COMMUNICATION AS COMMODITY: SHOULD THE MEDIA BE ON THE

MARKET?

Rutger Claassen
Published in: Journal of Applied Philosophy 28(1)(2011): 65-79

This is a pre-print version, identical to the final version except for corrections at the
proofreading stage. For referencing purposes, please use the published version available
at the website of the publisher.

Abstract: Should media communication be left to the market, or rather (partly) removed from the market?
This question is discussed by reconstructing an often-found “standard argument” in the literature on the
subject. This standard argument states that some form of market-independent media provision is required
since markets will fail to deliver a specific kind of high quality content conducive to the democratic
process. This paper argues that the standard argument is defective in several respects. By doing so, it
reevaluates the way we think about the contribution of the media towards democracy and the role that the
market is to play in this respect. First, the paper argues that the standard argument's normative premise
should not be couched in a welfarist theory but in terms of the capabilities that the media should strive to
realize. Second, it sets the normative expectations of the media’s contribution towards the public sphere
and democracy at too high a level. Third, the standard argument’s diagnosis of the market’s failure
incorrectly assumes that the market can never generate the demand for high quality content. An
alternative, more circumscribed claim about the market’s failure is presented, resting on two more
contingent types of demand failure.

Introduction

Many have held that media communication simply is a commodity, governed by

familiar laws of the marketplace. As a former chairman of the US Federal
Communications Comission put it: ‘Television is just another appliance… a toaster with
pictures.’1 The main question of this paper is whether we should agree with this
diagnosis. Should media communication be left to the market? Or do we have good
reasons to remove media communication – completely or partly – from the market, and
to provide it through alternative allocation mechanisms? In terms introduced by Michael
Walzer, should media communication be a ‘blocked exchange’?2 To answer this
question affirmatively would then be to establish one ‘moral limit to the market’, or

1
limit to ‘commodification’.3 The present paper is offered in the spirit of an exercise to
determine whether the media is an example of such a limit.
The practical urgency of this question is undeniable. Since their inception, the
mass media have played a crucial role in setting political agendas, shaping social
debates and informing the general public. During the 20th century, most democratic
societies have tried to regulate the media, so that they would perform these functions “in
the public interest.” For example, in many European countries commercial TV and radio
was not allowed until the 1980s and public broadcasters had a monopoly position.
Nowadays, the positions are reversed. After the break-up of these public monopolies the
audience share of commercial channels relative to public channels rapidly increased.
Similarly, in the US commercial channels were regulated to provide content of public
interest, but during the last decades these regulations were relaxed. As a consequence,
public broadcasting and public interest regulation now are on the defensive. Why spend
taxpayers’ money on services that are already available on the market? The central point
of sceptics has been that public intervention, bypassing the market criterion of satisfying
actual consumer demand is inherently paternalist or elitist. For example, Richard Posner
polemically remarked that people simply ‘want to be entertained’ and ‘want to be
confirmed in their beliefs’, so that ‘[b]eing profit-driven, the media respond to the actual
demands of their audience rather than to the idealized “thirst for knowledge” demand
posited by public intellectuals and deans of journalism schools.’4
Most political philosophers and social theorists writing about the normative
justification of media markets have generally been critical of the idea that media
communication would be a simple commodity. They have argued in favour of limits to
media markets, often along remarkably similar lines - for convenience sake I will refer
to their argument as “the standard argument”.5 The first, normative premise of the
standard argument is that consumer preferences should not determine what substantive
media content is available in a society. Instead a preference-independent normative
theory has to be adopted, which requires the media to deliver a specific kind of high
quality content, roughly equal to serious journalism conducive to the well-functioning of
the democratic process. The second, empirical premise of the standard argument is that
market-based media will fail to deliver the normatively required kind of media content.
The conclusion is that some form of market-independent media provision is required.
This paper provides a close evaluation of the standard argument. My aim will be
to show that it is defective in several respects. With respect to the first premise, I accept

2
the central importance of the media for democracy, but argue that this should not be
couched in a welfarist theory (the media as ‘merit good’). Instead I propose to rely on a
capability theory identifying those capabilities that the media should strive to realize
(section 1). Through a close examination of Habermas’ endorsement of the role of the
media in the public sphere, I will argue that, furthermore, the standard argument sets the
normative expectations of the media’s contribution to democracy at too high a level.
The media is best conceived, not as part of the public sphere itself, but as only having a
supportive role toward that sphere (section 2). With respect to the second premise, I will
argue that the failure of the market is not as straightforward as is usually thought. The
standard argument claims that we cannot rely on the market because it merely satisfies
existing preferences while the normatively desirable content forms new preferences
rather than satisfying existing ones. I will argue that this claim suffers from a confusion
of two different kinds of preferences (section 3). Alternatively I offer a more
circumscribed claim about the market’s failure to deliver the normatively desirable
content, resting on two contingent types of ‘demand failure’: in case of an emergency
situation where democratic content is threatened with extinction, and in case of
acollective action problem with respect to media consumption. At both levels, then,
there is a case for non-market media, although less sweeping, narrower and more
conditional than the standard argument acknowledges (section 4).

