SciELO - Scientific Electronic Library Online

 número15La obligación moral de cumplir los contratos: algunas consideraciones preliminaresDerecho y moral (razón práctica): conexiones en tiempos del Estado de derecho constitucional índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Servicios Personalizados




Links relacionados

  • No hay artículos similaresSimilares en SciELO


Problema anuario de filosofía y teoría del derecho

versión On-line ISSN 2448-7937versión impresa ISSN 2007-4387

Probl. anu. filos. teor. derecho  no.15 Ciudad de México ene./dic. 2021  Epub 31-Mayo-2022 


The role of morals in law: some reflections

Advertising as Guaranteeing: a Defense of Strict Liability *

La publicidad como factor de garantía: en defensa de una estricta responsabilidad

*University of California, Irvine, mail:


In many jurisdictions, advertisers are held strictly liable for false statements of fact. This may seem to depart from the ordinary moral treatment of speech, whereby individuals describing their activities -the food they cooked last night or the room they prepared for guests- are held only to being sincere and careful about what they say; if they tried hard to get things right, they are morally in the clear if they turn out mistaken. That may invite the thought that strict liability for false advertising is either a departure from ordinary morality or a special context justifying a different moral standard from the norm. Here, to the contrary, I argue that commercial claims often take the form of a familiar type of speech act, namely guarantees, for which, I try to show, speakers are already morally responsible for accuracy, regardless of fault. For that reason, I argue, strict liability for false advertising is not only justifiable, but natural and inevitable.

Keywords: Legal Strict Liability; Inviting Reliance; Guarantees; Ethics of Guaranteeing; Corporate Product Descriptions


En muchas jurisdicciones, los publicistas son considerados estrictamente responsables de las declaraciones falsas de hechos. Esto puede parecer que se sale del enfoque moral ordinario del discurso, según el cual los individuos describen sus actividades -la comida que prepararon anoche o la habitación que prepararon para los invitados-, sólo tienen que ser sinceros y cuidadosos con lo que dicen; si se esfuerzan en hacer las cosas bien, moralmente están libres de culpa si se equivocan. Esto pudiera dar pie a que se pensara que la responsabilidad objetiva por la publicidad falsa constituye una desviación de la moralidad ordinaria o un contexto especial que se justifica por un estándar moral distinto a la norma. En este caso, sostengo, más bien, que las deudas comerciales suelen asumir la forma conocida de una acción verbal, en concreto garantías, de las que pretendo demostrar que los interlocutores ya son moralmente responsables de la veracidad, independientemente del error. Por esa razón, sostengo que la responsabilidad objetiva de la publicidad falsa no sólo es justificable, sino que es natural e inevitable.

Palabras clave: Estricta responsabilidad legal; invitación a la confiabilidad; garantías; ética de garantía; descripciones de productos corporativos

Summary: I. Legal Strict Liability. II. Inviting Reliance. III. Guarantees. IV. Objections. V. References.

Clara is allergic to peas, so she asks her friend Jon if his homemade sugar-free iced tea has a pea ingredient. After considering every detail of how he made the drink, he answers her: “No, I used no such ingredient. Would you like some?” She drinks it and, despite his reassurance, suffers an allergic reaction to the tea’s sugar substitute, aspartame, which -unbeknownst to Jon- was derived from peas.2 Is Jon to blame? May he be held accountable? One reason why not, it might be said in his defense, is that his claim about the tea was reasonable, sincere and made in good faith. In short, he was blameless in coming to the false conclusion about the tea and voicing it to Clara, and for that reason, he should not be held accountable for it.

Imagine, instead, that Clara passes a display table offering sample cups of Slapper Ice Tea. Ringing the bottle are the stickered words: “Allergen-free: no pea, bean, nut or soy ingredients,” a claim backed up by the list of contents below. She drinks it and, again, has an allergic reaction, this time because Slapper’s supplier had secretly replaced the expensive sweetener it was contracted to provide with the much cheaper aspartame, of the form Jon had used. The company never found out. Can Slapper avoid responsibility in just the way Jon might, by claiming -plausibly enough- that its production of the tea, though injurious, was entirely without fault?

The law, at least in the US and a number of other countries, says no: Slapper -like any other company- is strictly prohibited from falsely describing its products in its public or promotional material.3 It does not matter whether it could have avoided the error; its good faith or reasonable judgments about the matter -even taking into account its greater access to information than a layperson like Jon- are irrelevant. While it may avoid criticism, it cannot avoid accountability of some sort for the harm it inflicted by way of its misdescription.

Examples of this type of strict liability in the law are by now familiar, as is the way they often depart from ordinary moral treatment of the behavior involved. As such it is controversial, effectively punishing behavior that is morally above reproach and beyond the agent’s control. But the practice has been defended on grounds that the parties being held to it are uniquely suited to adhere to it, or better bearers of the cost of the harm they inflict, even faultlessly, than typical moral agents, or because some valuable purpose is served by enabling it in this domain, unlike in others.4 These sorts of arguments tend to view strict liability as a special context, in which we may be justified in treating parties differently from how we would ordinarily treat agents engaged in the same or similar behavior.

