There hasbeen a wide range of normative debate concerning nudging, with many arguments focussingon whether nudging infringes on people’s autonomy. This essay will therefore focus specificallyon answering the question of whether we should nudge or not, with regards tothe concern over autonomy. Autonomy can be explained as self-government and howmuch control individuals have over their own decisions. Critics of nudge haveargued that nudging people takes away the control they would normally have tomake decisions.
This essay will firstlydefine what is meant by a ‘nudge’. It will then layout evidence which suggestsin some cases nudging does not undermine autonomy. After this, it will beexplained that our conception that individuals are normally fully autonomousmay be misplaced, bringing into question the high value assigned to autonomy inthis debate. From this, it will beconcluded that we should nudge. In 2008,Thaler and Sunstein brought nudging to the centre of the world’s attention. Theyargued that flaws in human decision making, including factors such asindividuals being too optimistic or overconfident, means that mistakes are madewhen making decisions.
Due to this, they stated that people should be ‘nudged’in the right direction. In their view a nudge can be defined as a particularaspect of choice architecture that steers individuals in the direction ofmaking a decision, but without forcing them to make this decision (Thaler &Sunstein, 2008). This means that nudges do not include financial incentives anddo not forbid any options that are present (Hausman & Welch, 2010). Thaler andSunstein tied nudging to the concept of ‘Libertarian Paternalism’. Toillustrate what this means, consider an example where food is rearranged in aschool cafeteria so that fruit was displayed first and dessert last, the intentionbeing to increase the number of children which pick fruit rather than dessert.This is a nudge because it has the libertarian aspect of not restricting anychoices; all foods are still available. Furthermore, the arrangement is alsopaternalistic because the aim of this was to modify children’s choices so thatthey would make a choice that was better for their health (Thaler , 2008).
Hence, Thaler andSunstein state that nudging doesn’t infringe one’s autonomy, as all the choicesare still available for the individual to make. Despiteefforts to show that nudging doesmaintain freedom of choice, a wide range of scholars’ object. Schubert (2005,p.9) states that autonomy is the key ethical concept that is “at stake” whennudging.
This is vital because having control over one’s actions and being ableto live freely is important in Western culture, and so, if anything was tothreaten this it would be a problem. Tracedback to Aristotelian times, it was stressed that it is important thatindividuals have autonomy for them to live a good life as part of the polis or political community (Hacker,2016). Furthermore,critics argue that nudges result in people not having a “reason that they can underwrite”(Yeung,2011, p. 135). In the cafeteria example,if someone does pick a piece of fruit when they were originally going to pickthe dessert, they are influenced due to this nudge and so cannot provide a motivationalreason as to why they picked the fruit instead. They were unconsciously nudgedto do so.
It can be argued that this nudge has undermined the autonomy of theindividuals; it has stopped people from having their own reasons for decidingto do things. From the above it can be stated, people who are being nudged arenot being rationally persuaded to make certain decisions. Due to this,Wilkinson (2012) argued that nudging is manipulative as it takes away some ofthe control that individuals normally possess when they make decisions. Furthermore,there is evidence to suggest that that nudges can in fact increase the autonomyof individuals. Barton and Grüne-Yanoff (2015) argues this by describing thenudge that pushes people to buy less cigarettes by disclosing information aboutthe dangers of tobacco on cigarette packets. They argue that this nudge could in factincrease the autonomy of the person making the decision; the individual willknow more about the risks of their potential purchase and so will be able toincorporate their knowledge of these risks into their decision. Thisinformation would mean they can make a better-informed decision, henceincreasing the control they have over the decision they make (Barton andGrüne-Yanoff ,2015).
To supportthis, Hacker gives another example. Heexplains a study done by Deb and Vargas. In this study, labelling food productssaliently with the number of calories they have, resulted in a reduction of thebody mass index of obese and overweight eaters (apart from obese women),(Hacker, 2016). Like the previous study, the nudge of providing salient informationon the food products does nothing to reduce the amount of control people havewhen deciding which food product to buy.
