Concepts are not individual occurrences but a whole set of occurrences. The concept is in our mind and it is not an image. First we conceptualise. Then we need to lexicalise the concept, then to grammaticalise it and then use it in a conversation. We should take into consideration the whole extension, which means the more data we process in the act of conceptualisation the better the concept in our mind. The problem with this theory is that it may work with words but not with units larger than words such as: phrases, clauses or sentences that also have meaning. Linguists were disappointed by the theories that have developed so far. Then there came the Chomski's theory in the sixties which was part of the generative semantics (transformational approach). Their goal was to split up the word meaning, to go beyond the smallest meaningful unit. They split the word meaning not formally into morphemes but they analysed the morphemes into further smaller units of meaning. Those units were called semantic markers (primitives, features or components). The whole theory is semantic decomposition or componential analysis. This componential analysis works fine with words that have some lexical relation such as: boy, son, woman, daughter or brother, e.g. boy - + animate; - adult; +male;
+human. But how can we describe the red colour using this theory, red -+colour. Such examples ruin the whole theory. Also the words which are used to describe other words such as: +male or +human for a boy are lexical words themselves and according to the theory can be divided into smaller units. In order to abridge the difficulties, the semanticists created semantic distinguish-ers that are semantic markers of a first, second and third degree whose role is to describe the words semantically.