Here, in contrast, I will defend strict liability for false advertising, like Slapper’s misdescription, as a direct application of pre-existing moral principles, specifically the principles governing communication. I will argue that such corporate claims as “Lactose-free” or “lead-free” or “contains no pea ingredients” are instances of a type of speech act I call guarantees, which are morally held to being uttered accurately, not just sincerely or otherwise blamelessly. I will motivate the strict moral treatment of guarantees generally, and explain how it applies to advertising, while it does not apply to ordinary assertions, like Jon’s claim that his tea was pea-free.

I. Legal Strict Liability

In the later half of the 20th century, the U.S. undertook new measures to ensure fair competition among venders and manufacturers, specifically by expanding upon the restrictions already imposed in tort against misrepresenting their products, especially as they compare to those of their competitors. Most famously, the Lanham Act and various subsequent state laws provided that commercial labelers, and even advertisers more generally, are held to accurate descriptions of their products, such as “caffeine-free” or “BPA-free”, regardless of whether any misrepresentation is intentional or even negligent.5 A merely “false… description of fact” that “misrepresents the nature, characteristics, [or] qualities” of the goods one is selling is grounds for tort damages, even on behalf of those who suffered no harm but merely “believe” they are “likely to be damaged” by it.6 Moreover, such misrepresentation grounds restrictions on import.7 Again, such legal sanction is imposed regardless of fault; what matters simply is that claims about the product are false. The Federal Food, Drug, and Cosmetic Act, in turn, bans labeling in a way that is “false or misleading in any particular”.8

Not all misrepresentation is prohibited by these legislative measures. Statements such as “The leading painkiller nationwide,” or “You will like the way you look!” are generally tolerated even if found false in some way or, at the very least, not known to be true by the speaker. These self-praising claims, known as “puffery,” are protected, at least in the U.S.9 Other countries have stricter standards, especially where the advertising is for products that bear heavily on public health, like food, beverages and medicine. Mexico’s laws against false advertising, for example, impose liability not only for claims that are untrue but for any that could be read in a way that misleads consumers, regardless of whether it is the most straightforward or reasonable reading.10

Exactly what constitutes misrepresentation for legal purposes remains the bone of vigorous contention and litigation: is it a literal misrepresentation or merely a representation that is likely to be interpreted misleadingly? Must such “false” interpretations be reasonable to be actionable, or need they only be likely? For present purposes, I will sidestep these sorts of questions, only because the focus here is on whether misrepresentation, however it is understood, should be treated with strict liability, as opposed to conditioning liability on whether the misrepresentation -again, however we understand it- was blameless. Of course, in both law and everyday morality, there are complicated questions about how speech is to be understood, and from whose perspective (speaker or audience or neither), for purposes of determining whether it is misrepresentative, deceptive, accurate, and so on. Granting all that, however, it remains less controversial that there are genuine cases of intentional misrepresentation, for which one is morally accountable as a liar or something similar, while there are, on the other hand, cases of unwitting and blameless misrepresentation. Also uncontroversial is that, as a general matter, we do not hold people accountable for expressing their factual opinion about some matter merely because it misrepresents the fact, assuming they tried hard not to do so. And that is why the legal treatment of advertising is noteworthy: in quite a few jurisdictions, the law does impose liability even for blameless cases of misdescription.

Of course, ordinary morality is not the same thing as morality; there may be good moral grounds for imposing strict liability for false advertising even if we would not hold people morally responsible for identical misdescriptions in other contexts, such as when they expressed them blamelessly. As Seana Shiffrin has argued, such “expansion” of liability fosters a morally indispensable form of trust, whereby we trust corporate actors not only to behave in good faith but to get the facts right.11 It is also arguably less prohibitive for businesses to bear the cost of even their blameless errors, when those costs would otherwise be borne by individual human consumers, who suffer harm as a consequence.12

Arguments of this sort, familiar to other discussions of strict liability, concede that the practice involves treating these players - advertisers, in particular - differently from how morality would otherwise regard and respond to people, such as ordinary citizens speaking freely about something they cooked or built. My focus, in contrast, is on whether strict liability in these contexts is even a prima facie departure from the moral treatment of speech. The coming sections constitute the case for why, in my view, it is not.

II. Inviting Reliance

Is strict liability for false advertising a legal innovation, however justified, or does it simply follow from how ordinary morality would treat a party like Slapper? To make the case for the latter, one would need to provide the moral grounds to treat such corporate claims as “No pea ingredients” or “no caffeine” differently from that of an ordinary person talking about the food he just prepared or the house he owns (“no pets ever lived here,” he might say to a guest). As a first pass, we might consider the hypothesis that what morally distinguishes commercial claims of fact, like “No pea ingredients,” from ordinary assertions is that the former invite reliance. They are directed at audiences so that they will act on these claims, for example by purchasing, consuming or otherwise using the products described, possibly to their detriment. In contrast, it might be claimed, Jon’s claim about his homemade tea was made in response to a question; he did not seek to get Clara to rely on his claim but simply provided his assessment when prompted. He did not ask her to try or taste the tea, using its alleged non-allergicness as a ruse (as in “You must have this, it’s completely allergen-free!”). Instead, she asked him about it, and he tried to answer honestly and accurately, only then asking if she’d like some. It would have altered his meaning had he added to his answer the phrase, “count on it,” or “you can drink it safely,” phrases that invite reliance. Slapper and other companies, in contrast, are expressing factual claims precisely so that they will be acted upon.