Like the example above, this nudge isproviding information about the products and so can lead to a better-informeddecision, meaning that the individuals are more autonomous. The individualswould spend time considering which option to choose, because they have anincreased amount of relevant information, and so their choices are moremeaningful (Mills, 2015). Furthermore, both nudges are presenting the individualswith information and allowing the decision makers to see aspects that would becovered up or invisible normally (Sunstein, 2015), hence can avoid the criticismthat autonomy is restricted.
It can beargued that although these informational nudges(providing more information about the product) may lead to an increase ratherthan decrease in autonomy, this is only one type of nudge. Not all nudges aimto work by merely displaying information. So, even if informational nudges donot infringe on autonomy, many other nudges do. This can be demonstrated by an example of a nudgeaimed at increasing charitable giving. This nudge aimed at increasing the numberof donations by the participants by manipulating whether they were presentedwith an ‘identifiable victim’ named Rokia, presented with only statistics, or acombination of Rokia and the statistics. It was found that there was a larger numberof donations elicited from Rokia alone (Small, Loewenstein & Slovic,2007). Itcan be argued that this nudge was manipulative; the choice architects influencedthe individuals’ decision, as they knew that presenting only the victim would playwith people’s emotions; causing them to donate more. Thus, the nudge didinfringe on the individuals’ autonomy.
Of course,from this example it is clear that there are many similar nudges which couldinfringe on autonomy. Should it follow from this that we ought to only allowcertain types of nudges? It seems thatthis would simply complicate matters more, it is not always clear-cut as towhen a nudge will decrease autonomy. Felsen and Reiner (2015) highlight that anudge that increases autonomy for one person, may in fact decrease autonomy foranother. They consider the cafeteria example, and state that being nudged topick the fruit could have different outcomes for people. The first person’sautonomy could be increased because they desire to be healthy and so pickingthe apple is the decision they would want to choose anyway; it is in line withtheir high-order volition to be healthy. However, the second person who picksthe fruit may have a decrease in autonomy if they desire to be happy.
Thebrownies may evoke pleasant memories of being with their grandfather, andreminiscing about these times may promote their well-being (Felsen and Reiner,2015). Thus, this person would have picked the brownie if the nudge hadn’t beenin place. This highlights that it would be extremely difficult to limit nudges;choice architects would have to possess knowledge about people’s high ordervolitions, which is impossible. However, thedifficulty of this task doesn’t mean that we therefore should nudge. Althoughit would practically be hard to limit certain nudges, practicality and normativityare different things. Nevertheless, itseems that it can be shown that the value we place on autonomy is too high. Ifthis can be done, then it can be stated that there isn’t as much of an ethicaldilemma as was thought of before. The definition of autonomy presupposes thatthe agent possesses a level of self-knowledge that cannot be found in the world(Schubert, 2015).
If we shouldn’t nudge because it imposes on our autonomy, itshould be the case that in every other instance when individuals are not beingnudged, that they are autonomous. These individuals should therefore have fullcontrol of their actions and decisions. However, there is evidence to suggest we don’tpossess the level of autonomy that is so commonly presupposed. Felsen andReiner (2015) argue that evidence from neuroscientific research can show thatour perception of the amount of autonomy the human brain has when makingdecisions is wrong. Although Berker(2009) and Kamm (2009) argue that neuroscience cannot contribute anything tothis question, it may demonstrate that the common assumptions we hold about thecapacities of our brain may be mistaken. Felsen and Reiner examined the neural conditionsof the brain that are needed to have full autonomy and found that neuroscientificevidence about everyday decisions being made is inconsistent “with the ideathat they are free from undue external influence” (Felsen & Reiner, 2015,p.476). Further, Schüür and Haggard (2011) found that external influences ondecision making are a norm rather than an exception.
From this, it can beargued that we have a common conception that our everyday decisions are moreautonomous than they actually are. This has heavy implications for objecting tonudging due to autonomy. If we don’t have the level of autonomy that we areassumed to have even in normal cases, then we can’t object to the governmentnudging people based on this idea. Fromthis, it can be stated that the value placed on autonomy by certain scholars istoo high (Felsen and Reiner, 2015). This allseems plausible; our common conception of autonomy is in fact not correct.