1. A Normative Theory for the Media

The normative premise of the standard argument is put in terms of the requirement to
provide specific kinds of media content. This is often elaborated in terms of what
economists call “merit goods.”6 The hallmark of a merit good is that it brings beneficial
effects (‘positive external effects’) over and above the value that is generated by the
transaction between a good’s producer and its consumer. The most prominent example
in the media context is content that keeps the public informed about social and political
affairs, enables it to form its own opinions on these affairs and to participate in
discussions about them (hereafter refered to as “democratic content”). One can however
also think of other types of merit good. For example, some argue that the consumption
of cultural programs has beneficial effects in civilizing people and increasing empathy
between them (sometimes all these merit goods are taken together under the heading of
“edifying content”). Finally, at the other end of the spectrum, there are also “demerit

3
goods”: goods with negative external effects, such as media products with racist, sexist
or violent content.
The normative claim cannot simply be that the media should – maximally or
sufficiently – deliver merit goods and refrain from delivering demerit goods. For
demerit goods, it remains to be seen whether or not the negative externality should be
borne by media producers and consumers or by the third party being benefited or
harmed by the externality.7 For example, offensive speech is often constitutionally
protected through the freedom of speech, even though it provides clear harm to the
offended person. Similarly, positive effects do not automatically qualify for
internalization. For example, a country as a whole may prosper economically if part of
the population has seen a winning match of the national soccer team and this causes a
consumption boom; but this does not oblige the media to broadcast winning soccer
games (and refrain from showing lost games). Any normative claim made on behalf of
(de)merit goods needs additional argumentation to establish a normative requirement in
its favor. This requires a general theoretical framework and its application to specific
potential merit goods.
With respect to the general framework, I propose not to rely on a welfarist
framework but instead interpret the normative claim in terms of capabilities. Thus
understood the normative claim of the standard argument will read: the media should
promote people’s capability to acquire democratic content (I will restrict my attention
to democratic content, presupposing other merit goods do not qualify8). Apart from
general reasons to prefer capability theories over welfarist ones 9, there are perspicuouss
reason to do so in the media context. The production of merit goods by media
organizations cannot guarantee their consumption by citizens, nor should it do so. If the
media broadcast democratic content, the realization of the positive welfare effect
depends on people’s free choice to actually watch this content. The distinction
fundamental to capability theories between ‘capability’ and ‘functioning’ is well-suited
to bring out this crucial difference between opportunities to consume and actual
consumption, the normative importance of promoting capabilities, not functionings, so
that people have the freedom to decide whether or not to function in the relevant way.
People cannot be coerced to actually consume democratic media content. Also, a
capability formulation of the normative premise fits better with the awareness in
communication studies that the reception of a message by an audience is not

4
unproblematic or self-evident. The actual consumption (functioning) may not be
identical to what the producer (sender) of the message intended.10
A cautionary note is in place. That it is objectionable to legally coerce people to
watch certain media content doesn’t mean that a normative conclusion to the extent that
people should actually consume democratic content is foreclosed by the adoption of a
capability framework. A capability theory is not committed to the argument that
capabilities should always be endorsed instead of actual functionings. Rather, the
capability framework is valuable for giving us the distinction between both (thus adding
analytical clarity compared to a welfarist framework). Even though this distinction is
often introduced to support normative arguments that capabilities should promoted, in
principle the distinction leaves open the question whether in a particular context
capabilities or rather functionings should be promoted. As we will see, this is important
for the justification of non-market media, which will partly rely on promoting
capabilities, partly on promoting functionings (see section 4).
This reformulation in terms of capabilities has not yet given us a reason to
endorse the normative premise. To do that, we need a normative criterion that
distinguishes normatively valid from invalid claims. If a capability is to have normative
force, we need to establish that it somehow is a ‘basic capability’; something that
deserves promotion. As a criterion to distinguish basic – or as I will say, “morally
required” – from non basic – or merely “morally permissible” – capabilities, I will use
the capacity for agency. Only those capabilities which are necessary to make people into
autonomous agents deserve to be listed in a moral theory that requires these capabilities
to be realized.11 Thus, the capability to acquire democratic media content is to be
classified as belonging to the category of morally required capabilities if it is somehow a
necessary condition for individuals to develop their capacity for agency.12
The next step is to apply this general framework to the capability to acquire
democratic media content: is this a morally required capability? A positive answer to
that question can be defended in either of two different ways. One form of justification
is to argue that to have this capability is a necessary condition for anyone to be a person.
Following this line of thought opportunities for benefiting from media content that
enhance the quality of one’s democratic participation are valued intrinsically. Edwin
Baker relies on this type of justification where he suggests that people’s collective
preference for non-market media can be justified by the fact that “many people would
like to be reflective, more self-reliant, more politically energized, more responsive to the

