Finish Line

As we wind up the semester, it is natural to reflect back on the course description for ‘lexical semantics’:

“Lexical Semantics,” informally speaking, is the study of the meaning of words. Both “meaning” and “word” are actually extremely challenging concepts to understand, so much of the class will be devoted to unpacking these two concepts. In the process, we will overview various dimensions of meaning, how meaning is structured within words, and, conversely, how the context in which a word occurs interacts with its internal meaning. We will also consider relationships that exist among words and among elements out of which words are composed.

Our course description raises many questions, and I look forward to your attempts to answer them . . .

• What are problems with the concept “word,” and why is it not used as a technical term in and of itself?
• How do we analyze the internal contents of words?
• How is the meaning of a word represented?
• What are some properties about the way the meaning of words is represented?
• How do we explain how words are related to one another?
• What do we know when we know a word, in addition to its meaning?
• What do we have to learn when we learn a word? What properties of words don’t we learn?
• What other information is there besides information provided by the meaning of the words we use?

I will be constructing our final test this week, partially in response to your feedback in this blog space. The test will contain three types of questions. Part I will be in an objective format. I’m interested in your understanding of concepts like “listeme,” “function,” “content,” “argument,” “semantic feature,” “synonymy,” “antonymy,” “hyponymy,” “semantic rule as opposed to syntactic rule,” and the relationship of word meaning to overall information. Here is a sample question:

Parker and Riley (Linguistics for Non-Linguists) point out in their textbook that “linguists have been unable to agree on exactly how many and which features constitute the universal set of semantic properties.” Which of the following features for the word “dog” looks most problematic, from the perspective of the argument over what constitutes a member in the universal set of features. That is, which of the following features would most likely not be a member of the set of semantic features we are born with?

1. [-human] 2. [+canine] 3. [+living] 4. [+countable]

Part II will concern your ability to problem solve, to carry out a linguistic analysis. I might ask you to look at a set of data, evaluate generalizations about the data, and then to test the generalizations against additional data, somewhat in the spirit of our work comparing Harley’s and Plag’s analysis of affixation.

Finally, part III of the test will be a question that you yourself write and answer in essay format. This question should serve as a platform for you to reveal an important theme in our course. The ensuing treatment of the theme should involve both a conceptual and technical dimension. You should connect to the class discussion, but you should also touch upon material in the reading that we haven’t discussed in class or in our blog space. You should reveal some ability to use the technicalities of the discipline and to open new ground, while making some sense out of it.

Plan on approximately 20 minutes for each part. Bring paper to class, but you will need to leave your reference material and class notes at home. Be prepared, you don’t want the wheels to come off ’round the final curve!

Attractive Opposites

Philosophers and others from Heraclitus to Jung have noted the tendency of things to slip into their opposite states; and many have remarked on the thin dividing line between love and hate, genius and madness, overlook (a mistake) and (scenic) overlook . . . . The paradox of simultaneous difference and similarity is partly resolved by the fact that opposites typically differ along only one dimension of meaning: in respect of all other features they are identical, hence their semantic closeness; along the dimension of difference, they occupy opposing poles, hence the feeling of difference.” D.A. Cruse

Last week, I approached the idea of synonymy from the perspective of semantic feature theory. If two words share all the same semantic features, they qualify as synonyms. Hence, “girl” and “lassie” are synonyms on the basis of sharing the feature set ([+person], [-mature], and [-male]). It is necessary to concede that even though these two words ‘mean the same thing’ according to this theory, they are certainly not always naturally interchangeable. Thus, there are other things which determine the appropriateness of a word in a particular situation, having to do with the linguistic location of the speaker (is the person in the US or Scotland, for instance?), what is the intended referent (is the named object a young female human or a personified entity, like a pet dog?), the feelings that the speaker has for the referent (does the speaker wish to express emotional neutrality or avuncular familiarity?), the formality of the situation (is the word used in a professional conference or in a personal reflection at home in front of the fireplace?), the modality of the expression (written or spoken?) and so on. So the total information conveyed by any expression will be much more than the meaning associated with a word. One advantage of this theoretical approach to synonymy in comparison to Cruse’s careful descriptive approach is that it establishes that synonymy is something that is real, and not a hypothetical ideal at the unreachable end of a gradient scale.