The hypothesis, then, is that such commercial claims count as utterances Judith Thomson identifies as “giving one’s word,”13 which -if true- would justify strict liability. On Thomson’s account, when someone utters an assertion like “the building is empty,” inviting others to rely on it -by adding, say, “I give you my word,” or “That’s a promise,” or “Count on it!”- and the other accepts this invitation, as in saying “OK,” the other “acquires a claim” against the inviter, or speaker, to the truth of his assertion.14 The speaker, in other words, owes the audience that the utterance be accurate, the word-giving be true. And this is so regardless of whether she would be blameworthy in misspeaking. If a claim like “The building is empty, you can take it to the bank,” was based on a thorough search of the premises, double- and triple-checking, and it was a perfectly reasonable and careful conclusion on the data available, the speaker still may be held accountable if she proves mistaken, despite her best efforts and intentions. Word-giving, like promising -which Thomson treats as a sub-class of word-giving- is a case of moral strict liability. The world has to actually fit what one says (or promises) about it; efforts, care and intentions do not matter.

Is a claim like “Lead-free” or “sugar-free” a word-giving, in Thomson’s sense? Is our working hypothesis correct? Like word-givings, these utterances invite reliance. The company is describing its products as part of an effort to be taken at its word and transacted with accordingly. On the other hand, there are two mismatches, at least on the surface. First, a company’s labeling or packaging is not directed at any particular person or customer, who then performs what J.L. Austin calls “uptake,” accepting the invitation to rely, as Thomson describes it.15 And it is only when there is both an invitation to rely, and an uptake or acceptance of that invitation, that the audience acquires a claim to the truth of the word-giving, on Thomson’s account.

This lack of uptake may be enough to show that Clara, in our first example, does not have a claim against Slapper for the truth of its label. It does not owe her anything directly. But this type of relationship, involving what are known as “directed” duties and claims, is special.16 What Thomson describes is more than simply a case of one party, the word-giver, having some moral obligation or reason to do a certain thing -like speak accurately- for the benefit of another. The audience, rather, has a “claim” or a right to the word-giver’s doing so, on her account. This is a unique interpersonal dynamic, one in which, as Margaret Gilbert puts it, the audience member has special “standing to demand” performance, in this case accurate representation, from the speaker.17

The paradigm case of such a special, directed relationship is promissory obligation, as would arise if the company CEO told Clara, “I promise you I will make sure it is allergen free.” A promisee is not only someone for whom the promisor ought to do as he promised. Rather, he owes it to her directly; she has a claim or right to his doing so. If he breaks it, he does not just do a wrongful act of which she is the victim; he wrongs her.18 Perhaps, then, we can draw from Thomson the conclusion that absent this invitation-acceptance exchange, no such special relationship arises. The audience, like Clara, does not have any special standing to demand anything of the word-giver, like Slapper, in the sense of acquiring a “claim” to its performance.

But that still leaves us with the question of what can be said about word-givers, like Slapper, who invite others to rely on the truth of a claim, reasonably foreseeing that they will do so even if no particular others are solicited or targeted. Are they accountable to being accurate, even if it is not something they directly owe their audience, and even if their audience does not have a directed claim or standing with respect to it?

One point we may note, by way of answering the question, is that even without uptake, much that is morally significant remains. Consider a paradigm case of word-giving without uptake, namely the public announcement: a health inspector, say, announces in a televised or radio broadcast that “The romaine lettuce scare is over; there is no longer any danger. Take my word for it, the hazard is gone; romaine lettuce is safe again.” Here, too, we have an assertion on which the speaker invites the audience to rely, even if no particular audience member is invited directly and given a chance to accept or “uptake” this invitation. And as with other word-givers, he can reasonably foresee such reliance occurring, and he can reasonably foresee the harm he would cause those who relied upon him if he were mistaken. If we take Thomson’s invitation image more literally, we can imagine the speaker sending out a number of invitations randomly in the mail, or even leaving them in a town square for whomever would like. Although there may be nobody who picks up such an invitation that can, on that basis, claim some agreement or shared commitment between them, they may all claim to have been given an invitation that they reasonably acted upon. If that to which they were invited -a party, a romaine lettuce fare- proved not to exist, the one who issued it would have much to answer for, even if not necessarily owing a particular person something in light of what passed between them.19

Indeed, the only significant difference to the moral evaluation of the speaker, between the televised health inspector and a standard word-giver, is the absence of this special, relational feature Gilbert and others have noted for the latter class of cases. The inspector is otherwise making the same sort of assertion, taking and imposing the same risks on others, all of which he reasonably foresees as he acts. If strict moral liability is justified in the one case, then, there appears to be no morally significant difference that could justify relaxing it for the other.

Still, it is not yet established that strict liability is justified in either case, even if they are on a moral par. This brings us to a second possible mismatch between Thomson’s account and corporate claims: Thomson’s account applies to speech that clearly does not give rise to the kind of strict liability we have noted. It is, in other words, overbroad. That is because one can invite others to rely on some claim without becoming thereby accountable for accuracy.