5
needs of others, and more interested in being informed.”13 This is a rather strong claim.
It may be the case that for some people these opportunities are crucial to become the
person they would like to be. For many others, however, participation in the public
sphere does not play an important role in their self-conception, nor is there a convincing
argument that it should do so. For example, what about people spending their life in
obsessive engagement with art, sports or science, without any interest in social and
political issues? Would they be agents to a lesser extent? That seems hard to defend.
It seems more promising to turn to an instrumental type of justification. People
need to have the capability to acquire democratic content, for it provides them with the
opportunity to participate in the public sphere, which makes this sphere function well.
This in turn is necessary to enhance the quality of decision making in formal democratic
bodies and to have checks against these bodies. A well-functioning democracy is
required because this form of government has – at least in complex modern societies –
the best chances of creating a society in which people can become full agents and have
equal standing.14 Thus, the capability to acquire democratic media content is required as
a prerequisite for the well-functioning of the type of political system that is best placed
to promote each person’s capacity for agency. Admittedly, this is a rather abstract
argument. Therefore I will hereafter take a closer look at the relations between the
media, the public sphere and democracy, scrutinizing arguably the most influential
account of these relations: Jürgen Habermas’ study of The Structural Transformation of
the Public Sphere.15
Before doing so, the sketch of our normative theory of the media needs one
addition. For the media do not only deliver democratic content, but also non-democratic
or – for brevity’s sake – “entertaining content”.16 Should our normative theory include
an imperative for people to have a capability to such content? Arguably, the media
should be allowed to let viewers have their daily portion of soap series, quizzes and
sports matches. None of these activities can be labeled as immoral. Neither can they be
considered morally required, however, for it is difficult to see what entertainment would
add to the development of agency. They are morally permissible. This classification is
important, for in cases of conflict the realization of morally required capabilities takes
priority over morally permissible capabilities. To the extent that it is practically
impossible to promote both, the promotion of the capability for entertaining content will
have to yield to the higher-order capability to acquire democratic content. Nonetheless,
the non-democratic capability represents a separate source of normative claims on the

6
media, even if only of the permissible kind. This is important, since our main question is
about the suitability of the market mechanism for the media context. Since it is rather
uncontroversial that the market is best equipped to realize entertaining content, we can
now conclude that media products should not be “blocked exchanges;” there is a
legitimate scope for market-based media. The difficult matter is whether the capability
to acquire democratic content can also be satisfied in a market context or requires non-
market media.

2. The Media and The Public Sphere

In this section I will argue for a specific relation of the media to democracy on the basis
of a close reading of Habermas’ book on the public sphere.17 I draw from it two
different theses about the effects of the mass media on the public sphere and democracy.
This will lead me to formulate more precisely the extent to which the media is able to
support the public sphere and democracy. I will focus on Habermas’ earlier work
because this text displays the features that are important to my argument in the most
exemplary fashion. To keep the discussion tractable, I will only briefly suggest how this
relates to his later work on the media at the end of this section.
Following Habermas’s original study on the subject, the ideal of the public
sphere is the ideal of a sphere of rational-critical debate on the part of citizens
deliberating among each other. This sphere arose in the 18th century as the rising
bourgeois class emancipated itself from the state and started to discuss political and
administrative matters. Discussions were not confined to politics; cultural and literary
matters were equally prominent, as the public sphere provided the opportunity to discuss
the new experience of subjectivity that simultaneously entered the sphere of the private
bourgeois household.18 The English coffee houses, the French salons and the German
Tischgesellschaften were the prototype institutional platforms of this public sphere, but
the role of the media was also essential. The press initially emerged in response to the
needs of merchants engaged in long-distance trade: along with intensified traffic came
the need for traffic in news. In a second step, many of the “political journals” fell into
the hands of state authorities who used them to make public their decrees and control
the stream of information. Only in the third instance these journals came to exhibit
criticism and debate, as genuine organs of the public sphere.19

7
 From this point on, two quite different themes can be stressed. The first theme,
dominant in Habermas’s account – and in media scholars following him – is that the
market subsequently captured the media. Once controlled by capitalist interests, the
media actively contributed to the decline of the public sphere. I will call this the “market
subversion” thesis. A second theme, more subterranean in Habermas, is that the mass
media in and by themselves, i.e. even when abstracted from their market-based
organization, subverted the public sphere. I will call this the “media subversion” thesis.
It is necessary to keep both theses analytically separated, even though in empirical
reality they may be intertwined.
According to the market subversion thesis, the market initially had a beneficial
influence on the public sphere because it provided the means through which people
could participate in the public sphere (books, journals, theater tickets, etc.). After a
while, however, the media turned against the public sphere and commodified culture
and critical debate itself, transforming it to fit prepackaged formats easily digestible by
large audiences. As Habermas writes, ‘Today the conversation itself is administered.
Professional dialogues from the podium, panel discussions, and round table shows – the
rational debate of private people becomes one of the production numbers of the stars in
radio and television, a salable package ready for the box office; it assumes commodity
form even at “conferences” where anyone can “participate.”’20
The underlying mechanism that Habermas identifies is that the standards of
debate are lowered so that a broad audience has access, but what it has access to in no
way resembles the original ideal of a public sphere in which participants discussed
social and political issues freely. The market not only gave the masses economic access
to cultural goods, but ‘it also facilitated access for broad strata psychologically.’ As a
consequence, ‘[t]his expanded public sphere, however, lost its political character to the
extent that the means of “psychological facilitation” could become an end in itself for a
commercially fostered consumer attitude.’21 When explaining why the media would
want to transform the public sphere in the way described, Habermas refers to the advent
of “the advertising business” that in the course of the 19th century came to dominate the
internal organization of the media.22
If we follow the media subversion thesis, the picture is rather different. In
Habermas’s account, the direct and live discussions in coffee houses (and their present
day equivalents) are emblematic for the interactions in the public sphere. If this is so,
then the mass media must always misrepresent the nature of that sphere. Debate in the

