Social Sciences
Unfortunately, the social sciences of today clearly perform much worse, than they could and also should, in order to support the civilization intellectually and efficiently in its struggle against the purely materialistic world view (invented, perfected and to this day assertively forced by the natural sciences). There seem to be multiple, diverse problems causing this underperformance, and in this section we try to identify (briefly) the most important of them, together with their possible resolutions.
First of all, the phenomena analyzed and described by the social sciences are obviously much more complex and convoluted, than that of the natural sciences. The physical processes can be effectively separated from each other (at least mostly), they can be reproduced in laboratories and analyzed repeatedly with small modifications; on the contrary, the social occurrences are often complicated and irreproducible, and the factors behind them hard to isolate and analyze. But this problem could not justify in itself the undeniable deficiencies in the theories of the social sciences, especially, since also the physics and other natural sciences have to deal with long and complex processes occasionally, and they are clearly more successful at doing this.
Secondly, the social science is due to its physics envy apparently too impatient and not willing to accept that in its current, immature state it is not ready yet to supply such scientifically well-founded and proven theories, which could compete with those of the natural sciences in terms of exactness, reproducibility and indisputability. The social science is approximately at the same level today, where the natural science stood in the 16th century, and it actually should measure itself to the natural science of that time not of the present. But it does exactly this latter, which results in a hypocritical and frustrated manner of theory creation and presentation, discrediting the social sciences even more.
A severe methodical problem originates from the social sciences' unreasonable striving to completely mimic the procedures of the natural sciences, and to use arithmetical models by all means for making the theories (seemingly) more accurate, which leads to the misapplication of mathematical (and especially statistical) techniques for defining and measuring social variables and in the search for relationships between them. E.g., let's suppose, we would try to measure somehow the psychological qualities of extraversion and intelligence and the connection between them. Today, the procedure would be to construct a questionnaire with scales (e.g. 1 to 5) that attempt to "measure" the various aspects of these qualities through different questions, then to apply an arithmetical model on the marked numbers of the scales to summarize those and get an aggregate figure for both qualities, and finally to search for statistical correlations between the sums in the dataset and (potentially) to interpret them as indications of a relationship between the two traits. This approach seems to be scientific (since it uses numbers and mathematical models), but in reality this accuracy is only confined to the current questionnaire, to the persons who took part on the survey and to the model that was applied to the dataset, and the results can hardly be reproduced (or even interpreted) by other teams and/or within (even only slightly) modified circumstances. The main problem is that we are fixated on using numbers for the sake of accuracy, and this drastically reduces the generalizability of the model, because we are now bound to this actual survey and to the scales contained in it. Furthermore, it is already strongly questionable, whether the usage of numbers for psychological qualities can be justified at all, since it seems not to have much sense of speaking about e.g. twice as large extraversion or an intelligence 20 percent less. Considering these problems, apparently another approach would be needed, which does provide accurate measuring capabilities, but without the need of the usage of numbers and arithmetical models for the psychological (or other social) qualities.
Those with computer programming background surely all know the
(at first rather surprisingly) fact, that in the various object oriented
languages and libraries every object type (even the most complex one)
can be placed into a perfectly defined scale according to one of its characteristic
without the need of using any numbers during the scale creation procedure.
E.g. in Java the
Comparator<T>
interface and its
int compare(T o1, T o2)
method can be used (in C#:
int IComparer<T>.Compare (T x, T y))
to compare any two objects according to a quality,
and to determine a relation between them (less, equal or greater).
The comparing method has to fulfill some very important requirements
(it has to be transitive,
must give the opposite result for a switched pair,
and must place all equal elements on the same side of any third elements),
but has not to use any numbers for the comparison
(it is actually indifferent, how exactly the comparison method works,
while it can fulfill these requirements).
Performing this very simple, elementary comparison operation multiple times
on object pairs (selected by some sorting algorithm), we can finally get
a perfectly defined scale built from the objects according to that quality
(or can conclude that our comparison method is faulty, if it supplies
inconsistent results during the sorting).
We can also easily put any new object into this scale at the right position,
by using the very same simple comparison method,
and with the help of it we are even able to compare two complete scales
for their equality and for their level of divergence.
It means that this perfectly defined scale (created without any numbers)
can be used as a ruler or a measurement tape for that quality it was built upon,
since any element can be measured against it,
and can be labeled with the position (or name) of the scale's nearest item to it.