Consider a volunteer alibi witness. Lara believes she saw Victor somewhere else while the crime was committed and feels obligated to come forward, testify and thereby exculpate him. She thus delivers an impassioned statement insisting with great conviction that she saw him at the fateful moment. In such a case, Lara is plainly inviting her audience, be they jurors, attorneys or the general public, to rely on her testimony, acting on it to free a possibly dangerous person. Yet if it turned out she was blamelessly deceived -Victor had sent a body double specifically to fool spectators while he committed the crime- she would not be responsible for the misstatement. Her blamelessness in arriving at her sincere opinion, and her solid evidentiary reasons for being so confident about it, would relieve her of any moral (or legal) responsibility for speaking falsely. She may feel bad, and perhaps some regret would be appropriate, but she would be blameless and owe nothing by way of redress or repair. And yet, she plainly invited someone to rely, possibly to their detriment. Indeed, many cases in which someone tries to convince another that he is right about something important to both of them amounts to an invitation to rely. Yet we do not hold interlocutors in high-stakes factual arguments to accuracy about their empirical claims.

There is, in short, something about the way the health inspector invited reliance, when she broadcast the claim about safety, that distinguishes her utterance from that of the alibi witness or anyone party to a heated argument. The issue is not whether she invited reliance, but how she did so, and in what way it differed from these other types of cases.

III. Guarantees

1. Introducing Guarantees

Taking stock, I have proposed that corporate product descriptions, like “lactose free,” resemble what Thomson calls “word-givings,” in that they invite others to rely on the truth of a claim. But they differ from interpersonal invitations to rely, where there is uptake and acceptance, and from argumentative invitations to rely, where one person tries to convince another to act on what she firmly believes, like the volunteer alibi witness. They are, instead, like a health inspector making a televised announcement about the safety of a place, product or resource. The question, then, is whether strict liability, moral if not necessarily legal, is justified with these sorts of speech acts, in a way that can, in turn, justify the legal treatment of advertising.

What, in other words, distinguishes the claims of a health inspector, or a public safety announcement, or even a news reporter declaring “the earthquake struck outside of Sacramento”? One key feature of such declarations is that they are authoritative. By “authoritative” I mean not that they are spoken by experts -though in these examples they are- but that they are presented as exercises of authority, whereby the speaker has removed for the audience any reason to evaluate the matter of themselves, directing them instead to act on the speaker’s say-so simply because the speaker has said so.20 They provide, in other words, exclusionary reasons not to consider the possibility of error or deceit; the audience need not entertain doubts or worry about whether the claim is accurate, as that task has been delegated to the authority of the speaker.21

A key function of authority, and a justification for having it structure and guide our normative roles and interactions, is that it enables people to act directly on the pronouncements of others without having to do the work of justifying or warranting those pronouncements, work they are often in no position to do. For example, I cannot read your mind or probe your inner emotions. So when you say, “I’m sad,” or “I’m curious,” I am licensed to take your word for it, to presume it is true simply because you say so.22 Of course, I may doubt it anyway, and may even have reason to be suspicious in certain instances. But as a default rule a person’s mere expression of their internal state suffices, not because it is sufficient evidence -it rarely is, especially with complete strangers- but because her expression of her own internal states or attitudes is authoritative. As is typical of authoritative declarations, they provide necessary information -we need to know what others are thinking or feeling- that we cannot know or even verify for ourselves, at least not without paralyzing difficulty; and they are uttered by speakers uniquely and perfectly positioned to provide that information. The same is true of the health inspector, once she’s inspected the lettuce.

Authoritative declarations of this sort I will call “guarantees.”23 More precisely, a guarantee is

  1. a claim of fact, that

  2. is presented as authoritative and

  3. invokes the authority of the speaker, either explicitly or implicitly, as one who knows the truth of the claim.

To clarify, (a) is meant to exclude expressions of opinion that are presented as such, as when Jon answers Clara to the effect that the tea he made has no pea ingredients. The answer may be confident, but it can be equally expressed as, “To the best of my knowledge, it has none.” It is, in other words, expressed as an opinion, perhaps an informed one. Guarantees, in contrast, express statements of fact, as known -rather than merely judged- by the speaker as such.

What is added by (b) is that the claim, in being “presented as authoritative,” directs the audience to exclude or discard reasons to suspect otherwise. In other words, it is to contrast such speech with argumentative debate or impassioned pleas, meant to convince the audience of some truth or other. Here, the audience is directed not even to consider the possibility of falsehood, as though the speaker has already taken care of all investigative requirements.

The third element (c ) is simply brought to exclude cases where the speaker has no standing to make the relevant authoritative claim of fact, as in saying, “I guarantee you there are planets somewhere in the universe with beings like us.”

It will be noticed that guarantees, so defined, are merely presented as authoritative, and as emanating from an authority, or at least someone with standing to make the relevant authoritative claim. The speaker could be mistaken or deceptive about this. In that case, the guarantee would be akin to fraud or at least fakery, even if unintended. This somewhat subjective understanding of guarantees is, however, necessary, for if authority is to serve its function of allowing people to take a directive or claim from another without investigating its merits, then its purposes would be frustrated if people were forced to investigate or doubt whether authoritative claims were themselves authentic, in the sense of being uttered by the right people in the right circumstances. The practice of authoritative speech would be undermined; indeed, it would not get off the ground. To take someone’s authority, then, is to take her as authoritative in light of how they or their statements are presented.