In contrast with ‘synonymy,’ the definition of ‘antonymy’ in the framework of lexical semantic feature theory declares this condition to hold whenever two words have all the same semantic features, though for one and only one binary feature, the two words are oppositely specified. If we return to our thesaurus entry for ‘girl,’ the one antonym that is given is ‘boy.’ We can confirm this antonymy by noting that ‘boy’ will have all the same semantic features of ‘girl’ in terms of personhood and maturity level, but will be specified oppositely on the feature of sex.

Interestingly, in our explorations of the synonymy of ‘girl,’ we ended up ruling out many of the candidates provided by the thesaurus, narrowing down the list substantially. Thesaurus lists of antonyms tend to be much smaller than lists for synonyms (why might this be the case?). In this case, it is natural to see if we can extend the list somewhat. Now our theory provides us with exactly the direction to explore, as we look for other antonyms of ‘girl’: What other words have all the same semantic features as ‘girl,’ with the exception of one oppositely specified? If we hold personhood and sex the same, we can modulate the feature for maturity and get ‘woman’ (and all of its synonyms) as another antonym of ‘girl.’ Psycholinguistic tests could be constructed to investigate whether ‘woman’ is as intuitively antonymous to ‘girl’ as ‘boy’ is. Note that we do not have the option of finding non-person antonyms for ‘girl,’ that is, something that would be identical in sex and maturity level, but a non-person. So ‘heifer’ (young, female, bovine) would not have a feature specifying personhood, but would have one identifying its particular place in the animal kingdom, and, hence, would not qualify as an antonym. This outcome may alert us to potential words that do not yet exist! Something like a “non-girl” would not help to fill this gap, because the prefix ‘non-‘ would overshoot the class of potential antonyms by covering everything that did not have all the features of ‘girl.’

Note finally that a single referent could be named in a variety of ways leading to all kinds of different antonymous sequences. So Miska, the little girl in our photograph, is a natural referent for ‘young’ (antonymous with ‘old’), ‘child’ (antonymous with ‘parent’), ‘small’ (antonymous with ‘big’), etc.

Conceptually, the theory of semantic features is quite attractive, but, as we have already considered, just what constitutes a feature is a difficult problem to resolve. Again, the close descriptive work that Cruse carries out in his chapter on ‘opposites’ gives us some sense of the challenges facing any theory of semantic features.

Sameness and Difference

It seems probable, and many semanticists have maintained, that natural languages abhor absolute synonyms just as nature abhors a vacuum.” D. A. Cruse

In my last blog essay, I defined “girl” as something that could be considered as comprised of the set of semantic features ([+human], [-male], [-mature]). I went on to suggest that other words that have the same set of semantic features would constitute synonyms for “girl.”

Now the thesaurus at lists the following entries as synonyms of “girl”:

babe, baby doll, bird, blonde, bobby-soxer, boytoy, broad, butterfly, canary, chick, coed, cupcake, cutie, dame, damsel, daughter, deb, debutante, doll, female, filly, gal, jail bait, lady, lassie, mademoiselle, maid, maiden, minx, miss, missy, mouse, nymph, nymphet, piece, queen, schoolgirl, she, sis, skirt, spring chicken, teenybopper, tomato, tomboy, virgin, wench, witch, woman

Just how many of these words are really synonyms, according to our technical definition? Many of these words are disqualified as synonyms because they differ according to the feature [-mature]. They may either be unmarked for this feature, as in the case of “female” and “she,” which are technical superordinates of “girl,” or they may be positively specified for the feature, in which case they would be antonyms (!) These include the following:

dame, gal, lady, and woman

So a first important lesson to be gleaned here is that relatedness does not mean synonymy. There are also many items in the thesaurus listing that contain additional semantic features distinct from ‘girl,’ yet without all the features constituting “girl,” making them technical overlaps . . .

