I will use this blog to help you study the linguistics course this year. I will explain the main concepts and give definitions and examples for the terminology we use in class. I hope this will make things easier for you. In the near future, I will try to make these notes available for you before class so you can prepare your questions about the parts that are difficult for you to understand. It is a good idea that you check this blog on a regular basis for comments, announcements and updates.
Good Luck

_____________________________________________________________________

Announcement
You can now post comments and ask questions on this page. Just go down to the bottom of the page, click 'comments', and post your comment.
______________________________________________________________________
Lecture 1
Concepts: Deixis, Definiteness

• Deixis is reference by means of an expression whose interpretation is relative to the (usually) context of the utterance, such as:


• who is speaking
• the time or place of speaking, or
• the gestures of the speaker.
  

Examples of deictic words in English: I, you, him etc., here, there; now, then, etc.; this, that
As I said in class, the interpretation of a deictic expression is context-dependent; yet there is a semantic class for each of such expressions that distinguishes it from other deictic expressions. For example, here and now are both deictic; but here refers to location and now refers to a point in time, no matter what the context is.

Practice: Identify deictic expressions (could be more than one word) in the following sentences and their semantic classes.

1. The Egyptian national soccer team won the 2008 African Cup of Nations 2 weeks ago.
2. It is 12pm now.
3. It was raining yesterday.
4. I will stop here now and get back to you tomorrow.

• Definiteness: I will get back to it soon.

______________________________________________________________________

Lecture 2
Concepts: Word, Reference, Polysemy, Synonymy, Semantic Extension, Prototypes

Today in class we discuss some of the central concepts in semantics. In the beginning I want to emphasize the point that semantics is concerned with NAMING the world, and giving LABELS to concrete things as well as abstractions.

The ideal scenario in this naming scenario would be that ONE <-> REFERENCE. That is, every word will have only one reference, and every thing is LABLED with only one word.

Reference: the relationship between a word or expression and the things in the world represented by those words or expressions. A referent is the specific thing that the word or expression refers to.

But what really happens in language is that one word could have more than one referece. For example, the word STAR can be used to refer to that object in the night sky or to a celebrity or a movie star. In this case, we have what we call POLYSEMY.

Practice: Give more examples of polysemy from Arabic and English.

On the other hand, we also have cases where the same thing is referred to using more than one word. As someone mentioned in class, the English words COAST, SHORE, and BEACH refer to the same physical thing in reality (of course with some slight semantic differences). In this case, we have what we call SYNONYMY.

Practice: Give more examples of synonymy from Arabic and English.

Back to POLYSEMY. When we have a case of polysemy what happens is that we use an existing word to refer to a new thing or concept. In other words, we extend the meanig of an existing word in order to be able to use it in naming this new thing or concept. This is what we call SEMANTIC (or MEANING or SENSE) EXTENSION. As you can see, using thing strategy in naming new things results in increasing the ambiguity in language. The other strategy would be to create or coin a new word for the new thing or concept. This strategy, however, will result in increasing the lexicon or dictionary in the language, which represents a serious load on human memory.

Question: Which of these strategies is more frequenlty used in naming new things?

When we use the strategy of semantic extension, we are actually doing a lot of semantic analysis unconsciously. Why? Because we do not just pick a random existing word and extend it to cover the new thing in reality. Instead we analyze an existing word and come up with a set of features of the things this word refers to. If these features are clearly similar to the features of the new thing, then we choose this word to describe this new thing. However, we do not have the images of all the things in our minds. We do not have in our minds for example images for all the cars we have seen in our life. That's why when we see a new car that we have never seen before, we know it is a car. We do this becuase we keep in our minds a record of the set of features that define what a car is. This is what we call a PROTOTYPE.


Click this link, what you see is a prototypical tree. It has everything every tree has: a trunk, leaves, branches, and height. To see what I am saying click this link. As you it is another tree. But if the trunk gets very small, and the height becomes less, then we have click this: a plant, where the trunk is called a STEM. For animal prototypes, click this link.

new If you like to read and know more about PROTOTYPES, here is a very usefull LINK.

______________________________________________________________________

Lecture 3 Sense Properties and Stereotypes (Unit 9)

In this lecture we will cover the following points:

[1] the distinction between SENSE and MEANING

[2] semantic classes

[3] some sense relationships and properties [synonymy and hyponymy]

• SENSE and MEANING

In semantics, some linguists like to differentiate between the SENSE and the MEANING of a word. Frankly, this difference does not have any practical consequences. However, the distinction we will follow throughout this course is a simple one:

Meaning is the semantic value (content) of a given word when not compared to other words; i.e., when it is not a part of a group or set of words.