8
media is always staged, artificially constructed. Whether a TV show is organizing a
debate around a social issue of the day or a journalist interviews citizens on matters of
public policy; the form and content of the resulting exchange is always in the hands of
the media provider. The provider selects the topic, the guests, the questions etc. From
this perspective media communications cannot escape being a product. It is always a
constructed item that is communicated, not the representation of a spontaneous
discussion with open access for all. This product nature of media communication is
given with its “one-to-many” character, whereas communication in the public sphere
typically is “many-to-many.”23
When Habermas tells us that in the early days of the public sphere this was
otherwise, we see how the media’s being part of the public sphere depends on a very
peculiar setting. For here Habermas describes the functioning of weekly periodicles that
were so localized and small-scale that they could be considered as adjuncts (“integral
parts”) to the live discussions in the public sphere.24 But these journals are only vaguely
reminiscent of today’s mass media; they look more like websites destined to (and
accessed only by) members of a specific club. From this perspective, the tale of the
decline of the public sphere emphasizes a rather different development: the decline of
the original discussions in the coffee houses due to the rise of state bureaucracies,
special interest groups and political parties that closed the void between the private
sphere and the state. This in turn provoked a new role for the media, much better
integrated with their large-scale technological expansion on a commercial basis; that of
“public opinion formation,” the conscious manipulation of the public for the sake of
parochial interests.25
The difference between these two explanatory accounts is crucial for the
market’s relation to the ideal of a public sphere. If the market subversion thesis is
correct, then the media’s role in sustaining a public sphere can in principle be salvaged,
by taking the media out of the hands of commercial (advertising) interests. The state
could take over the media and establish public broadcasting.26 If the media subversion
thesis is correct, however, then there is no escape from private interests. The
“administration of conversation” will happen anyhow, not only when the media is taken
over by the market, but also when media products are manufactured through alternatives
such as state provision. A non-market-based mass media will also be dominated by
groups who capture the relevant organizations. The objection against the market then

9
reduces to the market’s administration of conversations in favor of commercial interests
rather than other (potentially more benign?) interests.
We could also try to reconcile both theses, treating them not as mutually
exclusive but as mutually reinforcing explanations. The media subversion thesis forces
us to admit that the real public sphere always occurs outside of the manufacturing of the
media’s products. Nonetheless, just as a vibrant public sphere is an indirect warrant for
the quality of decision making in the official political sphere, the media are an indirect
testimony to the civic debate in the public sphere. The media support the debate in the
public sphere in their products. In this supportive function, the media channels
information from the society at large to the formal political arenas and vice versa. It
informs citizens about the plans of political actors, communicates about political
deliberations, exposes corruption within political bodies, etc. At the same time it
informs political bodies about the desires and actions of citizens within the public
sphere.
Thus formulated the ambition is more modest, not claiming a role for the media
that it cannot fulfill.27 Its supportive role would still be quite ambitious, however, in
aiming at a representation unhindered by commercial or other particular interests
(including state interests!), driven only by the wish to reflect what is going on among
the public and in formal political decision-making bodies. At this point the market
subversion thesis shows this supportive role is endangered as soon as commercial
interests predominate.
It remains somewhat unclear whether a reading of Habermas’ work after his
early Structural Transformation affirms or denies this conclusion. In a 2006 article
Habermas seems to throw all the blame for the artificial construction of debate on the
market’s influence.28 In this account the market intrudes the media, which in itself is
characterized by an emancipatory potential. The groundwork for this construction is
arguably laid in the 1980s Theory of Communicative Action, where Habermas subsumes
the mass media under the category of ‘generalized forms of communication’ (i.e.
influence and value commitment) rather than the steering media (money and power),
and stresses the emancipatory potential of the mass media.29 In an article on the public
sphere from 1992, however, we find Habermas faulting both the media’s own ever-more
centralized logic and the market’s influence for weakening the public sphere. 30 The
same line arguably is taken in Between Facts and Norms (1996).31 These texts support
the juxtaposition of the market subversion thesis and the media subversion thesis.

10
 This latter argumentative strategy is in line with my position. However, there
remains one point of dispute. Habermas still articulates very ambitious normative claims
for the mass media.32 In contrast, my position here is that even if the market doesn’t
have any influence in the media, the normative role to be attributed of the media in
supporting the public sphere will have to be more modest than Habermas’ version of the
standard argument allows.

3. Media and the Market

Now I turn to the second premise of the standard argument, that the market is unable to
promote the required capability; that it will tend to offer too little – if any – democratic
content. Sometimes an explanation of such low levels is sought on the supply side. For
producers, it is often more attractive to produce non-democratic content. The costs of
producing a soap opera are lower than the costs of producing high-quality drama and the
costs of producing investigative journalism are higher than the costs of newsroom
interviews.33 However, if consumers would press hard enough, it seems that cost
problems could be overcome. It is telling that even if democratic content is offered free
of charge – for example on public television – it attracts substantially smaller audiences
than non-democratic content. This shows that the problem doesn’t predominantly lie on
the supply side. Economies of scale make it attractive to produce democratic content at
sufficient levels once it is in wide demand, so there must also be explanations why
demand is lacking.
The standard argument argues for the lack of demand by employing a distinction
between preference satisfaction and preference formation. Media content either caters to
existing preferences (preference satisfaction) or leads to the establishment of new
preferences (preference formation). Democratic content typically triggers a process of
preference formation on the part of receivers. It stimulates a creation of preferences
about objects which were formerly not included in one’s preference ordering and aims
to challenge existing beliefs, so that people are required to consider a revision of
previously held preferences. Consumers will not exercise (sufficient) demand for this
kind of preference-forming content since their preference for it can only be formed by
already consuming it. And because consumers don’t demand democratic content,
market-based media will not offer it and confine themselves to content that appeals to
people’s wishes to have their actual preferences satisfied. For example, Russell Keat