Blonde, daughter, maid, queen, sis, virgin, and witch

or, in the case of words having all the features of ‘girl’ and then some, hyponyms . . .

bobby-soxer, cutie, deb, debutante, schoolgirl, tomboy, and teenybopper

Items in the list like the following appear to suggest sexual maturity, but not necessarily full social and intellectual maturity, and pose a challenge to our system of features.

Babe, baby doll, broad, coed, piece, and skirt

From these examples, it looks like the feature [mature] itself should be broken down further, if we are looking for the basic semantic building blocks of language in these semantic features.

Many of the words remaining on the short list of potential synonyms for “girl” are obvious metaphorical usages of words that have quite divergent meanings. If we understand metaphors as words which by definition violate a semantic feature, then we can eliminate the following:

bird, boytoy, butterfly, canary, cupcake, filly, jailbait, minx, mouse, nymph, nymphet, spring chicken, tomato.

We are left with the following ‘pure’ synomyms (I leave it to you to check whether they ‘cognitive synonyms’ according to Cruse’s definition):

Damsel, lassie, mademoiselle, maiden, miss, and missy

Saying that these words ‘mean the same thing’ or ‘have the same semantic features’ or are ‘synonymous,’ does not mean that they are interchangeable, can occur in the same contexts, or are exactly alike. That is because word meaning is only one ingredient in any instance of language use. Critically, in addition to meanings of words, there are speakers of words and referents for words. A speaker who refers to a girl as a ‘mademoiselle’ is saying something about his own social affliliations, quite apart from the meaning of ‘girl.’ Likewise, words occur in situations, which are always nuanced. So I can use “miss” to get a girl’s attention (maybe a young sales clerk at a store), where the use of ‘girl’ could perform the same function but with less politeness. “Damsel” is conceivable only in a somewhat archaic, poetic context. (Cruse would say that is has a very different “field” of usage), and “lassie” may say something about both the regional, dialectal affiliations of a speaker as well as something about the closeness of feeling the speaker has for the referent.

Recognizing that a language event is more than word ‘meaning,’ but also people interacting in a variety of relationships and with a variety of social affiliations and communicating in a variety of forms helps us to understand how there can be a significant amount of synonymy, while at the same time no two words are truly alike.

Parts and Pieces

A natural way to think about the meaning of a word is to break it down into its parts, something technically different from breaking a word itself into its parts. The latter case is, for example, the breakdown of a word like “unrecyclable” into the listemes un-, re-, cycl(e), and –able. Listemes, in turn, can be dissected according to their semantic contents. Take for instance, the listeme “girl.” The semantic parts that comprise this listeme include the notion of femaleness, humanness, and youth. Linguists have tried to formalize these semantic parts, not without some considerable controversy. One proposal is that all semantic parts come from a limited set of semantic features that we have stored in the mind, and that are part of our genetic endowment. These semantic features are posited to be the basic, binary stuff out of which all complex meaning is created.

Returning to our example of ‘girl’ above, we might formalize its semantic parts as features consisting of at least the following: ([-male], [+human], [-mature]). We would interpret this formalization as follows: the listeme ‘girl’ is comprised of three binary semantic features: it is negatively specified for the semantic features [male] and [mature], meaning that ‘girl’ is not a male and not someone who has reached the age of full maturity; conversely, it is positively specified for the semantic feature [human], meaning that ‘girl’ is a person and not another kind of animal.

Maybe this formalization process looks a little ugly to you, but there are many exciting dimensions of thought that it opens up. Ultimately, a test of its value is its usefulness in explaining things. One thing that any semantic theory should be able to explain is the idea of synonymy. With our formalization, we could suggest that words like ‘girl’ and ‘miss’ are synonyms in that they are each comprised of the same semantic features, that is both are ([-male], [+human], and [-mature]). Antonymy would arise whenever two words are alike in all there semantic features except one: so ‘girl’ would be an antonym with ‘boy’ because ‘boy’ would have all the semantic features of ‘girl,’ except that it would be positively specified for the feature [male]. On a different dimension ‘miss’ differentiates itself as an antonym to “Mrs” in a specification that we might call [mate]; thus, ‘miss’ and “Mrs” are alike in all features except that “miss” is [-mated] while “Mrs” is [+mated].

