NEWS & ANALYSIS

The logical errors of Pierre de Vos

The DA’s Gareth van Onselen takes the UWC professor to task over his critique of Helen Zille

Introduction
It is very unusual for an academic (and certainly a law academic) to make as many logical errors as are committed in this particular piece and it is for that reason I have decided to use it to illustrate some of the (logical) shortcomings often involved in constructing a coherent argument. Because, if a senior academic can make these mistakes, anyone can, and it is useful to identify and explain them.

The piece in question is the third in an exchange between DA leader Helen Zille and Pierre de Vos, Professor of Constitutional Law at the University of the Western Cape (and part-time political commentator). And I would recommend reading it in full, before continuing with this article. It is not my intention to involve myself directly in that exchange, rather to simply analyse the logic used in the piece to illustrate some broader problems (for Zille's side of the argument, see here).

Nor is this response meant to be comprehensive, there are numerous other mistakes in De Vos's piece, which I have not identified here, as space does not allow. Here then are four logical errors from De Vos's response to Helen Zille.

Example 1: Inconsistently applied principles

The first example is fairly straightforward. De Vos contests Zille's claims that she has been vindicated by a city-appointed commission by arguing that: "it is true that a Commission headed by a senior advocate cleared Zille and the DA of wrongdoing. But the terms of reference of this Commission were decided on by Helen Zille herself, and she was thus cleared of what she knew beforehand she would be cleared. The terms of reference of the Erasmus Commission are much wider, and thus have the potential to uncover evidence of wrongdoing not covered by the Zille Commission."

Put another way, De Vos contends that Zille set up a commission with limited terms of reference because she knew - within those terms - she had done nothing wrong, and thus it would inevitably clear her, which is what it did. Thus, De Vos argues, she should not oppose a broader provincially-appointed commission as it might uncover wrongdoing missed or outside the mandate of the city-appointed commission.

Implicit in this line of reasoning is a certain amount of guilt on the part of Zille - she acted to limit a probe which could potentially undercover corruption (and is thus trying to prevent a more comprehensive one). He also implies that the city-appointed commission was set up to serve a political purpose - to clear Zille and, therefore, that it is essentially dishonest because, according to De Vos, it had limited terms of reference and thus did not constitute a full and proper investigation.

(Remember, although I contest De Vos's assertion, I am not providing a response to it here - that Helen Zille's office can do - I am merely analysing the use of logic.)

In principle, then, De Vos asserts that the city-appointed commission served a political purpose (to clear Zille) as opposed to a principled purpose (to establish the truth). But De Vos does not use this principle consistently, and this undermines the coherence of his argument.

If one accepts that a commission can be established to serve a political motive, as De Vos implies is the case with Zille, then one must also accept that a different commission set up by a different politician could also have been established to serve a political motive (albeit a different one).

To disprove that it was, one would then have to turn to the evidence. This might vary in strength and prove that one or both or neither were established with a particular political motive, but the principle would have been evenly applied and a comparison then done on the weight of the evidence.

As De Vos's argument stands, he applies the principle unevenly - to Zille and not to Rasool - dismisses the evidence presented by Zille as "laughable" and thus reveals the bias in his argument, which as a consequence is dramatically undermined.

(Incidentally, De Vos makes the same error with regard to the difference between advocate and a judge, implying that the advocate who oversaw the city-appointed commission could be used as part of a political strategy, but that the judge involved in the provincial commission could not.)

Example 2: Circular logic

De Vos's quote, above, also constitutes an example of circular logic, when he states: "The terms of reference of the Erasmus Commission are much wider, and thus have the potential to uncover evidence of wrongdoing not covered by the Zille Commission".

It goes without saying that the wider the terms of reference are, the greater the potential to uncover wrongdoing. Indeed, if one was to follow that logic all the way through to its conclusion you would arrive at the absurd suggestion that South Africa needs a national commission of inquiry to investigate every single element of government's policy and practice since it came to power fourteen years ago; which, as I say, is simply mad.

Thus, there is a point where any commission's terms of reference become too broad. Here, we have a conflict between two views: Zille argues that the city-appointed commission's terms of reference were adequate, De Vos argues they were not (and implies those of the Erasmus commission are).

