Visual argumentation in the Hungarian Competition Authority's proceedings1


Advertisements are a shared subject of inquiry for media theory and argumentation the-ory. Commercial interests provide a prime field for observing innovative persuasion techniques. Marketers utilize verbal tools and visuality; these tools are usually analyzed in rhetorical terms and with good reason, for the persuasive power of advertisements is mainly rhetorical. Moreover, one might even go on to say that this is the only the kind of analysis available, since we cannot express arguments by visual means. However, informal logicians have claimed that visual arguments are not only possible but actually exist and can be analyzed and evaluated in roughly the same way as verbal arguments. In this paper we will argue that they are right. In particular, we will explore in some detail how visual arguments can be reconstructed and point out the similarities to and the differences from the reconstruction of verbal arguments. We will then substantiate these claims by providing a complete reconstruction of the visual (strictly speaking, multi¬modal) argument given by Unilever for the superiority of its product, Dove Intensive Cream, in a famous and controversial2  commercial involving the “tulip test”.

We will start with a brief description of the informal logic tradition and explain how it makes room for visual arguments. Then, relying on this understanding of visual arguments, we are going to explain what steps the reconstruction of visual arguments involves. Finally, we will use the Dove commercial to demonstrate how these steps look like in practice.

The Informal Logic Tradition

In the late 1970s a group of philosophers started to develop “non-formal standards, procedures of analysis, interpretations, evaluation, critique and construction of argumentation in everyday language”.3 Their main motivation was that formal logic is  rather hard to apply to everyday arguments. Everyday arguments – like the student’s argument for deserving a better grade, the husband’s argument for getting a new car – are never ex-plicitly formulated as deductive arguments and trying to put them in deductive form requires the addition of further premises. These additional premises, however, often seem arbitrary in the sense that there is little justification for supposing that the arguer would accept them. Indeed, these additional premises would often be obviously false. So in-formal logicians jettisoned the idea of deductive validity together with the argument forms which may be assessed in terms of deductive validity. The new understanding of argument structure and validity they developed has made it possible to raise the question whether visual messages can constitute arguments. The majority of theoreticians has answered this question affirmatively.4 

The Idea of Visual Argument

From the perspective of formal logic the idea of visual argument looks odd to say the least: premises and conclusions ar sentences, but pictures are not made up of sentences. But O’Keefe has suggested a broader conception of argument which is more hospitable to visual arguments. On his understanding arguments involve “a linguistically explicable claim and one or more linguistically explicable reasons”.5 This implies that arguments do not necessarily have to be linguistic, they only have to be linguistically explicable. Visual contents are certainly linguistically explicable, since we can describe in words what pic-tures show. To put it differently, what matters for arguments is propositional content, and propositions can also be expressed by visual means. This conception of argument makes theoretical room for visual arguments. Informal logicians then went on to argue that some pictures described in the way we usually describe pictures actually constitute arguments. Even though these arguments are rarely complete in the sense of explicitly containing the claim and all the reasons, verbal arguments are also often incomplete, for the simple reason that what the recipient of the message knows or can easily figure out does not have to be explicitly stated.6

The Reconstruction of Visual Arguments

So the only important difference between visual and verbal arguments is that the claim and reasons making up a verbal argument are linguistic, whereas those making up visual arguments are at least partly merely linguistically explicable. Verbal arguments thus consist of a linguistically formulated claim, i.e. conclusion and one or more linguistically formulated reasons, i.e. a single set or multiple sets of premises, whereas in visual argu-ments at least some of the premises or the conclusion is not expressed in linguistic form. In the case of a simple argument relying on a single reason the picture is this (Table 1).   
Verbal argument Argument Visual argument
linguistic Premises
  linguistically explicable
Table 1

