for the problems and the solutions with pretty much everyone, but not with the technology. But I said before that technology is not morally neutral. I am not contradicting myself. To say that technology is morally problematic does not require the assertion that technology is morally culpable. Equally, on the flip side: to say that the technology is not morally culpable does not require the assertion that technology is morally problematic. Just because we don’t put cluster bombs on trial for murder does not mean that cluster bombs are morally neutral or morally unproblematic.
So let us return to the people, concerning whom I have pointed fingers in all directions. The responsibility that each party bears — and the action required of them in mitigating the problem in which they are involved — is not uniform. It is worth tempering my apparently egalitarian finger pointing with a few comments regarding power and agency.
Ideally, a person buying a house would be aware of the social hazard involved in modern house design, and would select a socially inclusive house. What, however, if there was a standard layout of houses in the area, all of which were of anti-social design? Or what if all the rich people, being aware of the problem, had bought up all the socially inclusive houses, leaving all the houses with an isolating design to be picked over by those who could not compete on price?
These are called “structural” problems — not referring to how buildings are constructed, but to how society is constructed. They are ubiquitous, technology plays an integral part in how they play out, and they radically redistribute and concentrate power.
Your personal decision to not have a smartphone has a hard time when your gas company requires you to submit a meter reading via their App. Your personal desire to not have a car has a hard time when a motorway separates you from the shops. Your personal decision to reduce your carbon footprint has a hard time when two thirds of your carbon footprint comes from things outside your control, like the concrete used to build your office. Your personal decision to have face-to face conversations with real people has a hard time when your high-street shops and banks close because of pressure from online business.
Technology mediates all of these things. The scientist or engineer who has the power to chose what concrete to develop or to use cannot therefore absolve themselves of responsibility, saying that an individual’s carbon footprint is down to the individual’s actions. Nor can they they fob off an issue as “simply an economic problem” or “simply a political problem.” The responsibility for these things is inextricably linked to the science.
[1] Such doors are called “Norman Doors” after Don Norman, who has pioneered ideas in human-centred design, and worked to make doors better. He writes about them and many other design ideas in The Design Of Everyday Things (New York: Basic Books, 2013).
[2] At first glance, one might imagine that this problem is solved by creating an egalitarian society where men and women share the kitchen equally. Alas, this will merely change the problem: from one in which the house harms the social interactions of women, to one in which the house is even-handed in its social harm.
