As to yourself, although you should live in such a way that you trust your own self with nothing which you could not entrust even to your own enemy, yet, since certain matters occur which convention keeps secret, you should share with a friend at least all your worries and reflections.
I doubt it was foremost in Seneca’s mind, but I see in this short passage some useful reflections for current debates about privacy and autonomy in an era of mass surveillance. Seneca says three important things in it. First, he suggests that the ideal life is one in which you have no real secrets (i.e. there is nothing that you do that could not be entrusted to your own enemy). Second, he acknowledges that this ideal is unrealistic and that convention requires some degree of secrecy. And third, he suggests that controlled disclosures of private facts to discrete others (in this case close friends) is part of the well-lived life.
All three of these ideas surface in Marjolien Lanzing’s paper ‘The Transparent Self’. In the paper, Lanzing takes issue with some of the claims underlying the quantified self (or self-tracking) movement. Some of you will be familiar with this movement; some of you may even utilise self-tracking technologies on a daily basis. The gist of Lanzing’s paper is that there is a tension within the quantified self worldview. Although many of its proponents think that self-tracking enables greater autonomy, the reality is that it serves to undermine autonomy. The reasoning can be summarised in the following (informal) manner:
Claim1: According to its proponents, self-tracking enhances one’s autonomy.
Claim2: Informational privacy is a necessary condition for autonomy.
Claim3: Use of self-tracking technologies undermines informational privacy.
Conclusion: Contrary to its proponents, use of self-tracking technologies undermines autonomy.
Now, to be clear, Lanzing isn’t quite as explicit about the reasoning in her paper, but I think this is a pretty fair reconstruction. I want to go through each step in the argument. I’m going to spread my analysis out over two posts. In this post, I’ll focus on the first claim: the alleged link between self-tracking and autonomy-enhancement.
1. What is autonomy anyway?
I outlined a simple pro-autonomy argument for self-tracking in a previous post. I want to be a bit more thoughtful and systematic in this post. To do that, I will start with a particular account of what it means to be autonomous and modify it in light of some concerns.
Anyone who is familiar with the literature, will know that ‘autonomy’ is a contested concept. It is supposed to refer the concept of self government but there are several different accounts of self-government means in practical terms. Picking one particular account out of the morass might seem a little arbitrary, but it’s exactly what I am going to do here. I am going to pick out Joseph Raz’s definition of ‘autonomy’ and use it as the basis for an argument in favour of self-tracking. I do so because I think his account is reasonably expansive, and appropriately vague as to what is meant by some of its key conditions. Here it is:
If a person is to be maker or author of his own life then he must have the mental abilities to form intentions of a sufficiently complex kind, and plan their execution. These include minimum rationality, the ability to comprehend the means required to realize his goals, the mental faculties necessary to plan actions, etc. For a person to enjoy an autonomous life he must actually use these faculties to choose what life to have. There must in other words be adequate options available for him to choose from. Finally, his choice must be free from coercion and manipulation by others, he must be independent.
There are three important conditions of autonomy embedded in this quoted passage. They are (a) the person must have the minimum rationality to plan actions that will allow them to achieve their goals; (b) they must have adequate options available to choose from; and c) they must be independent, which Raz takses to mean free from coercion and manipulation when making and implementing their choices.
You can understand these as minimal or threshold conditions for autonomy. In other words, you can say that once a minimal degree of rationality is achieved, once there is an adequate range of options available, and once there is independence from others (in some appropriately defined sense) then there is an autonomous life. That’s all there is to it. It is also possible, I believe, to understand these three conditions as defining a dimensional space of possible degrees of autonomy. In other words, to hold that the higher the degree of rationality and independence, and the more options available to you, the more autonomous you become. That said, if one embraces this dimensional view of autonomy, one must be sensitive to the possibility that once you go beyond a certain level of independence, rationality (and so on) you may start to experience diminishing marginal returns. Indeed, at extreme levels it is possible that there is an inverse relationship between an increase along those dimensions and autonomy.
An example of this inverse relationship might be the so-called ‘paradox of choice’. This is an effect that has been discussed by psychologists which suggests that if you have too much choice you become unable to make a decision. You become paralysed by excessive choice. Think about those times when you sit down to watch a movie on Netflix and can’t make up your mind because there is a seemingly endless list to choose from. Since the ability to execute a decision is (presumably) central to being the author of one’s life, excessive choice may consequently undermine autonomy. I return to this point later.
It is also worth thinking about the complexities that arise in relation to Raz’s third condition (the independence condition). This states that in order to live an autonomous life you must be free from coercion and manipulation by others. While this certainly seems like a plausible requirement, it should be noted that some degree of external coercion may be necessary in order to achieve one’s goals. This is the central paradox underlying liberal modes of government. Many prominent liberal authors argue that some coercion by the state is necessary if you are to be free from coercion and manipulation by your fellow citizens (the classic Hobbesian argument). There is undoubtedly something faintly paradoxical about this but it seems to be borne out by our collective history: a lingering threat of coercive interference by the state (or other institutional actor) seems to be integral to maintaining peace and stability. Another important point is that not all autonomy-undermining manipulations are going to come from other people. Sometimes we are our own worst enemies. We have cognitive and motivational biases that draw us away from doing what we most want to do. Weakness of the will and procrastination are obvious examples of this. It’s possible that by consenting to some degree of external manipulation we may be able to overcome these biases and allow ourselves to do what we most want to do.