The question we have to address now is how this difference shows up in the reconstruction of visual arguments. What informal logicians mean by reconstruction is a fully explicit and transparent statement of the argument, which contains all elements necessary for its evaluation. So reconstruction involves more than a lay understanding of the argument – it is not a skill which everyone possesses but a learned art drawing on technical concepts. The reconstruction of an argument consists of the following elements:
  1. Identifying the conclusion.
  2. Identifying the premises.
  3. Rephrasing the sentences.
  4. Making implicit elements explicit.
  5. Building up the structure of the argument. 
These should not be conceived as consecutive steps of reconstruction, because reconstruction, which is a sophisticated process of understanding, like all other processes of understanding, moves in a hermeneutic circle. It is by identifying the conclusion that we may select the parts of the text which function as premises and set them apart from other parts, like explanations, incidental remarks, purely rhetorical elements, etc. But it is only by identifying the premises that we can understand exactly what conclusion the author of the text is arguing for. These two elements are present even in the lay under-standing of arguments. However, a reconstruction involves more. First of all, the possible ambiguities of the text need to be resolved. The terminology must be unified (e.g. in the student’s argument for a better grade which involves both the terms “unfair” and “unjust” we may have to substitute one for the other depending on how the argument goes). It is changes like these which the term ‘rephrasing the sentences’ signifies. In addition, the implicit elements must be made explicit otherwise the relevance or failure of relevance of the premise cannot be assessed. (E.g. the student’s showing his detailed notes of the readings is relevant only because this demonstrates that he has studied a lot – to which the teacher may respond that it is not the amount of studying which is relevant for the grade but whether the material has been learned.) When all the premises and the conclusion have been layed out, it needs to be spelled out how they are connected, how the premises are supposed to support the conclusion. (E.g. if the student explains that he has studied a lot and he has only one point missing for the passing grade, is he advancing two separate reasons for his claim of deserving a better grade, or is he arguing that it is in light of his hard work that the missing point should be ignored?)

When it comes to visual arguments, we cannot simply identify the conclusion and the premises, since we do not have a linguistic text in which we can isolate them. What we need instead is their linguistic formulation. Continuing down the list, pictures and films, being non-linguistic, are free of the occasional linguistic ambiguities and inaccuracies, and this renders rephrasing sentences superfluous; if there are not any sentences, there is nothing to rephrase. The rest of the elements remain the same. Visual arguments may contain implicit premises just as verbal arguments do. It is worth drawing attention to the distinction between linguistic formulation and addition of implicit elements. Linguistic formulation transforms the visual argument into a verbal one, whereas making the implicit explicit consists in providing what is missing. Linguistic formulation consists in changing the modality of content, making the implicit explicit amounts to enriching the content (Table 2).

Verbal argument
Visual argument
Identifying the conclusion.
Linguistic formulation of conclusion.
Identifying the premises. 2.
Linguistic formulation of premises.
Rephrasing the sentences. 3.
Making implicit elements explicit. 4.
Making implicit elements explicit.
Building up the structure.
Building up the structure.

Table 2

It seems, then, that the reconstruction of visual argumentation follows broadly the same method as the reconstruction of verbal arguments. It is worth pointing out that there are arguments termed “multimodal”,7 which feature both verbal and visual elements. Indeed, commercials making use of visual argumentation are typically multimodal, and the Dove commercial to be analyzed is no exception.

Given this picture of the reconstruction of visual arguments it is clear that the evaluation of visual arguments (e.g. identifying unacceptable premises or fallacies) is also fairly similar to that of verbal arguments. The reason is that reconstruction amounts to a verbal representation of the argument, and the verbal representation of an argument is a verbal argument, and as such, all the usual methods of assessment of verbal arguments are appropriate.

Argument Schemes

Before offering a reconstruction of the Dove commercial we need to say a few words about the apparatus to be deployed. Informal logicians have suggested various conceptual devices to replace the apparatus of the logical connectives geared to capturing deductive structure. The one we will make use of, the apparatus of argumentation schemes, bears some similarity to the logical forms of deductive logicians. A valid logical form is an abstract structure made up of linguistic elements characterized solely in terms of their identity or non-identity and logical connectives such that if it is filled in with linguistic elements in a way which renders the premises true, then the conclusion is also necessarily rendered true. An argumentation scheme is also an abstract structure which can be filled in with various linguistic elements. But filling it in with linguistic elements which make the premises true does not necessarily make the conclusion true. It makes the conclusion only presumptively true, meaning that we may accept the conclusion as true as long as we are not given a stronger reason against the conclusion or a consideration that undermines the argument. Argument schemes filled in with true premises thus supply only defeasible justification for the conclusion. These argumentation schemes are also constituted in parts by identical linguistic elements, but instead of logical connectives they involve non-logical expressions such as similarity, cause, sign. Here is a somewhat simplified example:

A is true in this situation.
A is a sign of B.
B is true in this situation.

