Everybody can very simply hear, see, smell, taste and feel many things. Throughout a whole a lifetime, we observe many thing through these senses. Generalizations are indeed something we usually make in order to use these observations. But here is the question, can we take the generalizations we make for true as sure as the particular observations we make. That is probably one of the most important questions about inductive logic and I think we are more likely to be mistaken when we generalize than the cases we observe one by one. But we are still in need of a fair judgment on inductive logic to say if this risk to fail totally makes our inductions invalid.
Inductive logic is one of the most important parts of our daily lives and it is a great helper for us to cope with world’s rules. Just think for a minute and you will find out that the words we use are results of inductive logic. When a small child sees a dog for the first time in his life, he needs to ask what it is to someone who, he thinks, would know, like his parent or an adult… Learning that it is called a “dog”, from then on, any four-legged, furred animal resembling the first one he saw is a “dog”.
He very simply generalizes that any four-legged, furred animal looking like the first one is a dog. This will be a useful piece of knowledge for him. He will use it deductively after on, saying that “All dogs are four-legged and furred, X is four-legged and furred, therefore X is a dog.” The generalization he made will be very helpful to him to face the complex organization of the life. This a very good and basic example of how inductions are important in human life. Without this kind of essential generalizations we make we wouldn’t be able to cope with indicating everything on their owns.
If we go on the example we are working on, a significant problem occurs. As the child generalized “the dog”, he might have problems on particular conditions. For instance, when he sees a wolf, it will also resemble the dog and it will mislead him to label the wolf as a dog. Here is the main POK of the generalizations that if the premises are not explicit, clear and sensitive enough, it would easily cause a hasty generalization. If he says that a four-legged, furred animal is a dog, so all four-legged, furred animals are dogs, he would be mistaken. Also, if he saw a dog having three legs due to a traffic accident, he would hesitate to label it as a dog. It is quite clear that we are quite fallible while making inductions.
Specific conditions that are to be investigated individually are more likely to lead us to the certain truth than the conclusions we make by observing them. It is very easy for us to comment on accurate quantitive and qualitive states which are the materials of a particular place and time. But we occasionally choose to make a general statement out of them. A very known one for me, for example, is about Carolina Kluft, a young Swedish heptathlete, who ,very usually by her fans, is said to be always getting the gold medal. At first it seems to be correct. As soon as she is categorized as an adult, she wins a gold metal at 2002 – Munich European Championships, 2003 – Paris World Championships, 2003
Word Indoor Championships… It is something agreed by many authorities that she is the most successful of the time, but here come a conflicting case to our argument, she wins the bronze medal in Budapest World Indoor Championships in Budapest 2004. Before then, everybody was sure to generalize that she was and would always be winning the gold medal. Indeed, it seemed to be like that but here we see the importance of thinking of exceptions before using inductive logic. A small disorder in her leg invalidated our generalization. If we were to use all this information individually, it was not a problem to state our results like “She won the gold in the competition A and B , but she won the bronze in competition C.” But here our claim was proved to have a fallacy because of an unexpected incident.
For the above example, you may say that it is my problem that I made a weak inductive argument and if I was just normally careful and rational, inductions were as reliable as deductions and my fallacy was just one of all those we made in daily life. You may seem to disproof my claim but here is an interesting example on even how “reliable authorities” are fallible in using generalizations. On 2nd January 1980, the Scientific American Magazine had some lines like this: “Last year, there haven’t been any improvements in car technology, so we can say that this invention has also completed its progression and evolution.” By only observing only one specific period of time they reached a conclusion, which is clearly wrong as we see far better cars than 1980 nowadays. Even if a most respected scientific magazine has fallacious inductive arguments, it should be very apparent to us that inductions increase the rate of fallacies in our arguments.
In addition to hasty generalizations and too few cases observed, there is another risk for us: ignorance of conflicting cases. In many cases which we take for “enough” misleads us to false conclusion as they are not purely accurate due to the fact that conflicting cases are unsought for or ignored. Here is an example of my own. In one of my TOK classes, I took the sentence “All black people have dark eyes” as a premise to one of my arguments. This sentence was a conclusion I reached using induction. I had seen many photos of black people having dark eyes and it was very normal for me to conclude this, that I was shocked to hear there was also blue-eyed black people. I was mistaken again because of the generalizations’ POK.
You may claim that my observations was the source of POK, not the format of inductive logic. But it would show that you lack the fact that even one exception can invalidate our argument and in any case we can face some “secret” exceptions which we are unable to observe. My argument seemed to be very strong for me , however, it was invalidated. Generalization again misleaded us to false conclusion which happens, as you can see from the examples, quite easily and usually. It is so hard to think of a “perfect” induction, even the strongest argument can be invalidated by tiny exceptions.
While discussing on inductions, it is nearly impossible to skip the famous “turkey” example of Bertrand Russell. The example is on an “inductivist turkey” . The turkey is brought to the farm and day by day it realizes that everyday it gets food at 9 a.m. It simply thinks that everyday at 9 a.m. , food will be received. At the end it was a very big disappointment when it came the Christmas Eve. Probably, it is the most popular and most defensive point of those who want to invalidate inductions completely. On the other hand, if we go on thinking of the example and, parallel to this , take our own example it will be clear that even if fallacious how important it is to make generalizations.
The first class on Monday , at 8.15 to 9.00 a.m. , is mathematics for 10F. It was on this way on the first week, then the second, then the third and it went on. So this sentence is a generalization implying that “Any class on Monday at 8.15 to 9.00 a.m. is mathematics for 10F. This argument was clearly disproved when it came 29th October when, due to the ceremonies, we couldn’t have the class. Our argument is disproved but do you think we could have a programmed and easy life if we didn’t make this induction? If we didn’t take this generalization for true because it’s epistemologically invalid, we wouldn’t have a timetable and our school-life would be a mess. This is the reason why we go on making inductions although we know that they are very weak in leading us to certain truth.
To sum up, generalizations are more likely to lead us to false conclusions then the specific observations we make, whereas we are still in need of them to have a meaningful and organized life. Logically, we always have a minimum risque that one day a space object will effect the earth and its gravity laws, but how can we live if we can’t generalize that when we stand on the floor, the floor will carry us and we will be able to move on the floor? This is why generalizations are so important and essential, ignoring all its defects.