2. The Ethics of Guaranteeing

Are we responsible for the accuracy of guarantees, regardless of fault? Here is a case for an affirmative answer, one which I believe already reflects our practice regarding speech acts like “I promise you the water is safe,” or “Take my word for it, the explosive has been defused.” A guarantee, if there is reason to believe it is authoritative, gives audiences exclusionary reasons to discard the possibility of falsehood, at least until contrary reasons or data emerge. In that way, they are reasonably relieved of the responsibility they ordinarily retain in verifying things for themselves. That, again, is part of the point of authority.

This feature of guarantees -that it gives the audience exclusionary reasons to discard alternatives- is known by the guaranteer. In issuing a guarantee, then, he knowingly reduces the latter’s defenses against being misled. As a result, he takes on for himself a parallel risk of being held accountable for any harm she risks inflicting on his audience. The defenselessness he imposes, on others, justifies imposing on him a corresponding defenseless against being held responsible or liable.24 Notice how the same cannot be said about ordinary assertions, like Jon’s reported recollection that he did not use a pea ingredient in preparing the tea (though he did innocently use what turned out to be a pea derivate). The assertion is not presented as authoritative; it simply comes as one person’s assessment, in response to an inquiry he did not solicit. The audience retains the ability to evaluate its epistemic merit and evidentiary bona fides.

True, many convincing assertions are likely to quell doubt, as in the compellingly sincere and impassioned alibi witness. In this way, they cause a similar state of reliance. Even there, however, we take the claim for the data that it is - a person’s confident recollection of what they believe they saw. And we can therefore evaluate it accordingly. In contrast, the guarantee quiets such evaluation before it starts; we simply take the word of the guaranteer and do not evaluate for ourselves, becoming thereby specially vulnerable. Hence the guaranteer’s special responsibility, and liability, in the event of misstatement.

All this is not to deny that some other invitations to rely may generate the same level of strict liability or responsibility. A physician’s assertion that a drug is curative might play that role in light of the great imbalance in expertise and vulnerability. Even there, the ethical grounds for heightened responsibility would be similar if not identical: the speaker imposes not only a false belief but a somewhat diminished state of alertness to its potential falsity, if only because the listener is not in a position to evaluate things for herself. With guarantees, though, the case is still more extreme: the speaker knowingly directs the audience to become altogether defenseless, by reducing to nothing their alertness to the possibility of falsity or error, at least if the audience takes their word, as the speaker’s authority licenses them to do. It is, after all, not presented as an assessment or evaluation to be judged on its perceptual and inferential grounds, but as a fact, pure and simple, already known by the speaker.

A guarantee, in other words, is a special subclass of invitations to rely, in that the invitation is based on the authority of the speaker and the utterance, such that the reliance amounts to a complete suspension of scrutiny and reassessment, or even an openness to it, at least if the audience does what it is invited to do (and may reasonably do).

Still, holding guaranteers responsible for accuracy can be harsh, particularly when the speaker had every reason to be certain and to provide an authoritative guarantee. The health inspector, for example, must eventually lift the warning against a widely available and economically important resource, particularly when many people depend upon it for livelihood or health. He cannot withhold the guarantee altogether on the grounds that some contrary evidence may arise; indeed, the whole point of having authoritative declarations is to allow people to get closure on certain matters and shift their efforts and energies elsewhere. So the need to provide a guarantee may prompt him to assess the situation, and an extremely thorough investigation might justify issuing it decisively. In short, his guaranteeing as he did would be above criticism. And it would remain above criticism even if, despite his best efforts, he proved wrong.

This sort of harsh strictness can, however, seem more palatable when we recognize that guaranteeing something in particular, even where there is pressure to do so, remains a matter of choice, and one that a reasonable person can make only when conscious of the vulnerability it is likely to impose upon others, who are in no position to investigate or even second-guess it. In taking the action of guaranteeing, the inspector knowingly imposed a special risk on others, and so knowingly assumed a risk of his own liability.25 It is like someone who promises for reasons that are completely justified, and for similarly blameless reasons is forced to break the promise. While his actions may be above criticism, he has nevertheless wronged the promisee, at least in standard cases of blameless promise-breaking, and he knew this was a possible consequence when he first set out to make a promise. In this way guaranteeing belongs to a category of actions like promising and taking loans, whereby one assumes the risk of being held responsible if things do not turn out a certain way, regardless of fault, just as one is imposing a special risk on others.

3. Corporate Product Descriptions are Guarantees

Having argued for holding guaranteers responsible as to their accuracy, regardless of fault, I now wish to make the case that statements like “caffeine free” or “allergen free” on product packaging, or “No Sweetener” in a commercial, count as guarantees.