Could not we possibly use this procedure outside the object oriented programming realm too, for instance in the social sciences to measure variables, to build scales for these measurements and to compare these scales regarding their equalities and level of divergences? E.g., in order to measure the psychological quality of the extraversion with a scientific accuracy, it would be (perhaps a bit surprisingly) completely adequate to possess only one single elementary comparison function, which is able to determine (consistently) the relation between any two persons' extraversion characteristic, i.e. can say, which one of the two is more extraverted (or if there is no (determinable) difference between them). Applying this function (and a sorting algorithm) on a sufficiently high number of individuals (tens or even hundreds), we would finally get a very fine graded scale for this quality, which could also be used to measure (similarly to a ruler) the extraversion quality of any other person. And, analogously, we could construct another scale for the quality of intelligence, and then analyze the level of congruence or divergence between the two scales (with the classical methods of the mathematics and geometry).
Let's go (briefly) over the advantages, which this measuring method could provide, contrary to the arithmetical approach enforced today. First, the measuring function is obviously much simpler and more elementary, than the number scales used in the questionnaires. Actually, when we classify a quality with a number from 1 to 5, we perform unconsciously a very similar procedure, during that we compare the current item with the minimum and maximum we know (or could imagine), and then try to assign a number based on these comparisons. It would be much easier (and much less error-prone), if we could (and also would be urged to) perform these comparisons explicitly, and the scale would be created only as the final step, after we have already compared each (or enough) pair of items.
Such a comparison function (comparing two items from a certain aspect) is an elementary and very common operation of the human mind; in fact, this seems to be one of the strongest capability of our brain, which runs automatically and unconsciously behind a lot of more complex thinking and analyzing mental procedures. This simplicity and elementariness could make this method eminently suitable for using it as a survey tool, since this function is known to everybody and performed everyday (many times), and it is much easier and common to compare two items directly, as to give both of them (independently from each other) a classifying number from 1 to 5. Furthermore, the brain is much more consistent (because of their basic nature) in providing such types of comparisons, while assigning a scaling number to an individual object or occurrence can vary even for the same person from survey to survey (or yet from question to question).
Very useful (from a scientific point of view) is the "brutality" of the comparison function, which forces us to compare two items directly from a certain aspect, and does not allow us to avoid making a potentially impolite or insulting relation between two persons or other objects (which we would certainly try to avoid in the everyday life, when we should decide e.g. which one of two women is the more beautiful, or which one of two men the better and closer friend). The science cannot tolerate any tactfulness or sensitivity when determining the scientific truths, since this could decrease significantly the accuracy of the theories (to an extent threatening even their validity), and such a "merciless" function could considerably help eliminate the delicateness from the process (like in the natural sciences the invention of "tyrannically" unquestionable and indisputable measuring methods for lengths, time intervals, forces, etc. was essential for the further development). This brutality could also be beneficial for the person itself who performs this comparison, as he/she would be compelled and urged to take a position in this question, and to analyze and clarify his/her own thoughts and feelings (in fact, this method on its own could already be a valuable help in e.g. the psychological or psychoanalytical practice, or anywhere else, where there is a requirement to uncover hidden attitudes, opinions and sympathies).
A significant advantage would also be the fact, that the sorting algorithm itself already automatically validates the comparing function (at least technically), and (at a sufficient high number of objects and comparisons) it turns out without any further analyzation, if the function is faulty or not working consistently (for this purpose it would be advisable to compare all items pairwise, even if the sorting algorithm itself would not require this, especially at the beginning of using a function).
Applying this measuring method could facilitate considerably the exchange of the scientific results between various groups and laboratories, since the elementary comparing function could be publicized, analyzed and exchanged already in itself, before any further conclusions or without any complex theories. In fact, the primary activity of the scientific groups could be to construct and verify such comparing methods, and to assure that these are functioning in a consistent and coherent way, since the conclusions about the qualities and their relationships could be almost mechanically deduced, once these comparing functions perform flawlessly. The results of one group could be much more easily verified and reproduced by other teams, since they could check the suggested comparison method technically and substantially by performing (re)sorts using it on their own objects or with their own comparing staff. Essentially, a well constructed comparison method should sort consistently the same objects always the same way, and also new elements introduced to it should be immediately placed into their right positions, completely independently from the team, who actually performs the operation. E.g., in the example above, one scientific group could work out a method, which compares two persons from the aspect of their extraversion (giving a <, = or > result), and then the following requirements should be fulfilled in order to accept this method as scientifically correct: 1. a given person set must always be (re)sorted into the very same order (the possible equal elements could of course be in arbitrary order among each other); 2. any other team performing the sort on this person set must get the very same result; 3. any two persons selected by any team must be comparable by this method; 4. the method must function properly and consistently on any person set selected by any team in the world. A comparing method can be as simple as taking a look at two persons and determine the relation between them intuitively, or as complex as running through days or even weeks of various tests and analyzation to choose the right one (the lesser or the greater according to the method) from the two persons; but it must always supply a relation between two items from a certain aspect, and should not even try to evaluate them independently from each other.