Exploring further relationships, we could say that one word is a “hyponym” of another word if it contains all the semantic features of the other word, and then some. So, for example, “girl” is a hyponym of “youngster” because in order to specify “girl” you need all the semantic features ([+human], [-mature]) that are needed to specify “youngster.” Conversely, we could call “youngster” a “superordinate” of “girl.” Hyponymy can be differentiated from concepts like overlap in ways that I will let you explore on your own.

There are some exciting philosophical implications of this model for understanding lexical semantics: words are formed from connecting already present semantic features; one result is that very complex results can be established in next to no time: Consider a word like “persuade,” used by children of any language at early ages without specific instruction. Its meaning is actually wonderfully complex, as it can be broken down into something like . . .

‘to cause someone by means of the process of disputation to do something which is successfully accomplished, and accomplished voluntarily and not through any act of coercion.’

Because these semantic features can be quite complex, humans learning a human language will learn something like the concept “rabbit” long before fathoming the ostensibly identical “undetached rabbit parts.” That is, words will come in the form of well-defined “parts,” not any old conceivable piece of a word will be able to make up a part.

Conversely, just as we are designed by our genetic endowment to package the world in certain ways, it follows that there will be a whole universe of possible intellectual conceptions that are in principle impossible for human beings to conceive. Of course, these impossible conceptions are difficult to fathom because they are by definition unfathomable. Perhaps certain mathematical concepts point us in the direction to look. Here we have concepts that we are led to posit by a logical process, even if the concepts themselves are unfathomable . . . I’m thinking of notions like “instantaneous change” (fundamental to Calculus), “imaginary numbers” (like the square root of negative one), and “quant” (a physical unit that is neither mass nor energy).

Role Playing

This week I would like to “problematize” our reading on role playing. . . as I am on the lookout for prospects for technical papers . . . The challenge is how to get underneath the surface of things to reveal the dynamic hidden depths, the “round” character of issues which tend to be “flattened” when literature has a non-specialist audience.

When choosing a point for investigation, it is usually best to choose something small and pointed; one never knows what Pandora’s box will be lurking around. Harley, in last week’s reading, tells us that English has a suffix -er, which turns verbs into nouns. She writes that “ –er makes nouns referring to the Agent or Causer of the verb it attaches to”(209). She gives the following examples:

a) employer, climber, fighter, rider, writer, sleeper, singer

So an “employer” is someone who employs somebody, a “climber” is someone who climbs something, and so on. Harley’s generalization leads to the prediction that if we try to attach ‘-er’ to a verb that does not have an agent among its role players, we should get something ungrammatical. Consider the verb ‘rain’; as there is nobody or nothing that ‘rains’, we predict that such a verb will not support the agent –er ending:

b) *Adam rains.
c) *The weather rains.
d) *Today was a rainer.

Likewise, verbs that are ambiguous in that they optionally project an Agent role player will be disambiguated as a consequence of Harley’s generalization. The verb “melt,” for example, can take either an agent or theme role player:

e) Dorothy melted the wicked witch of the west

If I asked you who was the melter (as opposed to the meltee), you would respond, “Dorothy,” presumably. So far, so good, right? However, Ingo Plag (Syntactic category information and the semantics of derivational morphological rules) points out the that situation is not as simple as Harley implies. That is, Plag notes that –er may attach to other things as well as verbs, a fact that Plag emphasizes is often glossed over in the standard literature (a researchable conjecture, here!):

f) loner, foreigner (adjectives turned into nouns)
g) banker, Londoner, pin-striper, weekender (nouns that continue to be nouns)
h) four-wheeler, five-leafer, left-hander (phrases turned into nouns)

So all of a sudden, the position that –er creates agents out of verbs is completely shaken: a “loner” is not one who “lones” (compare “loaner”). The base form “lone” is not something associated with an agent role player; indeed, it is not a verb. Likewise, a “Londoner” is not one who “Londons,” a “left-hander” is not one who left-hands, and so on. Plag, citing Ryder (1999), suggests that “-er nominals are interpreted on the basis of the most plausible event schemas associated with the base”(13). Now that’s quite a mouthful, but note he makes no reference to the syntactic category of the base, nor to argument structure.