Recall that guarantees are (a) claims of fact, (b) presented as authoritative, that (c) invoke the authority of the speaker with respect to the claim. Returning to the case of Slapper Ice Tea, whose packaging proclaims that the product contains “no peas, beans, or soy” ingredients, we can see that the elements are met. It is a claim of fact - it would have a different meaning if it were appended with the words, “as far as we know,” or “given the information available to our experts here.” It’s not that those words would clarify or expand upon the claim; they would, rather, alter the type of claim it was. Second, it is presented as authoritative: the claim comes as a directive to take it as given, to discard alternatives, to close the matter. In this way, it provides exclusionary reasons to avoid evaluation of the epistemic or evidentiary merit. It is not offered to persuade or engage the audience’s reasoning, but to close it down by the declaration of one in a position to foreclose all doubt. In issuing this type of declaration, then, the corporation invokes its authority as the manufacturer, who knows, beyond doubt, what ingredients are involved; element (c), in other words, is met.

Guarantees, as I have tried to demonstrate, are reasonably held to accuracy, regardless of fault. So if such corporate advertisement counts as guaranteeing, as I am claiming they do, that would justify the strict liability standard to which they are held. It would also explain the legal tolerance of puffery, those transparently evaluative claims praising one’s own products.26 An aspirin manufacturer’s claim, along the lines of “cures pain like no other,” would not reasonably be interpreted as a claim of fact, rather than an (obviously biased) evaluation, nor would it be taken to invoke the speaker’s authority to foreclose any need to verify it.27 In short, corporate puffery would not meet the criteria for guarantees set out above, and that could justify the current legal practice, at least in the U.S., of relaxing the speaker’s responsibility for accuracy. In contrast, a claim like “pain relief in 30 minutes” would be a guarantee, and therefore subject to responsibility for accuracy. The legal strict liability in such cases would, then, be a straightforward application of the ethical treatment of guarantees, at least on the arguments set out above.

IV. Objections

It may be noticed that the preceding argument for strict liability in advertising depends on a claim about the power of authoritative guarantees, namely that they remove, or at least reasonably can remove, the defenses a consumer may ordinarily have against false belief. One defense, in particular, was mentioned: ordinarily, even the most convincing assertion comes to us presented as an opinion, a judgement or evaluation reached by the speaker on the basis of evidence or perceptual skill or both. In short, it is itself a kind of data, which we remain free to scrutinize for its evidentiary value. But an authoritative guarantee - “the dynamite is defused” or, more to the present point, “safer than nicotine” - warrants closure. We may discard the possibility of falsehood. Or so I have argued here. It may be worried, however, that to credit a declaration with this sort of power relieves audiences of any responsibility for assessing things for themselves; indeed, it appears to deny them their indispensable role as free thinkers and deliberators.28 This is both morally problematic, as Robert Paul Wolff has argued,29 and plainly inaccurate; we often can and do remain able to reassess even the most authoritative guarantee.

Nothing here should be read to disagree with the claim that people may reasonably reassess guarantees if the thought occurs to them. But just as autonomy and critical scrutiny are indispensable to moral, social and political life, so is the ability to defer to the authority of others, at least in some contexts. Indeed, that ability underwrites the entire system of communication on which these forms of life and thought depend: as I noted above, we routinely and reasonably take people at their word when they express their internal states, such as “That hurts,” or “I’m thirsty,” or “I agree.” These utterances are considered sufficient to move on and close the matter for scrutiny, even if doing so is not demanded. People are, simply put, authorities about the inner workings of their minds, and their expressions of this knowledge is likewise authoritative, functionally equivalent to guarantees. Treating them this way does not only pose no noticeable threat to our autonomy; rather, as Kant noticed, it is necessary for interpersonal communication, cooperation and social life.30 Once we allow, drawing on such examples, that authoritative declarations have their place and do not undermine autonomy and critical scrutiny, we can note that the type of guarantees at work in commercial claims like “safe to drink” function in a similar way, enabling us to close scrutiny about something for which we need, at any rate, to defer to others.

Of course, the argument just now rehearsed may itself be vulnerable to what has been described, in other contexts, as a “wrong kind of reason” objection.31 In a nutshell: the practical or moral value of a belief is not an acceptable basis for forming it. We cannot believe a claim like “Lactose free” because of how well we would be served by taking corporations at their word at least sometimes. There may be great value, even urgent necessity, in having authoritative guarantees, but that is not an epistemic reasons to believe them, and no other such reason should be admitted.32

In response, it may be easy to miss that what has been said about authoritative guarantees is not yet an argument for believing them. Nothing here, in fact, supports believing that a corporate claim like “safer than cigarettes” is actually true. Rather, I have treated guarantees here as a sub-class of invitations to rely. They direct listeners to act on the embedded claims and discard the possibility of falsehood. And, for reasons I laid out, it is reasonable -and so epistemically (and otherwise) permissible- to do so, all else equal. But that is not at all to say it is required or even advisable to do so in any particular instance, much less to suggest we ought to believe guarantees if we do not already. The only implication of the foregoing analysis is that if we are told, by a product label, for example, some fact like “wildcaught,” we may act on it without worrying about whether it is false. That, however, is compatible with recognizing that we should only do so if we do not already have reasons to doubt the claim, or to be risk-averse about it and so refuse to avail ourselves of the epistemic or practical option that authoritative declarations provide.