11
argues that ‘[t]elevision “ratings wars” provide plentiful examples, with programmes
carefully constructed to provide audiences with “just what they (happen to) want,”
forcing out others which, by virtue of their transformative value, present something to
their audiences which may challenge those preferences.34
On the basis of this premise, combined with the previously discussed normative
premise, the standard argument concludes that a collective decision should be made to
provide non-market media. Such a decision shouldn’t be seen as illegitimately
overriding people’s preferences (paternalism), for preferences are (partly) a function of
the setting in which they are expressed. The content of preferences expressed in the
market and that of preferences raised in a process of collective decision making may
therefore legitimately diverge. By expressing a preference for the provision of
democratic content in the latter setting, citizens correct their market behavior. It is a case
of auto-paternalism.35
In my view this argument must be rejected. I can see no reason why a media
consumer couldn’t express a demand for media content which is ‘transformative’ of her
beliefs in social and political matters, helping her to reflect on issues important to actual
democratic deliberations. ‘Preference formation’ itself can be considered as a property
of certain types of media content (‘democratic content’), and consumuers can demand
content which has that property. Just as consumers can demand to be entertained they
can demand to be challenged in their beliefs. Whether or not they are likely to ask for
such transformative content is another matter; but there is no principled obstacle here.
The fact that many consumers do ask high quality content (e.g. for newspapers) proves
this point. A defender of market-based media might therefore argue that the market is
very well capable to cater to the demand for democratic content to a sufficient extent.
She could in support introduce the following thought experiment.
Imagine that ideal-typically the media market consists of two types of
consumers. The first type has a preference for democratic content, but is in the dark
about which media goods contain this kind of content. Consequently these consumers
may fail to exercise a demand, even though they have the required preference. Call a
person suffering from this problem the democratic-content-seeking consumer. The
problem arises because most media goods are either “experience goods” or “credence
goods.”36 For experience goods, one has to become acquainted with their content to
know their value. Only after consuming it for some time one knows its value. A soap
series is an experience good. With credence goods one cannot evaluate their value, even

12
after consumption. The daily news is a credence good. It is difficult to evaluate its
accuracy in reporting on “reality” since one watches it precisely to learn about the state
of reality. Both types of goods lead to an information problem. Here the market however
may provide a solution, through the possibility for producers of establishing a
reputation. Review sites, consumer organizations and independent prizes may all help to
inform consumers which providers serve their mission well. Thus, the reputation
mechanism helps consumers to act upon their preferences by buying media content from
providers reputed to deliver the desired content. If one prefers being a better-informed
citizen, one can take a subscription to “Citizen Channel.” Democratic-content-seeking
consumers are auto-paternalist but the market serves their auto-paternalism well.
A second type of consumer is simply too lazy, short-sighted or occupied to
purchase democratic media content. Whatever the exact motivation, he has no
preference to that end. The problem of these uninterested consumers does not originate
in a purported lack of information. They know it would be socially better if they
consumed democratic content; nonetheless they neglect their capability to acquire
democratic content. They may complement this attitude with two different policy
stances. Either they prefer to have democratic content available to all (delivered by non-
market media, financed by taxpayers), as an insurance scheme in case of the unlikely
event that they relieve themselves of their own lack of motivation; or – more likely –
they refuse supporting such a scheme and give up on their prospects for preference
formation with regard to democratic content altogether. The existence of the latter group
raises a dilemma: should we coerce them into collective payments for the supply of
democratic content or respect their autonomous wish not to be involved? If we grant
these persons their way of life (as it seems we should), then the market gives them all
they want. Collective action for non-market media cannot be justified on the basis of
this group of consumers.
Proponents of a completely market-based media will say that the two groups of
consumers exactly represent the actual media audience. Either consumers are auto-
paternalist and the market can help them out or they are not auto-paternalist and then
democratic content needs to be offered outside of the market, but this qualifies as real,
unjustified paternalism. Some are spontaneously interested in watching democratic
content and find the market rewarding their demand; others are not interested and
should be left alone. The morally required capability to acquire democratic content is
safeguarded by the market, since market demand expressed by the first group creates

13
supply of democratic content, so that those from the second group have supply available
in case they change their mind.

4. A Limited Justification for Non-Market Media

Is there a way around this conclusion? Can the standard argument be saved? I think it
can, but only to a limited extent. In this section I will argue that there is a justification
for non-market provision of democratic content, in two narrowly circumscribed
circumstances. It is worth noticing at the outset that the justification is contingent on
these circumstances actually obtaining. The market need not fail.
The first situation is where, due to whatever reasons, democratic content is
threatened with extinction. This occurs where the actual level of market supply is so low
that we cannot reasonably expect prospective consumers to be able to find this content
and become acquainted with its value through consumption. For example, if markets –
on the basis of actual demand – would only provide one hour of serious political
discussion a week, this is probably insufficient to become accustomed and develop a
taste for it. In these circumstances the market fails to realize the capability to acquire
democratic content to a meaningful extent. A fall below this threshold justifies provision
of non-market media. This justification could be rightfully invoked at an imaginary
“first day” of a society, where citizens have no recollection of democratic content
whatsoever (for example, after a revolution overturning decades of dictatorship and state
propaganda); or a society that has suffered a very gradual but ultimately near-to-
complete erosion of the demand for democratic content. At any rate, this justification
applies to a kind of emergency situation where the level provided by the market is too
low. It will not obtain very often; normally market supply will be above this level.
A further-ranging justification arises in a second type of situation, in which
market supply of democratic content is higher than this emergency level, but lower than
the level needed to make the public sphere function properly. The emergency level may
be too low for this because it only ensures consumption of democratic content by a
small niche of dedicated citizens. For a proper functioning of the public sphere, wider
citizen participation may be needed. This diagnosis changes the normative claim itself.
In this situation, the reason for justifying attempts to push consumption to a higher level
by taking collective action simultaneously at both sides – raising the level of supply and
trying to stimulate demand – can only be found in an independent requirement for high