Sense is the semantic value (content) of a word when compared to other words; i.e., when it is part of a group or set of related words.

For example, when we talk about synonymy we prefer to use the term SENSE. Remember that, synonyms are words that refer to the same reference. That is: coast, shore, and beach refer to the same part of the sea, so they are synonymous. Accordingly, synonymy is a kind of sense relations between a group of words. Some linguists like to define synomymy as follows:

Synonymy: the relationship between two expressions that have the same sense. Synonyms, therefore, are expressions which share a sense; perfect synonyms would share all their senses.

MORE EXAMPLES: conceal/hide purchase/buy freedom/liberty

Don't worry, all these definitions are the same, they just use different words.

Another interesting sense relation is HYPONYMY/HYPERNYMY. A very clear and simple definition of this term is:

Hyponymy: the relationship between expressions such that the meaning of one expression is included in the meaning of the other. Usually this sense relation is expressed in structures like X is a kind of Y. (Here is a paper I will present in a conference on the automatic acquisition of hyponymy).

EXAMPLES:

[1] cat is a hyponym of animal, AND animal is a hypernym of cat.

[2] apple is a hyponym of fruit, AND fruit is a hypernym of apple.

Sometimes, this semantic relationship is referred to as an IS-A relationship. For an easy way for the memorization of these two terms, you can think of hypernyms as fathers and hyponyms as sons.

• SEMANTIC CLASSES

A very important and legitimate question in this context is that: why do we care about clustering related words into sense groups?

To answer this question, let's first consider the following sentence:

He went to the BANK to cash some money.

If you check an English dictionary, you'll realize that the word BANK has about 18 different senses, (a highly polysemous or ambiguous word), the most frequent of which are "financial institution" and ''geographical formation" as in 'the bank of the river'. However, it is easy to determine that in the sentence above, the first sense (i.e., financial institution) is the sense that fits the context here. The question is: how do we come to this conclusion? In different words, how do we disambiguate the ambiguous word (BANK)? Or what type of inforamtion and/or processes do we used to decide the proper sense for the context?

One feasible answer is the following. We understand that BANK here means 'the financial institution' because we have words in the sentence such as CASH and MONEY. And these words are usually more related to FINANCE than to 'GEOGRAPHY'. Here are basically using key words in the context for disambiguation. But this means that if we are going to use key words in disambiguation this way, we need to list all the words that are related to this particular sense, which is time-consuming and frankly very boring. A good improvement of this method is to classify or clusters words into semantic classes or groups. This way we reduce the burden of memorizing long lists of words. We can find these semantic classes in thesauri, specialized dictionaries and glossaries.

MORE ON THIS TOMORROW

More examples of Sense Relations for NEXT CLASS

Opposites: the relationship of being "opposite in meaning". There are four basic types of opposites (or incompatibility of meaning):

1. Complementary (antonyms): expressions which come in pairs and which, between them, exhaust all the relevant possibilities. Being "not X" automatically means being "Y" and being "not Y" means being "X". EXAMPLES: dead/alive on/off married/unmarried
2. (Gradable) Antonyms: expressions are gradable antonyms if they are at opposite ends of a continuous scale of values (a scale which can vary according to the context of use). Thus, with gradable antonyms it is possible to be both "not X" and "not Y", but somewhere in the middle. EXAMPLES: hot/cold tall/short love/hate
   (Note: A test for gradability is to see whether the term can combine with something that quantifies it, like very or very much or a little. So one can be "a little hot", but what does it mean to be "a little dead"?)
3. Converses: expressions which express a relationship between two things such that one of the expressions conveys the relationship in one order and the other expression conveys the relationship in the opposite order.EXAMPLES: buy/sell husband (of)/wife (of) above/below
4. Non-binary antonyms (incompatible sets): expressions in sets of more than two members which are incompatible in talking about the same thing. All the terms in the given set are incompatible and together all the members of the set cover the entire semantic area. EXAMPLES: Seasons (spring/fall/winter/summer liquid/solid/gas
   (Note: in this kind of system of X, Y, Z, being "X" means being "not Y" and "not Z"; however, unlike the binary opposition of complementary antonyms, being "not X" does not imply "Y", it implies "Y" or "Z". In other words, something which has a physical state must be either a liquid, a solid, or a gas; if it is a solid, it is neither a liquid nor a gas. If it is not a solid, you know it must be either a liquid or a gas; however, you do not know which one it is merely from knowing it is "not a solid".)
   

Meronymy: the relationship of a part to a whole. Nose is a meronym of face.

See you soon insha'allah.