Life is uncertain. We know this from our empirical experience. Consequently, Praxeology would be incomplete if it would not take uncertainty into consideration.
Uncertainty in Praxeology
Previously we explained that the concept of human action is based on choice. When we act we choose to behave one way and not another. We do so based on our expectation that this behaviour and not the other one will produce the goal that we seek, but there is no guarantee that we will be successful. When we choose one behaviour over another we assume that this is the correct behaviour. We do so because we don't know the future. If we would know the future then we would know exactly which behaviour we require and there would be no need to choose. We would simply move from one behaviour to the next one and we would do so automatically. As we do not do so, it becomes obvious that the idea of uncertainty is already embedded in the Praxeological concept of human action.
Classification of Uncertainties
In terms of dealing with uncertainties science provides us with probabilities. We can use probabilities to forecast the future (but not human behaviour) within a certain margin of error. In this area, the best we can do is to describe the most probable outcome based on a series of events and conditions. If we would try to apply probabilities to humans, we would find out rather quickly that they are useless because we can't describe the conditions and we don’t know the events. In other words, there are two things that are not-knowable. We can say that these constitute the classification of uncertainties:
Information about the current state of events (information about reality or nature that we don't have)
- The choice humans will take.
- These two uncertainties are inherent or built-in in any human action.
Reaction to Uncertainties
Now that we know that uncertainties are present in all human action, we need to look at how humans deal with it when acting purposely. They do one of three things. The difference between those three is the amount of available information.
1 - Gambling
This sits at the lowest informational level in terms of dealing with uncertainty. We have some information about the events we are dealing with (for example frequency) but nothing more. This information is patently insufficient to forecast each event. As we need to act to eliminate uneasiness, we do the next best thing. We rely on luck and as such the risk is very high.
For example, we can easily calculate the probability of a win we have for each number in roulette, but nothing more. We go to a casino and we gamble with this little info.
2 - Speculating
This is the next level up in the informational level. We control our decisions based on how we perceive the future. We do this as a one-time operation or as an ongoing operation. We make adjustments as the future evolves. We do collect more information and use more robust tools for the task.
In our previous roulette example, we would sit by a roulette for a few hours, collate data and use a calculator to estimate statistical parameters. We would then use these parameters to bet and as new numbers would be selected, we would further adjust our calculations. However, none of those calculations will be able to provide us with absolute certainty. We would still have to rely on luck, albeit less than in gambling.
3 - Engineering
This is the highest position up in the informational level. We can analyze the statistical characteristics of a phenomenon very accurately. We can understand which are the best or the poorest conditions to win. We can understand the mechanics of the process, the frequency and the rules. Through this knowledge we can control all the elements of the action and build its means. We can even add safety measures and hedging capabilities. The primary objective of any engineering process is to minimize uncertainty. The only thing that we cannot do with all this information is to predict the outcome of each independent event.
For example, the statistics of roulette are exceedingly well known and so are its mechanics. If casinos would let us, we could sit by a roulette for days, gather sufficient statistical information and describe that particular roulette table in terms of probabilities. We could manufacture a portable computer to take with us into the casino. We could even build an exact duplicate of it at home to test our models. We can also hedge our bets through gambling insurance. However, even with all this information and technology we would still be incapable of forecasting the outcome of each run.
We must also take into consideration that this classification is subjective. What may be gambling for one person may be engineering for the next.
The application of Engineering to people
Basic political philosophies supporting any type political theory specify or define the concept of "good". In almost all types of political theories the "good" of the people is specified through somebody else's will. In a democracy is through the will of the representatives. In a dictatorship is through the will of the leader and so on. As such, these actors execute human actions to try to bring about the goal of "the good of the people". But in so doing, they must face uncertainty and therefore they must choose a way to deal with uncertainty. All governments choose the Engineering method. They do so because it is the one requiring the least amount of luck hence providing maximum control. This is called "social engineering". These types of processes they all fail eventually. This is so because the human actions of individuals are directed by their personal and subjective goals. Social engineering methods attempt to replace those individual goals with the goals of the actors. In other words, people are forced to follow artificial goals. These artificial goals are typically pre-conceived or static. For example in dictatorships it is typically nationalism and opposition to a different political theory (communism or capitalism). In democracies the typical goals are socialism and the perpetuation of the democratic system. All these goals are one-size-fits-all while human action is one-person-one-action. They are both inherently and basically incompatible.
In addition, how these artificial goals are imposed, varies from political theory to political theory. They are all based on coercion but the logistics changes somewhat. In dictatorships the imposition is through indoctrination, brute force and fear. Behave or be beaten, arrested and killed. In democracies artificial goals are imposed through conditioning by state education, punishment and fear. Behave or be arrested and jailed or fined. Within those parameters people will find it in their own best interest to do as they are told and as such their human action will seem to align with the artificial goal of the actors. Most people do not hold political protests in dictatorships. Most people do pay taxes in democracies. For this reason social engineering seems to work, but this is only an illusion.
Eventually, people realize that their individual, personal and subjective goals do not coincide with the artificial goals of the actors. This is so because artificial goals do not remove uneasiness from people's lives. At that point, people evolve politically in a direction where their goals can be acted upon. As only in government-less Libertarianism the will of the people is specified by every person individually, political evolution happens towards that direction.
Forecasting the future
Uncertainty also plays a role in the forecasting of future events or behaviours. As we have seen in previous lessons, Praxeology rejects all quantifiable methods to do so. Uncertainty is yet another reason why quantifiable methods must be rejected. This is so because all methods must take into consideration uncertainty, and this includes how humans will act which is not quantifiable.
This is not to say that no forecasts can be issued. Praxeology does produce forecasts but they are qualitative in nature. If interest rates go up the number of new mortgages will decline, this much can be said (assuming no other conditions have changed). But what cannot be forecasted is the exact decline (for example as percentage) of new mortgages. This is so, because of the existence of Scales of Values (described in a previous lesson) which are qualitative. These scales provide us with a general direction toward which people will move which in turn enables Praxeology to make forecasts. The manner in which Praxeology uses these Scales of Values is through the laws of action. Praxeology can do so because the laws of action have been constructed in such a manner as to account for uncertainty, this is, to restrict the amount of information dealt with as not to be affected by uncertainty.
But people (Praxeologists or not) still issue numeric forecasts. How does Praxeology look at these forecasts? As entrepreneurial activity. This is, these forecasts are personal and subjective estimates of the future issued by people with their personal goal in mind. This quantitative estimate allows them to operate in markets, but they are not Praxeological forecasts.
For this very same reason, the fact that we can have scientific predictions has no necessary implications for the future in terms of human action. There are many companies selling quantitative evaluations of cars based on empirical (objective) evidence. If our goal is to buy the best car we can, then we will use this information because the forecast is accurate. On the other hand if our objective is to buy the cheapest car, this forecast is useless. However, in the big scheme of things, when trying to forecast human car purchasing habits, the only thing we can say with certainty is that as prices rise car quality will be of less importance because not too many people will be able to afford them and therefore sales of quality cars will drop. Even with all the scientific evidence of car quality, we cannot forecast what the drop is sales will be (in terms of percentage or sales).
Note: please see the Glossary if you are unfamiliar with certain words.