14
levels of actual consumption. We do not rest content with a situation in which everyone
has the capability to acquire democratic content but few convert it into functioning. A
sufficient level of functioning is now required. Thus, this second type of justification
presupposes that the public sphere needs an active citizenry of democratic content
consuming citizens. Supposing this to be the case, let us see how the argument could be
developed.
The justification for providing non-market content up to this higher level can
best be explained by a criticism of the market proponent’s thought experiment from the
previous section. In addition to the democratic-content-seeking consumer and the
uninterested consumer, we introduce a third ideal-type consumer, who has adequate
information about the reputation of providers and is not uninterested. However, he does
not want to engage in consuming democratic content if others do not do the same. This
calculating consumer conspicuously observes other citizens’ behavior before deciding
to watch democratic content. Elizabeth Anderson explains the general problem his
general attitude gives rise to: ‘Because markets don’t give consumers control over
others’ decisions, they tend not to be effective vehicles for satisfying the preferences
individuals have that are conditional on their confidence that a large number of other
people will behave likewise. Call these large-scale conditional preferences.’37
If the preference for democratic content is a large-scale conditional preference,
this may seem to save the case for non-market media.38 For now we can explain the lack
of market demand for such content by reference to the classical free-rider problem. The
consumption of democratic content is a civic duty media consumers will only bear if
they receive signals that others are doing their part. This would explain why many
consumers do not grasp the opportunity to become acquainted with democratic content
through consumption of content already available (as a response to the demand
expressed by the small group of democratic-content-seeking consumers). For they think
their individual contribution will not have much of an effect on the overall quality of the
public sphere.
This solution faces a problem of its own. It assumes collective action will be
effective in remedying the individual’s motivational defect. With normal examples of
collective action problems this assumption is relatively unproblematic. For example,
citizens can decide to correct their consumer preferences for products whose price
doesn’t internalize negative pollution externalities by forcing producers to obey
environmental regulations. Or they can protect themselves against the consumption of

15
unhealthy products by severely restricting opportunities for their sale and consumption
(e.g. smoking in public places). For the media, however, citizen action at the level of
policy making cannot on its own resolve the collective action problem. Here individuals
have to take up their citizen role at two levels. First at the political level they have to
make a collective decision; and then at the level of media consumption they have to
actually watch the collectively established content. Even if people vote for producing
democratic content (e.g. by a public broadcaster) this doesn’t guarantee they actually
watch it sitting at home. Normally collective action is effectively designed as a legal
obligation backed up with sanctions; smokers in public buildings and polluting firms are
penalized for their behavior. For the media, a collective decision to provide democratic
media content is insufficient as long as media consumers cannot be penalized when not
actually watching (in the absence of a media police force intruding their homes, forcing
them to watch…).
This dependence on voluntary action is partly a blessing. After all, a defining
characteristic of civic virtue is that it involves taking upon oneself civic obligations
voluntarily; without strategically waiting for assurances that others will move too. In
this regard, the obligation of citizens to inform themselves and reflect upon political
affairs with the help of media content is analogous to the obligation of voting. One is
supposed to engage in it not because others do so, but because one wants to be member
of a community in which others do so for that very same reason. The voluntary nature of
the additional effort needed on the part of the citizen is a constitutive part of the end
goal (the establishment of a democratic society); not a merely instrumentally necessary
effort that one may abstain from in the absence of simultaneous efforts by others. The
flip side of this coin is that the collective effort to provide democratic content is
vulnerable to disintegration. When some abstain from watching democratic content,
others will wonder why they themselves should keep on paying the required taxes and
watch. Arguably, other policy measures (like education) will then be required to
convince people of the value of actually consuming democratic content.

Conclusion

This paper has argued for several revisions of the standard argument about the
justification of non-market media. The normative claim was formulated as a claim about
the capability to acquire democratic content, the substance of this claim implying a more

