A Kantian Analysis of Technological Change

Photo credits: Financial Tribune. Are companies using their employees as merely means to ends?

Kantian business practices are well defended in the literature. Yet technological advances in the workplace continue to challenge the compatibility of Kant and modern business. In this essay, I argue a) that companies cannot treat their employees as merely means to an end, but that individual managers could, and b) that by introducing workplace surveillance, managers might coerce employees, using them as merely means to ends. Although some have argued that business practices agreed to in the contract are not mentally coercive, and so do not use employees as merely means, I argue that workplace surveillance may result in companies using their employees in this way. If so, this would be unacceptable for the Kantian. I first outline why a company is not the kind of being that can treat its employees as only a means. This will help explain how individual managers could treat their employees as such by coercing or deceiving them. Finally, I introduce the issue of workplace surveillance, arguing that it often meets the criteria for coercion. Companies that introduce workplace surveillance might consequently treat their employees as merely means to ends and, even worse, force their employees to deceive customers. This is why I argue that the Kantian manager should take great care when implementing workplace surveillance.

Why the Kantian law does not apply to businesses

It would make no sense to argue whether companies do or do not treat people as means to ends. This is because, for Kant, only moral agents can be judged against the standards set by the categorical imperative (CI). As Altman (2007) has argued, businesses, being abstract and non-rational entities, cannot themselves act in a way that is morally right or wrong and are not moral agents. Businesses are an aggregate of many individuals’ actions. To consider a company to be a moral agent because it is constituted by a collection of moral individuals would commit the fallacy of composition (ibid.:262). This is because the company does not make decisions – it is not rational. And, for Kant, exercising rationality is synonymous with being a moral agent.

Kant IV
Photo credits: Philo-Sophie’s World. Immanuel Kant, the 18th Century Philosopher.

Rationality, in Kant’s sense, is intentional. It is concerned with an agent setting and pursuing their own ‘ends’ and acting to fulfil those aims from good motivations. These ends are premised upon the concept of our self-legislation – our adherence to moral rules, or ‘maxims’, that we choose to follow in recognition of the fact that they “could serve everyone (as universal) and could bind everyone (as practically necessary)” (ibid.:256). Companies are not beings that can set their own ends, though. Indeed, as Altman points out, if a business could be thought of as rational in this way, it would be a moral agent, and would thus need to be afforded the respect it is owed in “accordance with the dignity that all rational beings have” – an absurd conclusion (ibid.:258). Employees, shareholders and other stakeholders would be using the artificial person of the business itself as merely a means to profit generation. Thus, we can reject the claim that companies can treat their employees as merely means to ends. Nonetheless, although the business cannot treat its employees as mere means, it does not follow that those individuals that constitute it cannot use one another in this way.

What is it to treat someone as a mere means to an end?

Kant’s second formulation of the CI recognises that we each think of ourselves “as a rational creature who is entitled to dignity and respect” and Kant extrapolates from this premise, for reasons of consistency, to hold that this should also apply to other rational people (Bowie, 2000:192). Thus, his moral theory is fundamentally egalitarian and requires us to treat all beings that are similarly rational to us as if they are worthy of our respect in line with the CI. In another formulation, the ‘Kingdom of Ends’ formulation, Kant states that “one should act as if… one was subject and sovereign at the same time”, which illustrates what mutual recognition and respect entails (ibid.). Central to this formulation is that one’s maxims could be both universalisable (i.e. held by all without contradiction) and would be accepted by all rational agents like us in a hypothetical social union. The business can be thought of as one such union. Adopting this outlook, the ways in which the manager can interact with their employees are constrained, if they are to act in line with the CI and avoid treating their employees purely as means.

Sarah O'Connor
Photo credits. Financial Times. Sarah O’Connor, writer for The Financial Times, put herself under scrutiny from her superiors for a week, in an effort to investigate the effects workplace surveillance could have on employees. Watch it here.

Immediately, we should note that “no one is used as merely a means in a voluntary economic exchange where both parties benefit” (Bowie, 1999:7). The CI is compatible with business. What the CI does not allow for is the unconstrained usage of autonomous individuals as if they were mere tools, through means of coercion or deception, or otherwise diminishing their rationality (Korsgaard, 1996:140). Thus, if I lied about my business capabilities to win a contract, through my deception I would be using you merely as a tool and not respecting your rationality. Further, I could not will that lying be universalised in the Kingdom of Ends without contradicting my own rationality. If everyone lied like I did, then people would not be able to make informed, rational decisions – I would have undermined their dignity and rationality as well as my own. This might pose a problem for business practices, though. After all, is it not the job of a manager to coerce his employees into working harder and more efficiently?