We can reformulate the Razian account of autonomy to take account of these considerations:
Autonomy: If you have (a) the rationality to select the appropriate means to your desired ends; (b) an adequate range of options; and c) are free from most external and internal sources of interference with, and coercion and manipulation of, your will, then you are autonomous. Indeed, the more you have of each, the more autonomous you are (up to some point of diminishing returns).
I’ll have more to say about autonomy as the post develops, but that suffices for the introduction.
2. Does Self-Tracking Enhance Autonomy?
With Raz’s definition in tow, we can start to build a case for the autonomy-enhancing powers of self-tracking technologies. Let’s start by getting a clearer sense of what these technologies do. There are many examples out there, at least some of which will be familiar to readers. Smartphones and other wearable technologies now contain within them a variety of sensors. These sensors enable the devices to collect information about the user’s activities. The FitBit is an obvious example. It collects information about heart rate, number of steps taken, distance traversed, sleep movements, calories burned and so on. When this information is processed by appropriate software, it enables the user to build up a sophisticated profile of daily activity. Other wearables can do similar or indeed more impressive things. Since the number of such sensor-laden devices and the array of sensors with which they are equipped is growing, we can expect to have even more self-tracking power in the future.
What’s more, most self-tracking technologies are not purely about self-tracking. They typically incorporate some element of behaviour change into their design as well. The idea is that the information collected and processed by these devices can be used to motivate you to change your behaviour in various ways. This desire for behaviour change is very clear from the marketing on the FitBit webpage. The makers hope that by collecting more detailed information about your physical activity they will be able to prompt you to become fitter, sleep better and generally improve your well-being. Makers of such devices often achieve this through some degree of gamification, i.e. by using the data to create a game-like motivation system in which you are rewarded when you achieve certain goals (that are tracked and verified by the device).
Understood in these terms, there are two important ways in which self-tracking technology could be said to enhance autonomy:
Gaining insight: The devices allow you to learn more about yourself, i.e. what you actually do, what prevents you from doing what you want to do, and so on. This information can facilitate self-experimentation, which allows you to gain even greater insight into what works for you and what doesn’t. All of this supports your autonomy by enhancing your means-ends reasoning power (i.e. rationality), which was the first of Raz’s three conditions.
Motivational scaffolding/outsourcing: The devices provide an external motivational scaffold that allows you to overcome internal motivational biases. This facilitates autonomy by ensuring that you are able to pursue your goals with minimal interference from your internal biases. In other words, it enables greater independence, which was Raz’s third condition (as modified by my comments towards the end of the previous section).
(There are also ways in which these technologies could alert you as to the existence of additional options or choices. Thus, they may enhance autonomy along the third dimension too, but I won’t develop that idea in this post.)
The upshot of this is that we are able to make the following pro-autonomy argument in favour of self-tracking technologies:
- (1) If you have (a) an ability to select the means appropriate to achieving your goals; (b) an adequate range of options to choose from and c) you are free from most forms of internal and external interference with, and coercion and manipulation of, your will, then you are autonomous. Indeed, the more of (a) – c) you have, the greater your autonomy (up to some point of diminishing returns).
- (2) Self-tracking technologies enable you to enhance your ability to select the means appropriate to achieving your goals (by helping you to gain insight into yourself) and provide motivational scaffolding that minimises internal sources of interference with your will.
- (3) Therefore, self-tracking technologies enhance your autonomy.
Is this argument any good? There are some obvious criticisms. For starters, as noted above, the account of autonomy I have defended is far from being the only game in town, and my modifications of the basic Razian position are not uncontroversial. I think they are defensible, but there are undoubtedly flaws that need to be ironed out. That would create problems for premise (1). It’s also possible that self-tracking apps are not that good at helping us to gain insight into ourselves or in providing valuable motivational scaffolding. For example, there has recently been some criticism of Fitbits on the grounds that the data they collect is not that accurate. If this is true, then they may not enable much insight. Furthermore, I have some psychologist friends who argue that the behaviour change aspects of many self-tracking apps are not appropriately grounded in the science of behaviour change (though some apps are better than others). So they may not be great at providing motivational scaffolding either. This would cast premise (2) into doubt.
These criticisms are worth pursuing, but I set them to one side here. This is because Lanzing’s argument against self-tracking is largely willing to grant that the apps enable insight and help with motivational scaffolding. She simply argues that the devices undermine autonomy in other ways. We’ll address those in part two. Doing so will bring us back to Seneca’s thoughts about privacy, controlled disclosures and the well-lived life.