Filling in “There is smoke over there” for A and “There is fire over there” for B, we get a cogent, even if not conclusive argument for B. The argument would be defeated if it turned out the smoke was generated by a high-powered smoke machine. This basic idea has been spelt out differently by different authors. Here we will be drawing on Walton, Reed, and Macagno’s argumentation schemes.8 

The Case Study

In 2006 Unilever started to air a commercial for a New Dove Intensive Cream.9 The commercial, intended to convince customers that the Dove product is a better moisturizer than Nivea’s market leading product, runs as follows.

A female hand touches first the Dove then the Nivea product, placed left and right, respectively (Figure 1/1). Then we are presented with two containers with the names of the two brands, in which the amount of cream appears to be the same. After that, the camera focuses on the containers (Figure 1/2–3–4). A dying tulip is placed first in the Dove cream (Figure 1/5), then in the Nivea product (Figure 1/6). The flowers are in bad shape, drooping in opposite directions; they obviously need water. The camera shows them from the side, which makes their miserable condition perfectly clear (Figure 1/7). At the 12th second of the commercial the tulips are left alone to give them time to absorb the creams (Figure 1/8). The changing light and the ticking of a clock suggests that time passes. The camera focuses on the tulip of the Nivea and we see that its condition does not visibly improve (Figure 1/9).

Figure 1

The camera zooms out and we see both tulips now. The passing of time is shown on a virtual stopwatch (Figure 2/10). After ten hours, the tulip left in the Dove product looks perfectly healthy, while Nivea’s tulip is still somewhat drooping - a humiliating defeat (Figure 2/11). The examiner chooses for the tulip treated with Dove (Figure 2/12). The abandoned tulip is left in the Nivea moisturizer (Figure 2/13). The tulip is retrieved from the left container and placed on the right beside the moisturizer. The text reads “New Dove Intensive Cream” and “Better moisturization, beautiful skin” (Figure 2/14–15). Notably, in the Hungarian version of the advertisement the slogan was “Better mois-turization and beautiful skin” (Figure 2).

Figure 2

The Reconstruction of Visual Arguments in the “Tulip Test”

Following the procedure outlined earlier, in reconstructing a visual argument we must start with the linguistic formulation of the conclusion and the premises. The former presents no difficulties: since this is a commercial for Dove, the conclusion should be something like “You should use Dove”. What about the premises? One clue is supplied by the text appearing at the end of the commercial, “Better moisturization, beautiful skin”. Having superior moisturizing effect and thus making the skin more beautiful is certainly a good reason for choosing Dove.

Notice, however, that it is at the very end of the commercial that this text appears, which suggests that it might be a conclusion deriving from what we saw before. So what did we see? We saw that Dove improves the condition of the drooping tulip much better than Nivea does. As we all know, flowers need water, so it is by supplying water, i.e. by moisturizing that Dove improves the condition of the tulip. So one premise leading to the conclusion presented in text (which, in turn, is a premise for the final conclusion that we should use Dove) is something like this: “Dove moisturizes the tulip better than Nivea does”.

The next question is how we move from this premise to the conclusion than Dove moisturizes the skin better. It is at this point that the idea of argumentation schemes can be invoked, as structures linking premises to conclusions. Since the commercial derives a conclusion about the skin from a premise about the tulip, it presumably relies on the idea that the two are similar. This suggests that it is an argument from analogy. This argumentation scheme is characterized by Walton, Reed, and Macagno as follows:

Argument from analogy:
  • Generally, case C1 is similar to case C2.
  • In case C1, A is true.
  • A is true in case C2.10 
In the present case C1 is the case of the tulip, C2 is the case of the skin, thus the analogical argument offered in the commercial is this:

Argument from analogy in this case:
  • The skin is similar to the tulip.
  • Dove moisturizes the tulip much better than Nivea does.
  • Dove moisturizes the skin better than Nivea does.

Notice that in identifying the two premises we perform different reconstructive operations. In the case of the second premise we merely put what we saw in the commercial in verbal form, which we called linguistic formulation. But the pictures do not show anything like the first premise. We find out about it by asking how the first premise might lead to the conclusion, and its specific form is identified with the help of an argumentation scheme. So what we do here is performing the reconstructive operation of making the implicit explicit.