Drawing on the work of Arnold and Bowie (2003), we can distinguish two kinds of coercion in a Kantian sense. Firstly, there is physical coercion, which we can grant is always wrong in business situations, as Sollars and Englander do (2007:122). Secondly, we can recognise psychological coercion, though there is less agreement on what counts as coercion in this sense. For Arnold and Bowie, to have psychologically coerced an employee, three criteria must be met:

(1) the coercer must have a desire about the will of the victim;

(2) the coercer must have a desire to compel the victim to act in a way that makes the coercer’s first desire efficacious; and

(3) the coercer must be successful in getting the victim to conform.

(2003:229, as quoted in Sollars & Englander, 2007:122)

Taking the example of employees being forced to work overtime to avoid being fired, Sollars and Englander reject that in such cases psychological coercion has taken place. This is because they believe that in businesses where working overtime is a routine part of work, the threat of firing is merely “a statement of the terms of employment” that does not constitute a legitimate coercive threat (ibid.:123). The employee entered into their employment contract knowing that they may be required to work overtime, and so cannot claim to have been deceived as they knew that this was a possibility. As such, it would not meet Arnold and Bowie’s criteria. The coercer “need not have a desire to compel a worker” to work overtime by changing their will, but simply a desire for the worker to carry out their contract, which the worker already agreed to by signing it (ibid.). Consequently, such manipulation does not amount to treating someone as a means to an end.

Is Workplace Surveillance Coercive? 

Accepting Sollars’ and Englander’s defence of threatened firing as non-coercive, I turn now to explore ‘workplace surveillance’ and whether it might meet the criteria for coercion. Workplace surveillance can be defined as the “sustained monitoring” of an employee by their employer whilst they are working (Macnish, 2017:175). This usually involves tracking various aspects of an employee’s workday – for instance, logging their keystrokes on a company computer, recording which websites they visit, who they call and for how long, or tracking their location via GPS in a company car (ibid.:177). Clearly, this will capture everything an employee does with company equipment, totally erasing their privacy. Importantly, many employees are not aware as to the extent to which they are monitored, as “American companies generally aren’t required by law to disclose how they monitor employees using company-issued devices” (Solon, 2017). Are such practices treating employees merely as means to an end?

I argue that some employees are mentally coerced by workplace surveillance, according to Arnold and Bowie’s criteria for coercion. Although not necessarily coercive, such practices have a contingent tendency to produce coercive behaviours because of its highly intrusive nature. Trivially, it should first be noted that if the employer does not disclose their surveillance, as Solon claims is the case in America, above, then they deceptively infringe upon their employees’ rights to privacy, which would straightforwardly mean treating someone as only a means to an end through that deception. So, many instances of workplace surveillance are straightforwardly morally impermissible for Kant. Assuming that employees know that they are being surveilled does not solve the problem, though. As the following example shows, great care must thus be taken by the Kantian manager when implementing workplace surveillance.

Fitness Trackers
Photo credits: Disruption. Fitness trackers are one of the ways in which companies could monitor employee performance in and outside of work. 

Consider the call centre worker who is unsatisfied with her job. A customer asks if she enjoys working for the company. In such a case, thanks to the chilling effects of surveillance, she might not answer truthfully. She would know that her employer could be surveilling her calls and so might be expected to answer that she “enjoys working here”. Despite the fact that the intention of the surveillance might have been legitimate – perhaps to ‘chill’ only inappropriate behaviours like customer abuse or increase efficiency (Macnish, 2017:180) – the desire on a manager’s part to alter an employee’s behaviour (reducing abuse or increasing efficiency) results in them self-censoring and deceiving the customer. Thus, the company’s surveillance meets the three criteria for coercion outlined by Arnold and Bowie – even with the clarifications made by Sollars and Englander – by ensuring that the employee is coerced to act as she otherwise would not have. The manager both has a desire to change her will (1), and their desire is successful in bringing about that change and ensuring that she conforms (2), (3).

Not only this, but it also coerces the employee to treat the customer as merely a means to an end by ensuring that she deceives him – undermining his rationality, and so reflexively hers, too. Kant, as we recall, would not allow such behaviour, as this would not be universalisable in the Kingdom of Ends. The employee could not wish that all people lied like her. Thus, all parties’ rationalities are undermined, and so workplace surveillance of this sort could not be accepted by the Kantian on account of the coercion and deception that the coercion brings about.


I have argued two points. Firstly, that companies – as non-rational abstractions – cannot possibly treat their employees as merely means to an end. And, secondly, that individual managers could treat employees in this way if they introduced workplace surveillance of a certain kind. As such, although the benefits of workplace surveillance (such as limiting abuse or encouraging productivity) may be economically valuable, for the Kantian it would be morally impermissible for managers to implement it in cases where it coerced employees to change their wills – particularly if it inadvertently coerced them to deceive customers. Clearly, I have only shown that such practices are impermissible in one contingent application of the technology; workplace surveillance may not necessarily be coercive if carefully implemented. Even so, I have shown that there is at least one plausible way for managers to treat their employees as no more than a means to an end, even with good intentions, and so care must be taken by the Kantian pursuing such policies until we can better ascertain what surveillance is morally permissible.


By Charlie Smith


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s