Just like the first example above, any proper analysis or interrogation of these two conflicting viewpoints must be conducted using the available evidence. What were the city-appointed commission's terms of reference and what are the provincially-appointed commission's terms of reference; how do they differ and how do you explain that difference?

To simply dismiss the one on the grounds that the other is bigger (and therefore better) is dishonest and misleading. The scope of the terms of reference is not what defines the quality of the commission. It is the nature of those terms of reference that need to be interrogated (among other things - the impartiality of the chairperson and the purpose behind the commission).

Example 3: False assumptions

The third example is more general. There is false assumption underlying much of De Vos's argument - that the provincially-appointed Erasmus Commission is not a political tool. That may or may not be the case, but one cannot simple assume that, without presenting any evidence in support of that assumption. If one does merely accept it as given, the credibility of one's argument is dramatically undermined.

The following three sentences, for example, are all based on this assumption, and would all be rendered invalid were it to be proved that the Erasmus Commission was, in fact, established to serve a political purpose. Indeed, in each case, the opposite would be true:

·"The terms of reference of the Erasmus Commission are much wider, and thus have the potential to uncover evidence of wrongdoing not covered by the Zille Commission."

·"But by pulling out all the stops to try and prevent this broader investigation, reasonable people without an axe to grind will inevitably become suspicious and will begin to wonder why we should trust Zille."

·"A judge does not stop being a judge merely because he sits as the head of a commission of inquiry, and to suggest that he does is to engage in mental gymnastics of the highest order."


Once again, De Vos's assumption might or might not be true, but its validity can only be tested against the available evidence. If, however, it is false, then his entire thesis and many of the subsequent assumptions he makes are false. It is devious and misleading to base an argument on a false assumption because it misdirects debate.

Example 4: A false choice

A classic logical fallacy is the false choice. It works like this: the impression is created that there are only two options/realities/principles available and, by disproving the one, the other must therefore hold true. In truth, there might be several options available which simply haven't been mentioned.

De Vos commits this error repeatedly, in numerous different ways but, essentially, the false choice he offers is this:

He suggests that anyone choosing between Zille's opposition to the Erasmus Commission and the Erasmus Commission itself is, in effect, choosing either to support values such as "freedom of expression", the "independence of the judiciary" and the pursuit of "the truth", or oppose them. In other words, he implies that the Erasmus Commission and these values are one and the same and, if you oppose the one, you are, by default, opposing the other.

However, it is quite possible that the two are not one and the same and that by opposing the one you are not, in fact, opposing the other. History is cluttered with examples of commissions being used for political ends - certainly South Africa's history before and after apartheid is.

It is quite possible to embrace those values, but still oppose the Erasmus Commission. To imply otherwise is misleading and devious and presents a false choice which inevitably leads to a false conclusion on the part of the reader.

In his book ‘Thinking from A to Z", Nigel Warburton describes a false choice (he calls them false dichotomies) as follows: "false dichotomies can be set up accidentally or deliberately... When accidental they result from an inaccurate assessment of the available positions; when deliberately they are a form of sophistry."

Sophistry he describes like this:

"A display of cleverness which doesn't respect the principles of good reasoning but smuggles in unlikely conclusions under the cloak of a sham argument."

Conclusion
There is almost always a common characteristic to arguments that make logical errors - they represent the triumph of the subjective over the objective and personal opinion over considered reasoning.

There is a nothing wrong with an entirely subjective argument, indeed they often serve to stimulate debate, but they should be easily identifiable as such. It is when subjective opinion is couched in ostensibly objective (but actually flawed) reasoning, that logical fallacies are inevitably committed. If it is done well an argument can be very weak yet strangely compelling - which is misleading and, if it is done badly, it can be fundamentally dishonest.

The best way to avoid these sorts of logical errors is to fashion your argument around evidence. The stronger the evidence, the stronger an argument.

Gareth van Onselen is the DemocraticAlliance's director of special issues. This article was first published on the Democratic Alliance weblog, Inside politics, April 25 2008