 Once a conceptual structure or part of it has been generated, the process of linguistic encoding can be initiated. This process can be subdivided into two successive stages. The stage of grammatical encoding and the stage of phonetic planning. Both levels of language production will be considered in detail in the following. But first of all, we look at the input to the level of linguistic encoding and we will use a certain working model to do so. Let's look at the input first. Before the level of linguistic encoding can come into action, we have to generate a conceptual structure or some people call it a pre-verbal message. This structure is language independent. Normally such representations are given in English as a meta language. Well, here are two conceptual representations for the same sentence. The first one goes back to Roger Schenck and it is a representation that has been developed in the context of conceptual dependency theory or CD theory. The second one goes back to Ray Jackendorf and it has been developed in the context of his so-called conceptual grammar. Now, both representations stand for the same sentence. The sentence is simply the man gives or gave. Well, it depends on what sort of attribute you include the book to the woman or the man gives the woman the book. Now, in Schenck's approach, you have this sort of template which defines the act of giving as an act of abstract transfer, a-trans. You have an actor which is normally the subject of the sentence. You have the object of giving which is the book and you have a direction from the man, the actor to the recipient, the woman. In conceptual grammar, well, it is virtually the same the event which here is explicitly mentioned in the past. You could do the same thing for the present. It is a causative action where a man applies the event and here is the event go possession which is the equivalent of abstract transfer in Schenck's approach. The thing is the book, the object and there is a path, the equivalent of the direction component in Schenck's theory. Well, the path is from the person himself to the woman. So this is a sort of language independent conceptual representation that serves as the input to the level of linguistic encoding. Now, here is the working model which specifies the level of linguistic encoding. It can be subdivided into two central components. On one hand the formulator which consists of several linguistic levels and on the other hand the mental lexicon which we will not discuss in this little clip because we have an extra clip that is dedicated to the structure of the lexicon and the lexical entries. Now, within formulating, the level of formulating you have two central levels. The first one is called grammatical encoding. Now, this is really the first level which retrieves from the mental lexicon the syntactic and conceptual information of lexical entries and generates as an internal output the so-called surface form. The second level is the level of phonetic planning and here we translate the surface form by means of retrieving the formal aspects from the mental lexicon about each entry in the input string. Translate it into some sort of phonological, first phonological then eventually phonetic structure which serves as the output of the entire formulator and can be forwarded to our articulators which eventually will be set in motion in order to produce an audible sentence. Let's look at the level of grammatical encoding first. Now, as we said grammatical encoding, this level of grammatical encoding translates the pre-verbal message into an ordered string of elements so-called lemmas. Now, the lemma as you will find out in the clip about the lexicon is the information that can be generally associated with syntactic and conceptual aspects of each lexical item in the input string. The lemmas are combined into phrases of certain kinds so they have to incorporate primitive notions such as subject, verb, object, so the basic syntactic functions, they have to incorporate aspects of word order. So all basic syntactic notions that are of course now language specific. Whereas the conceptual information is language independent, here the level of grammatical encoding must be language dependent because now we integrate all those aspects that apply to the morphosyntactic principles of a particular language. For example, case marking and as I've already said word order. Let us now convert our pre-verbal message that we've seen at the very beginning into four different languages into their language specific surface forms. One for each language. Now here we have the example in English and as you can see in English grammatical functions are largely expressed by word and phrase order. That is by the configuration of surface structure. So English is what we call a configurational language. And these word order principles have to be adhered to by the translation of the conceptual structure into a surface form. Now this is of course a representation that is by no means in line with let's say X bar syntax in generative grammar but it is a representation that expresses the formal and functional relationships within the sentence. Now here's Tagalog, a language that is spoken on the Philippines. Well let's listen for a change. So this is something like the man gives the book to the woman in this case. So we have in this case we have to convert the indirect object into a prepositional phrase in Tagalog and this is quite interesting. In Tagalog grammatical functions are expressed through case marking morphology and are not encoded in phrase structure configuration. So Tagalog is a non-configuration language with a great freedom of word order and the surface form itself seems to be flat. Flat means that all the main functional elements are on one general level. Interestingly in Tagalog the verb is sentence initial. Our next language is Finnish. Let's listen first of all. So this is the Finnish version of the man gives the woman the book. And as you can see in Finnish like in Tagalog we have a non-configuration language which has a great deal of freedom of arranging its main functional elements. And we have once more this flat structure of functional elements if we represent them hierarchically. The grammatical functions subject verb object etc are expressed through case marking at least subject to an object and are not encoded in phrase structural configurations. Nevertheless the verb and this is quite interesting in Finnish normally occurs in the second position. Finally here is Turkish and again let's listen first. Again this is our sentence the man gives the woman the book. And first of all in Turkish we have the verb in the final position. Again we have a flat structure. So Turkish like Finnish in Tagalog is a non-configuration language and the verb as I said is generally in the final position. So as a result of these observations of four different languages we know that the grammatical encoder delivers the language specific surface form in terms of a functional representation that can now undergo phonetic planning. Now phonetic planning is the process that successively translate the segments or fragments of the surface form into a representation that eventually will be used by the articulator. This is achieved by retrieving the formal information for each lemma from the mental lexicon. Let's go back to our working model. Here is our working model and the formal information is of course everything that has is represented in green here. So that is morphological and phonological aspects and they will be translated into a eventually a phonetic form. So let's illustrate that on the basis of just one example. So this could be our surface structure and let us now select one item to illustrate the level of phonetic planning. So first of all we have to represent to some extent the morphological structure of each item in the surface form. So for example give could be represented like this morphologically. However as we all know it's much better to represent it in terms of phonemes in this sort of representation. In looking at gives we know that gives has the following syllabic structure. It consists of an onset, a peak and two consonants in the coder. So this is the syllabic structure that has to be translated somehow because we have to know the structure to arrange words within connected speech. And finally we have to look at the phonological aspects and here of course the mere representation in terms of phonemes is certainly not sufficient. So if we take out the central vowel in this item then we know of course that phonologically in terms of distinctive features this is a high vowel which has the features plus high and minus low. To define its exact position on the cardinal vowel chart we need the criterion or distinctive feature minus advanced tongue root. And we know that it is a vowel which is of course not back and which is produced with spread lips so it is minus round. This sort of distinctive feature representation which will be explained in another video clip and in another virtual session has to be applied to each segment of each item in the surface form. As a result of the level of phonetic planning a string of pronounceable patterns is generated. Well let's summarize. We have outlined the basic steps of translating the pre-verbal message, the conceptual structure into a phonetic plan that eventually can be forwarded to the articulators. At the same time we have outlined the complexity of this process, the process of grammatical encoding and phonetic planning which in reality only lasts a few milliseconds ranging from syntax to phonetics. However we have for reasons of didactic simplification dispensed with discussing several additional phenomena such as for example questions of feedback, feedback between the various levels within the linguistic process or the level of linguistic encoding. Or we have dispensed with the structure of the mental lexicon where all the information about the items that we have to consider in the input string are contained. Well consult the virtual session, the linguistic encoding session for detailed information and also there is a second clip about the structure of the lexicon.