16
modest role for the media in supporting the public sphere than usually acknowledged.
The second premise of the standard argument was transformed into a more
circumscribed picture of two specific occasions on which the market will fail to provide
democratic content. This led to a limited and conditional justification for non-market
media. This leaves the following tasks ahead.
First, the two justifications for non-market media are contingent on the portrayed
circumstances actually obtaining. Strictly speaking, this assumes one would have to wait
until supply actually falls below one of the two levels before one starts providing non-
market content. If one is somewhat less strict, one can also argue non-market provision
should be delivered prospectively, as a precautionary measure, because it would be too
harmful if democratic content actually falls below one of these levels. Only if we grant
such a relaxation we can argue in favour of permanent regulation or public broadcasting.
But this relaxation is not innocent, for it harbors the danger that non-market provision
continues for long periods of time when there would be no justification for it (because in
the counterfactual situation levels of market demand for democratic content would have
been high enough). Therefore, more argument is needed to make this leap from the
presently offered justification to such a permanent scheme based on a precautionary
argument.
Second, the line of thought offered in the third and fourth sections has
presupposed a somewhat abstract picture of markets obeying standard laws of supply
and demand. In this abstract model democratic content only becomes available to the
extent that consumers ask for it. Actual market configurations may diverge from this
standard in different ways. Our conclusions would have to be modified to the extent that
the supply of democratic content is different from what the abstract model would
dictate. For example, some media suppliers may offer more democratic content than is
strictly asked for by consumers, since they feel it their social responsibility to do so, or
because they have the financial means to make a loss on it. They might use cross-
subsidization (as publishing houses sometimes do with their books), paying for
democratic content out of the profits made on entertaining content. Another factor might
be that markets in small countries may have more difficulty to deliver democratic
content than larger countries (given economies of scale), making the case for non-
market provision in these countries easier to make than in larger countries. These and
other differences in market configurations will have to be taken into account to fine-tune
the conclusions to be drawn from the general argument offered in this paper.

17
 Third, it is still open whether non-market provision should take the form of
public regulation of private media or of public provision by a public broadcaster.
Arguments can be given for both these main options (and several variations thereof).
This issue cannot be resolved on the basis of the justificatory question dealt with here.
This paper has merely provided the groundwork for a debate on the more specific
institutional form of non-market provision. Finally, this also leaves ahead the task of
identifying conflicts with market-based media, usually but not exclusively specialising
in all kinds of entertaining content. A solution for these conflicts requires both
democratic and entertaining content finding a satisfying place in the modern media
landscape.

1
Reported in C. Edwin Baker, Media, Markets and Democracy (Cambridge: Cambridge University Press,
2002), 3.
2
Michael Walzer, Spheres of Justice. A Defense of Pluralism and Equality (New York: Basic Books,
1983).
3
For important positions in this debate, see Elizabeth Anderson, Value in Ethics and Economics
(Cambridge Massachusetts: Harvard University Press, 1993). Margaret Radin, Contested Commodities.
The Trouble with Trade in Sex, Children, Body Parts, and Other Things (Cambridge Massachusetts:
Harvard University Press, 1996).
4
Richard Posner, "Bad News," The New York Times Book Review July 31 (2005).
5
For closely resembling versions, see Baker, Media, Markets and Democracy, 80-95. Cass Sunstein,
"Television and the Public Interest," California Law Review 88 (2000): 511-23. Jürgen Habermas,
"Medien, Märkte Und Konsumenten," Süddeutsche Zeitung, May 16th/17th 2007. Robert W. McChesney,
The Problem of the Media. U.S. Communication Politics in the 21st Century (New York: Monthly
Review Press, 2004), 198-204. Russell Keat, Cultural Goods and the Limits of the Market (London:
MacMillan, 2000), 151-61. There also is a debate on the justification of public broadcasting indirectly
discussing these themes. See Mark Armstrong, "Public Service Broadcasting," Fiscal Studies 26, no. 3
(2005). Shaun Hargreaves Heap, "Public Service Broadcasting," Economic Policy (2005). John O’Hagen
and Michael Jennings, "Public Broadcasting in Europe: Rationale, License Fee and Other Issues," Journal
of Cultural Economics 27 (2003).
6
E.g. Baker, Media, Markets and Democracy, 44. Sunstein, "Television and the Public Interest," 517.
7
About the difficulties of making “externalities” the basis of normative judgments, see Michael J.
Trebilcock, The Limits of Freedom of Contract (London: Harvard University Press, 1993), 58-77.
8
As regards the main candidate, cultural goods, I must confess having some scepticism. However
intrinsically valuable, it is hard to see that people cannot be full agents if a society would not harbor