With the help of well-functioning comparison methods, reference lists could be constructed, which then could be used to measure other items and to assign them a value or a name according to their (hypothetical) position in this list. For this purpose, well-known, simple and unchangeable reference objects should be chosen, e.g. (imaginary) persons of novels, theater plays, comics, films and computer games, or new ones should be created, if there are no suitable objects existing yet for a certain quality or aspect. After creating the reference list, any living person could be measured against this scale, and it could get a name or a label based on its position in relation to the reference objects. With this procedure, an individual's personality could be determined quickly and described in a unified way, which description could be interpreted by anybody, exclusively on the basis of the reference objects and the applied comparing methods.
A (consistent and already established) comparison function could also be used to test the comparing abilities of a person, and to teach him/her about the aspect behind the method and the procedure to perform the comparison. E.g., if we already have a well-functioning method to compare the extraversion quality of two persons, then everybody must get the same results performing this operation on a selected pair, and if somebody does get a diverging result, then it shows, that he or she has an improper notion of the extraversion quality, and should modify his/her own conceptions and ideas to make them congruent with the scientific way of thinking and understanding.
A well-behaving comparison method could also be used in reverse mode, i.e. to define new qualities and variables, and to teach and train this new conception to anybody, exclusively with the help of this comparing function (and possibly a reference list). E.g., let's suppose, we have a method to measure consistently "something", but we do not have a notion for this quality yet; instead of characterizing it textually, the comparison function and a reference list is perfectly sufficient to define this new quality, and to make anybody able to use it in itself or as a basis for other, more complex properties. Of course, in the everyday practice the two approaches (direct and reverse mode) would usually be followed simultaneously: at first we would try to create a function to measure a common quality with scientific accuracy, but at some point we would most certainly realize that it is too complex and impure to get handled by a method consistently, and then we would start to clean it and separate it into (sub)conceptions, which would be defined accurately and precisely with the help of the constructed comparison methods and the created reference lists. By using this method consequently, the social sciences could get finally a completely new and exactly defined terminology, where each conception is entirely characterized and described with its corresponding comparison function and (optionally) a reference object list.
After discussing the measuring problem and suggesting a (potential) solution to this problem, let's review briefly the remaining most important possible issues behind the failures of the social science. First, the activities in it should obviously not be organized already to the extent, as it is the case today (mainly due to the unhealthy mimicking of the natural science), since at this phase (analogous to the state of the natural science in the 16th century) rather genial individuals would be needed and not professional scientists working in over-organized and bureaucratic conditions. It is also questionable, whether the social science should ever be professionally organized in the manner the natural science is, because ideally it should also have the task (additionally to the knowledge acquiring) to cultivate the public and to make everybody to a learned and experienced psychologist, sociologist and philosopher eventually. Perhaps the best solution could be, if the social science activities were organized mainly through the Internet, and the majority of the works were carried out by (half)dilettante "scientists", who could share and discuss their own everyday experiences in a standardized way, for example using the constructed comparison methods described above. In this model, not final theories with exactly formulated conclusions would be created by professional scientists, but instead everybody could classify his/her own current life situation and problems with the help of the comparison methods (e.g. by choosing, which one of two shown situations is closer to his/her), and this way could find analogous case-studies and problems, together with the applied and suggested solutions. In this system, there would be no explicitly formulated and deducted hypotheses really needed, because the phenomena themselves would already be documented through the Internet, and everybody could use his/her own reasoning to discover regularities and relationships in them, eventually becoming an experienced and learned social scientist him/herself. The main end product of the social science would be (instead of theses and publications) a cultivated and enlightened society, with socially highly intelligent and wise individuals.