We have already noted that the final conclusion of the commercial is that we should use Dove and that it is inferred from the premise that Dove moisturizes better and makes the skin more beautiful. But how exactly does the inference go? Moisturizers are supposed to make our skin more beautiful, which we think is a good thing. This suggests that the inference utilizes the argument scheme from positive consequences. This scheme is described by Walton, Reed, and Macagno in this way:

Argument from Positive Consequences:
  • If A is brought about, then good consequences will plausibly occur.
  • Therefore, A should be brought about.11

Variable A is in this case using Dove, and the good conseqences in question consist in having better moisturized and hence more beautiful skin. So the argument runs as fol¬lows:

Argument from Positive Consequences in this case:
  • If you use Dove, then it is plausible that your skin will be better moisturized and be more beautiful.   
  • Therefore, you should use Dove.

What remains is the final reconstructive operation, building up the structure of the argument. The argument from positive consequences takes us to the final conclusion of the commercial, and the role of the argument from analogy is to support the premise of the argument. However, the conclusion of the argument from analogy is not exactly the same as the premise of the argument from positive consequences, since the latter mentions beautiful skin (italicized above), which the former does not. This gap is filled by the textual element of the commercial, “Better moisturization, beautiful skin”, which can be construed in this context as saying that better moisturized skin is more beautiful. Construing it in this way involves the reconstructive operation characteristic only of verbal arguments, rephrasing the sentences.    

So the argument can be put together as follows (Table 3):

The skin is similar to the tulip.
  implicit premise
Dove moisturizes the tulip much better than Nivea does.    explicit visual premise
Dove moisturizes the skin better than Nivea does. from 1. and 2. by argumentation from analogy
Better moisturized skin is more beautiful.
  textual premise rephrased
If you use Dove, then it is plausible that your skin will be better moisturized and be more beautiful.
from 3. and 4.
Therefore, you should use Dove. 
  from 5. by argument from positive consequences

Table 3


We have argued first that the reconstruction of visual arguments follows by and large the same procedure as that of verbal ones. We have found only two differences. In place of the identification of premises and conclusions in verbal arguments we have the linguistic formulation of premises and conclusion. Also, in the case of visual arguments there is nothing corresponding to the reconstructive operation of rephrasing the premises. What is especially interesting and might even be surprising is the similarity that the operation of making the implicit explicit is also part of the reconstruction of visual arguments.

To see how the reconstruction works in practice we have provided a detailed re-construction of a commercial, pointing out how the theoretically motivated reconstructive transformations actually show up in practice. The reconstruction also allows to draw a more specific conclusion, namely that the apparatus of argumentation schemes can be applied to the reconstruction of visual arguments as well.

[1] Supported by the ÚNKP-16-3 New National Excellence Program of the Ministry of the Human Capacity. We would like to thank István Danka, János Tanács, and Réka Markovich for their invaluable contribution to the research. We would also like to thank the Hungarian Scientific Research Fund (OTKA) for the financial support provided by grant K 109456.

[2] The Hungarian Competion Authority initiated a proceeding against this advertisement and found it deceitful.

[3] F. H. van Eemeren et al., Handbook of Argumentation Theory, Amsterdam: John Benjamins, 2014, pp. 373–374. Eemeren is here referring to R. H. Johnson and J. A. Blair, “The Current State of Informal Logic”, Informal Logic 9 (1987), pp. 147–51.

[4] Birdsell’s, Groarke’s and Blair’s papers in the 1996 special issue of Argumentation and Advocacy are especially important.

[5]  Anthony Blair, “The Possibility and Actuality of Visual Arguments”, Argumentation and Advocacy, vol. 33, no. 1 (1996), p. 24.

[6] Opponents of the existence of visual arguments often claim that pictures are unsuitable for the expression of arguments because they are intrinsically ambiguous (David Fleming, “Can there be Visual Arguments?”, Argumentation & Advocacy, vol. 33, no. 1 [1996], p. 11). That is a serious concern which cannot be easily dismissed; nevertheless, we agree with Blair (op. cit., p. 24) that the difference between the verbal and the visual in this respect is merely a difference in degree. We trust that the reconstruction of the commercial below at least illustrates that this concern is unfounded.

[7] J. Anthony Blair, “Probative Norms for Multimodal Visual Arguments”, Argumentation, vol. 29, no. 2 (2015), pp. 217–233.

[8] D. N. Walton et al., Argumentation Schemes, Cambridge: Cambridge University Press, 2008.

[9] Unilever Germany – tulip test: http://www.tvspots.tv/video/42773/unilever-germany-tulip-test.

[10] D. N. Walton et al., op. cit., p. 315.

[11] Ibid., p. 332.



  • There are currently no refbacks.