18
opportunities to engage in consuming art works. A way around this conclusion would be to argue that
certain forms of art are necessary to the functioning of democracy, so that they qualify as democratic
content. I remain open to the possibility that such an argument could be successful, but will restrict myself
to democratic content that directly supports the public sphere.
9
E.g. Martha Nussbaum, Women and Human Development. The Capabilities Approach (Cambridge:
Cambridge University Press, 2000), 112-48. Amartya Sen, The Standard of Living (Cambridge:
Cambridge University Press, 1987), 1-19.
10
I thank one of the anonymous reviewers for this point.
11
For an elaborate defense of this theory, see Rutger Claassen, "Institutional Pluralism and the Limits of
the Market," Politics, Philosophy, and Economics 8, no. 4 (2009).
12
This formulation doesn’t say anything yet about the appropriate level of democratic content that people
should have a capability to acquire; for that see section 4.
13
Baker, Media, Markets and Democracy, 93-94.
14
Admittedly, a defense of this claim requires a full-scale comparison between the scores of different
political systems. In the absence of such a comparison, we rely on the conceptual connection between an
ideal-type democracy’s principle that each person deserves equal political influence (one man one vote)
and the society-wide ideal of equal protection for each person’s capacity for agency.
15
For the influence of Habermas in media scholarship, see Baker, Media, Markets and Democracy, 138
ff. R. Randal Rainey and William Rehg, "The Marketplace of Ideas, the Public Interest, and Federal
Regulation of the Electronic Media: Implications of Habermas' Theory of Democracy," Southern
California Law Review 69 (1996). James Curran, "Mass Media and Democracy Revisited," in Mass
Media and Society, ed. James Curran and Michael Gurevitch (London: Arnold, 1996).
16
This dichotomy is an ideal type. It allows for nuances in application, such as that certain popular
programs normally considered to be in the genre of entertainment can occasionally arouse interest in
democratic matters. Also I leave open the possibility that the dichotomy could be supplemented with
separate categories of non-democratic, non-entertaining content (such as cultural content).
17
By focusing on one text we hopefully gain precision; the drawback is that other accounts of the
relations between media and democracy don’t get specific attention. Although warranted by the influence
Habermas had on many other accounts, this restricts my argument. Ideally the same exercise would have
to be repeated for other accounts.
18
Jürgen Habermas, The Structural Transformation of the Public Sphere (Cambridge: Polity Press, 1989
[1962]), 29.
19
Ibid., 15-16, 20-22, 24-25.
20
Ibid., 164.
21
Ibid., 166 and 69.
22
Ibid., 184-88. Habermas is somewhat unclear about the exact causes of this transformation. He refers to
the fact that the press “could abandon its polemical stance and concentrate on the profit opportunities for a
commercial business” because of the “establishment of the bourgeois constitutional state and the
legalization of a political public sphere.” Habermas, The Structural Transformation of the Public Sphere,

19
184. However, he also refers to the “necessity” of attracting capital for the “upgrading and perfection of
the technical and organizational apparatus.” ———, The Structural Transformation of the Public Sphere,
185.
23
Baker, Media, Markets and Democracy, 302. The internet, insofar as it is a sphere where people can
communicate directly and spontaneously with one another, escapes this predicament of the other media.
For a criticism of high expectations about the internet, see C. Edwin Baker, Media Concentration and
Democracy. Why Ownership Matters (Cambridge: Cambridge University Press, 2007), 97 ff.
24
Habermas, The Structural Transformation of the Public Sphere, 42. Habermas’s position that the
periodicals are an integral part of the public sphere directly contradicts his earlier statement that the
products of the media market remained outside of the public sphere itself.
25
Ibid., 197-203.
26
Ibid., 187-88.
27
A more detailed account would be needed to elaborate on the structure of the public sphere itself and
the demands upon the media, but this formulation of the media’s “core business” suffices for our
purposes. One further issue is the tension between the liberal and the republican perspectives on the
public sphere. Nancy Fraser, "Rethinking the Public Sphere: A Contribution to the Critique of Actually
Existing Democracy," in Habermas and the Public Sphere, ed. Craig Calhoun (Cambridge,
Massachusetts: The MIT Press, 1992).
28
Jürgen Habermas, "Political Communication in Media Society: Does Democracyt Still Enjoy an
Epistemic Dimension? The Impact of Normative Theory on Empirical Research," Communication Theory
16 (2006): 422.
29
———, The Theory of Communicative Action Volume Two. Lifeworld and System: A Critique of
Functionalist Reason (Boston: Beacon Press, 1987), 390.
30
———, "Further Reflections on the Public Sphere," in Habermas and the Public Sphere, ed. Craig
Calhoun (Cambridge, Massachusetts: The MIT Press, 1992), 436-37.
31
———, Between Facts and Norms. Contributions to a Discourse Theory of Law and Democracy
(Cambridge: Polity Press, 1996), 376-7.
32
For Habermas’s later formulation of the normative claim, see Ibid., 378.
33
McChesney, The Problem of the Media. U.S. Communication Politics in the 21st Century, 201.
34
Keat, Cultural Goods and the Limits of the Market, 158. Later Keat recognizes the problem I discuss in
the following paragraphs, but doesn’t provide a satisfactory answer (to my view).
35
Baker, Media, Markets and Democracy, 86-87. Sunstein, "Television and the Public Interest," 522.
36
John H. McManus, Market-Driven Journalism: Let the Citizen Beware? (London: Sage Publications,
1994), 65. Experience and credence goods contrast with normal consumer goods, which are “search or
inspection goods… whose quality is evident merely from examining them.”
37
Elizabeth Anderson, "Consumer Sovereignty Vs. Citizens' Sovereignty: Some Errors in Neoclassical
Welfare Economics," in Freiheit, Gleichheit Und Autonomie, ed. Herlinde Pauer-Studer and Herta Nagl-
Docekal (Wien: R. Oldenbourg Verlag, 2003), 383. Both the democratic content-seeking and the

20
calculating consumer use a “precommitment strategy”, the first a private strategy, the latter a collective
strategy.
38
Do people have different preferences in private compared to collective contexts, or are their preferences
continuous but they are unable to attain their highest-valued preference for cooperation in a private
context? In the latter case, the dilemma takes the form of an assurance game (not of a prisoner’s dilemma,
where the highest-valued option is non-cooperation). See Daphna Lewinsohn-Zamir, "Consumer
Preferences, Citizen Preferences, and the Provision of Public Goods," The Yale Law Journal 108 (1998):
391-96. In the media context, see Cass Sunstein and Edna Ullmann-Margalit, "Solidarity Goods," Journal
of Political Philosophy 9, no. 2 (2001): 141 and 43-44. Sunstein, "Television and the Public Interest,"
520.

21