Our best scientific result
In the past we have done more than 10 years research of political decision making. This research was published in 1996. But in 2020 Willem read a book about informal logic of Toulmin and realized that we didn´t do research on "political decision making" but on "argumentation about decisions". The idea was that our so called decision rules where rather implicit logical rules to derive conclusions in arguments about decisions. Then we started to study the literature on argumentation and this led to the best scientific result we have produced in our life. This research is now accepted for publication. We are very proud that we could do this work, being already 77 and 75 years old. It is of course based on the research we did 25 years ago.
What is the problem
Through time, philosophers have argued that one cannot deduce a proposal for action from arguments for or against the proposal because the consequences may belong to different dimensions, which are not comparable. They lack commensurability. This problem also has been mentioned in the context of deliberative argumentation.
Nevertheless people in general and politicians especially are making arguments for decisions. They also present their arguments to others, believing that they make convincing arguments. To detect why they think that their arguments for actions are convincing, we studied a random sample of documents of government meetings where politicians had to make decisions in complex situations for their country.
Through time, philosophers have argued that one cannot deduce a proposal for action from arguments for or against the proposal because the consequences may belong to different dimensions, which are not comparable. They lack commensurability. This problem also has been mentioned in the context of deliberative argumentation.
Nevertheless people in general and politicians especially are making arguments for decisions. They also present their arguments to others, believing that they make convincing arguments. To detect why they think that their arguments for actions are convincing, we studied a random sample of documents of government meetings where politicians had to make decisions in complex situations for their country.
An illustration
The documents showed that the politicians spoke about possible actions, their consequences, the probabilities of these consequences and their evaluations.
As an example we present an argument of the Governor-General of the Dutch East Indies which was at that time still a colony of the Netherlands. In 1945 freedom fighters had declared the independent Republic of Indonesia. The Dutch government ignored this declaration and wanted to restore its authority in that region. Below we present the literally translation of one of his telegram to the government :
If we were to do nothing, it is very improbable that the Republic will respect the
cease-fire order, but the chance is rather high that a new untenable situation will
be created. However, if we occupy the seat of the republican government it is highly
probable that a political reconstruction will be achieved while the risk of creating
a new untenable situation is very small
In this argument the Governor-General mentioned two possible actions with their consequences and the connected probabilities. With this argument, he tried to convince the government in the Netherlands of his preferred action, occupy the seat of the republican government. He thought that this was a convincing argument, The consequences are incomparable as expected in the literature. Nevertheless, the speaker at least hopes that he makes a convincing argument for the action he prefers.
The documents showed that the politicians spoke about possible actions, their consequences, the probabilities of these consequences and their evaluations.
As an example we present an argument of the Governor-General of the Dutch East Indies which was at that time still a colony of the Netherlands. In 1945 freedom fighters had declared the independent Republic of Indonesia. The Dutch government ignored this declaration and wanted to restore its authority in that region. Below we present the literally translation of one of his telegram to the government :
If we were to do nothing, it is very improbable that the Republic will respect the
cease-fire order, but the chance is rather high that a new untenable situation will
be created. However, if we occupy the seat of the republican government it is highly
probable that a political reconstruction will be achieved while the risk of creating
a new untenable situation is very small
In this argument the Governor-General mentioned two possible actions with their consequences and the connected probabilities. With this argument, he tried to convince the government in the Netherlands of his preferred action, occupy the seat of the republican government. He thought that this was a convincing argument, The consequences are incomparable as expected in the literature. Nevertheless, the speaker at least hopes that he makes a convincing argument for the action he prefers.
Our hypotheses
In general, also in this case, politicians do not indicate why their arguments should lead to their choice. Based on the information (nominal or ordinal) about probabilities and utilities of the consequences we distinguished 4 types of arguments:
In the example above the probabilities are ordinal, rank ordered, while the utilities are nominal , not rank ordered. We specified hypotheses for each type of arguments what the implicit argumentation rules, “warrants” called by Toulmin, could be to derive a conclusion. We specified 7 rules (see Gallhofer and Saris 1996)
In general, also in this case, politicians do not indicate why their arguments should lead to their choice. Based on the information (nominal or ordinal) about probabilities and utilities of the consequences we distinguished 4 types of arguments:
- Probabilities and utilities are nominal
- Probabilities are ordinal and the utilities nominal
- Probabilities are nominal but the utilities and ordinal
- The probabilities and utilities are ordinal
In the example above the probabilities are ordinal, rank ordered, while the utilities are nominal , not rank ordered. We specified hypotheses for each type of arguments what the implicit argumentation rules, “warrants” called by Toulmin, could be to derive a conclusion. We specified 7 rules (see Gallhofer and Saris 1996)
Nearly 100% good predictions
Then we checked whether these warrants could predict the choices of the politicians based on their arguments. Based on a random sample of historical arguments, we detected that the seven implicit informal logic rules, we had specified, nearly perfectly predicted the choices made. In the above presented argument the rule which leads to the conclusion is the "Risk Aversion rule" i.e. Choose the strategy with the highest probability of a positive result.
We also have shown that a random sample of the Dutch population made the same choices based on the same arguments and that they were able to indicate the warrants we had expected.
This suggests that we have detected the basic warrants by which people argue about decisions. This was a surprising result because each type of arguments could contain many arguments on which our rules could not be applied.
Then we checked whether these warrants could predict the choices of the politicians based on their arguments. Based on a random sample of historical arguments, we detected that the seven implicit informal logic rules, we had specified, nearly perfectly predicted the choices made. In the above presented argument the rule which leads to the conclusion is the "Risk Aversion rule" i.e. Choose the strategy with the highest probability of a positive result.
We also have shown that a random sample of the Dutch population made the same choices based on the same arguments and that they were able to indicate the warrants we had expected.
This suggests that we have detected the basic warrants by which people argue about decisions. This was a surprising result because each type of arguments could contain many arguments on which our rules could not be applied.
Conclusions
We drew the conclusion that politicians simplify their decision problems so much that simple warrants, which everybody knows, can be used to make a convincing argument. Therefore, these warrants are implicit rules which don´t have to be mentioned. When people talk with each other, they also don´t have to explain the grammar to understand each other. The same seems to be true for the arguments concerning choices: the warrants don´t require explanation because the speaker and the audience are intuitively aware of them.
Beware that these argumentation rules will not necessarily lead to good decisions. This is so, because the decsion makers ignore a lot of information which may be relevant for the decisions. As argumentation rules, however, these rules are generally accepted.
We drew the conclusion that politicians simplify their decision problems so much that simple warrants, which everybody knows, can be used to make a convincing argument. Therefore, these warrants are implicit rules which don´t have to be mentioned. When people talk with each other, they also don´t have to explain the grammar to understand each other. The same seems to be true for the arguments concerning choices: the warrants don´t require explanation because the speaker and the audience are intuitively aware of them.
Beware that these argumentation rules will not necessarily lead to good decisions. This is so, because the decsion makers ignore a lot of information which may be relevant for the decisions. As argumentation rules, however, these rules are generally accepted.