Still, a critic of nudging could state that in cases where people are nudged; thewill of one agent is “imposed” on another (Hausman & Welch, 2010, p.133). Whereas,in cases where there isn’t a nudge, although the individual may be influencedby external factors, no one’s will is being imposed on another. So, we should distinguish between choicesindividuals can make that are distorted by the nudge, and cases without anudge, where, at least the person’s choice isn’t distorted by any choicearchitecture that is intentionally designed (Hausman & Welch, 2010). Therefore,nudging is still objectionable.
However, howcan it be said with certainty that in cases where people’s choices areinfluenced, although it is not due to nudge, that there definitely hasn’t beenan influence of one agent’s will on another? Furthermore, from nudging it cannot be statedthat one agent’s will always influences another. To be able to say this, justas was stated earlier on in the essay, we would have to have knowledge of everyperson’s volitions when they make a particular decision. Someone’s will would only be imposed onanother if the action they perform goes against their volition; this isimpossible to know. It followsfrom the above that if nudging is not as morally impermissible as it wasthought to be. The concept of autonomy has been central in debating whether itis right for governments to nudge individuals, but such heavy emphasis need notbe placed on this.
This is because it is debateable whether nudging alwaysundermines an individual’s control and it has been stated that we don’t exhibitas much autonomy when making decisions that we thought we do, so less valueshould be placed on whether nudging undermines our autonomy. It has beenstated that the criticism from autonomy is subject to debate on many levels.Firstly: nudges do not always decrease, but sometimes increase an individual’s’control’ over a decision. Secondly: we have less autonomy than we thought. Buildingon the idea that autonomy is hard to define and a controversial concept, shouldwe really stop nudging people to improve their welfare because of this? Fromthe above reasons, it seems difficult to argue that we should not nudge basedon the criticism that individual autonomy may be undermined.
This can be emphasized especially with nudgesthat have an obvious benefit to the welfare of society. Nudges suchas those to do with public health policies have strong benefits. One example isthe nudge concerning organ donation. Instead of individuals being required to’opt-in’ to whether they want their organs to be donated after their death,individuals are required to ‘opt-out’ and so it will be assumed that individualsgive their consent.
In one study, it was found that the number of organ donorsdoubled as a result of this (Johnson & Goldstein,2003). The critic of nudges may again state that thisis controlling an aspect of an individual’s life which they should in factcontrol. However, the collective benefit is greatlyenhanced by this nudge; this is in terms of the number of organs that areavailable for donation which the same people might need one day (Raihani, 2013). Furthermore,Yeung explains that drivers slowing down when they are driving around a bendbecause they see the illusion created by painted red lines, may undermine theirpreference for driving at a higher speed. However, this illusion will bebeneficial because ultimately it will stop accidents from happening (Yeung,2011). Hence, it can be argued that thecomplications that arise when discussing the objection from autonomy are notworth it, given the obvious benefits that can result from nudging. It isimportant to go back to why autonomy is important in society.
The Aristotelian reasonwas that autonomy aligns with being able to lead a good life in the Polis. Canit not be argued that certain nudges such as the opt-out strategy for organdonation also contribute to the ability for individuals to live a good life?(Hacker,2016). Additionally, the ‘SaveMore Tomorrow’ programme aimed at increasing employees’ saving rates byautomatically increasing the percentage of their wage devoted to saving(Benartzi & Thaler, 2004). Again, itcan be argued that this increases people’s ability to live the ‘good life’. To conclude,this essay aims to show that the objection from autonomy is not strong enoughto give up nudging.
This objection is complicated and subject to a large amountof debate; it can be argued that many nudges do not infringe on autonomy.Moreover, the value we place on this is toohigh. This objection doesn’t seemworth stopping nudges, especially when they benefit the collective.
Thus, thisobjection needs to be thought about a lot more before it demonstrates anyconsiderable force against the argument to nudge.