Of course, it may be worried that not all corporate claims about products are invitations in any natural sense of the word. While Slapper’s claim that its tea was allergen free arguably counted as a kind of selling point, designed to entice allergic consumers (whereas Jon was simply answering a question), other corporate claims are prompted by third parties.33 Consider the listing of calories. If it came in response to a statute that required publicizing such information, would it still count as an invitation to rely? Could it be an invitation at all, if it was initiated by the statute and not the company on its own? If not, that might challenge the idea that a corporation that releases such information is assuming the risk of liability that comes with guaranteeing.

In response, it is worth noting that all guarantees are issued for some reason or other, some interest served not only by the audience but often by the speaker. The health inspector must make her announcement by a certain date, lest producers of the relevant product suffer irreparably; the bomb squad has to declare the device safe before it can move on to other business; a vulnerable person may solicit a guarantee because anything less would fail to quiet his concerns (“I know you believe it, but I need a guarantee”). If we exclude from the class of guarantees, or authoritative declarations of fact, all utterances prompted by others in one way or another, we risk losing the category altogether. How, then, do such “prompted” or incentivized guarantees still differ from the assertion of Jon, who responds to Clara’s question by asserting that his drink is allergen-free? A key difference is that he does not intend to guarantee. He means only to assert his assessment or recollection. He is not seeking to induce any reliance by answering, and he is not seeking to be taken at his word simply because he is saying so. If, however, both elements were present -if he answered, “Yes, there is no pea ingredient, I guarantee it,” or “Yes... Bank on it!”- there would be no reason to disqualify it as a guarantee, even though he was in effect required to answer by her question. He still had the option of not issuing the guarantee and chose to do so anyway, at least on the revision of his answer just now considered.

Similarly, a corporation required to list some information -chemical contents, for example- on its packaging or promotional material still has the choice of not releasing the product and thereby avoiding any declaration about the matter. In releasing it anyway, it still seeks to have would-be customers read the label and take its claims, including the legally solicited ingredients, as authoritative, taking the corporation’s word and acting on it. This point parallels the difference between a coerced promise and one that is merely highly incentivized, as when the promisor desperately seeks a loan from the promisee. The latter is still a promise, even if it is compelled by the lender, as in “You must promise me you’ll pay me back next month.” The promisor in such cases still has the option of withholding the promise, despite the costs. The same is true of advertisers compelled by law or some other outside demand to describe their products. They still choose to issue the authoritative statement as such and invite reliance on it, which is to say, they choose to guarantee, when they could have spoken less authoritatively, and are for that reason held responsible, and liable, for any inaccuracy.

V. References

Austin J. L., How to do Things with Words (Oxford University Press 1962). [ Links ]

Barnhill, Anne et al. (eds.), The Oxford Handbook of Food Ethics (Oxford University Press 2018). [ Links ]

Gilbert, Margaret, Rights and Demands: A Foundational Inquiry (Oxford University Press 2018). [ Links ]

Kant, Immanuel, Critique of Practical Reason and Other Writings in Moral Philosophy (Lewis White Beck ed and tr, University of Chicago Press 1949). [ Links ]

Lewis, David, ‘The Punishment that Leaves Something to Chance’ Philosophy and Public Affairs 18 (1989). [ Links ]

May, Cabulea, Directed Duties, Philosophy Compass (2015). [ Links ]

Migotti, Mark, “All Kinds of Promises,” Ethics 114 (2003). [ Links ]

Owens, David, Shaping the Normative Landscape (Oxford University Press, 2012). [ Links ]

Raz, Joseph, Practical Reason and Norms (1990). [ Links ]

Shiffrin Seana Valentine, Speech Matters (Princeton University Press: 2014). [ Links ]

Sreenivasan Gopal, ‘Duties and their Direction’ (2010) 120 Ethics 3. [ Links ]

Thomson, Judith Jarvis, ‘Giving One’s Word’ in Thomson, The Realm of Rights (Harvard University Press 1990). [ Links ]

Waldron, Jeremy, “Moments of Carelessness and Massive Loss,” in David G. Owen, Philosophical Foundations of Tort Law (Oxford University Press 1995) . [ Links ]

Wolff Robert Paul, In Defense of Anarchism (1970). [ Links ]

Wolff Robert Paul, In Defense of Anarchism (University of California Press 1970, 1998). [ Links ]

*For invaluable feedback on earlier versions, I thank Scott Altman, Brian Bix, Barbara Herman, Greg Keating, Jonathan Quong, Nomi Stolzenberg, Wil Waluchow, Gary Watson and especially Seana Shiffrin and Juan Vega, along with audiences at the workshops they led at the University of California, Los Angeles and the National Autonomous University of Mexico (UNAM), respectively.

2This is no longer universally true; other protein sources can be used to synthesize aspartame.

3The Lanham Act, 15 U.S.C. §§ 1051 et seq. At one point, it might have pleaded “contribution,” citing the supplier’s role in the affair, assuming it somehow found out. In re Masters Mates & Pilots Pension Plan, 957 F.2d 1020, 1028 (2d Cir. 1992). But as a false advertiser it would fall under the Lanham Act, for which courts have resisted such arguments. Getty Petroleum Corp. v. Island Transportation Corp., 862 F.2d 10 (2d Cir. 1988). The legal prohibition on false advertising, however, does not necessarily give rise to a tort claim on Clara’s part.

4A most compelling argument of this sort, involving the type of trust that such liability enables, appears in Seana Shiffrin, ‘Deceptive Advertising and Taking Responsibility for Others’ in Anne Barnhill, Mark Budolfson, and Tyler Dogget (eds), The Oxford Handbook of Food Ethics (Oxford University Press 2018) 470-93.

515 USC § 1151.

615 USC § 1125 (a)(1)(b).

715 USC § 1125 (b).

821 U.S.C. § 301 et seq.

9See, e.g., Vulcan Metals Co. v. Simmons Mfg. Co., 248 F. 853 (2d Cir. 1918) (holding that claims about the quality of a product, like that it would last a lifetime and perform perfectly, unlike claims about its manufacturing and sale history, are not actionable for false advertising).

10General Health Law, Article 25 of the Products and Services Health Control Regulations, the Mexican Official Standard “NOM-051-SSA1/SCFI-2010, General labeling specifications for prepackaged food and beverages-Commercial and health information” (NOM-051s)

11Seana Shiffrin, ‘Deceptive Advertising and Taking Responsibility for Others’ in A. Barnhill et al. (eds), The Oxford Handbook of Food Ethics (Oxford University Press 2018), especially pp. 471-82.

12For a discussion of the strict liability aspect of enterprise liability and its policy rationales, see Gregory Keating, ‘The Theory of Enterprise Liability and Common Law Strict Liability’ (2001) 54 Vanderbilt Law Review 1285-1337.

13 Judith Jarvis Thomson, ‘Giving One’s Word’ in Thomson, The Realm of Rights (Harvard University Press 1990) 294-321.

14Ibid 302.

15 J. L. Austin, How to do Things with Words (Oxford University Press 1962) 139.

16For a general explication, see, e.g., Gopal Sreenivasan, ‘Duties and their Direction’ (2010) Ethics 120: 3, pp. 465-94. See also Simon Cabulea May, ‘Directed Duties’ (2015) Philosophy Compass.

17For a full account of this type of standing, see Margaret Gilbert, Rights and Demands: a Foundational Inquiry (Oxford University Press 2018), especially 58-59.

18For an analysis of this distinction, see David Owens, Shaping the Normative Landscape (Oxford University Press 2012) 45-67.

19For Gilbert, standing-to-demand can only arise from a joint commitment between the party, so that when A breaks a promise to B, B appropriately regards the promise as “our” thing, something A and B share and which neither can rescind unilaterally. B’s keeping the promise is in some sense A’s. See Gilbert (n 16) 169.

20This is an essential feature of legal authority, for example; we obey the law because it’s the law, not because of its content. Robert Paul Wolff, In Defense of Anarchism (1970)5-10.

21For the canonical discussion of such exclusionary reasons, see Joseph Raz, Practical Reason and Norms (1990) 55-58.

22See Seana Valentine Shiffrin, Speech Matters (Princeton University Press 2014) 11.

23This is not to deny that the term “guarantee” has been used differently, especially in legal contexts, often to involve, in some way, the conditional promise to redress if the assertion proves false. Others have understood “warranty” along similar lines. Mark Migotti, ‘All Kinds of Promises’ (2003) Ethics 114, 78-79.

24This principle —imposing a risk on others justifies, or is tantamount to, assuming a parallel risk of accountability in the event the first risk materializes— has been invoked famously as a defense of imposing liability for all the harm that results from even minor negligence. See Jeremy Waldron, ‘Moments of Carelessness and Massive Loss’ in David G. Owen, Philosophical Foundations of Tort Law (Oxford University Press 1995) 401-405 (discussing David Lewis, ‘The Punishment that Leaves Something to Chance’, (1989) Philosophy and Public Affairs 18 53.

25See Waldron, supra.

26One reason is that

27For example, a federal Court of Appeals for the 8th Circuit concluded that the phrase “America’s favorite pasta” is, despite its descriptive form, “not a statement of fact.” American Italian Pasta Company v. New World Pasta Company, 371 F. 3d 387 (8th Cir. 2004) at 391.

28Thanks to Brian Bix for raising this objection.

29Wolff acknowledges the difficulty of knowing what an authoritative commander or guaranteer might now; “Nevertheless, so long as we recognize our responsibility for our actions, and acknowledge the power of reason within us, we must acknowledge as well the continuing obligation to make ourselves the authors of such commands as we may obey”. Robert Paul Wolff, In Defense of Anarchism (University of California Press 1970, 1998) 17.

30I take the phrase from Pamela Hieronymi, ‘The Wrong Kind of Reason’ (2005) 102 The Journal of Philosophy 437-457.

31Actually, this last point is no longer orthodox among epistemologists, but I will ignore the controversy for now, as I can respond even to the most extreme position that denies any grounds for belief other than evidence.

32As perhaps first observed in Immanuel Kant, ‘On a Supposed Right to Lie from Altruistic Motives’ in Immanuel Kant, Critique of Practical Reason and Other Writings in Moral Philosophy (Lewis White Beck ed and tr, University of Chicago Press 1949) 347-48.

33Thanks to Seana Shiffrin for bringing this concern to my attention.

Received: January 20, 2020; Accepted: August 06, 2020

Creative Commons License This is an open-access article distributed under the terms of the Creative Commons Attribution License