Ok, . . . but Plag himself recognizes that his technical sounding statement is itself not well-defined (at this point I’m asking: What is an ‘event schema’? What makes an ‘event schema,’ whatever it is, ‘plausible’? And finally, what exactly is the nature of how the ‘event schema’ is ‘associated’ with the base?) Plag admits in a footnote that “[s]uch ‘event schemas’ could be conceptualized and formalized as frames (e. g. Fillmore 1982, Fillmore et al. 2001) or scripts (as in Meyer 1993) . . . . [but that a]n implementation of these formalisms is beyond the limits of this [=his] paper”(39).

As is the nature with so much technical literature, the research solves some little things, but ultimately has the function of begetting much more research. To write a technical paper is to start a conversation, not to end one, or to say the last word. Here “event schemas” “plausibility” and “association” would need to be formalized in a way that would help us understand our understanding of this last sentence:

i) We want Eve to be the dropper Adam to be the breaker in this scene.

We interpret Eve to be the one doing the dropping, not the one being dropped . . . but Adam is the one being broken, not the one doing the breaking. And at this point, what you’ve got is a nice little research agenda for an end-of-semester technical paper.

How closely does your native language parallel English on these issues?

Harley, Heidi 2006. English Words: A linguistic introduction. Oxford: Blackwell Publishing

Fillmore, Charles J. 1982. Frame Semantics. In Linguistic Society of Korea (ed.), Linguistics in the Morning Calm. Seoul: Hanshin, pp 111-138.

Fillmore, Charles J., Charles Wooters, & Collin F. Baker. 2001. “Building a large lexical databank which provides deep Semantics” In: Benjamin Tsou & Olivia Kwong (eds.), Proceedings of the 15th Pacific Asia Conference on Language, Information and Computation. Hong Kong: City University of Hong kong

Meyer, Ralf. 1993. Compound comprehension in isolation and in context. The contribution of conceptual and discourse knowledge to the comprehension of novel noun-noun compounds. Linguistische Arbeiten, Band 299. Tübingen: Niemeyer.

Ryder, Mary Ellen. 1999. “Bankers and blue-chippers: an account of –er formations in Present-day English.” English Language and Linguistics 3, 269-297.

Argument Structure

Lately, we have been considering different dimensions of meaning. So, for example, when we were looking at the English pronoun “it” in the following expression:

(1) “Look, I’ve got an apple, and I want you to hold it”

We can describe its grammatical meaning as ‘third person, singular.” Another way of looking at it; whatever “it” is, it cannot be the speaker, or the person spoken to, and it cannot be more than one. We know, furthermore, that ‘it’ may refer to ‘an apple’ in our example sentence, it can refer to something else besides an apple (perhaps the speaker is a beautiful but troubled blonde woman and she would like you to hold her pet snake, while she inspects an apple). Finally, ‘it’ may not refer to anything at all. That is, the expression ‘hold it’ can be interpreted, under the right circumstances, as the message ‘halt’!

Now I would like to dedicate this week’s blog entry to an aspect of this last dimension of meaning mentioned in the paragraph above. We have called “it” in the expression “hold it” (meaning ‘halt’) an expletive. In this case, ‘it’ has nothing to refer to. Note that “hold it” in this sense cannot be paraphrased by *“hold your body” or “hold your motion.” We have considered that the presence of ‘it’ is mandated by the fact that ‘hold’ is a so-called transitive verb, a word that requires both a subject and an object. Hence the impossibility of (3) below in comparison with (2):

(2) I want you to hold it!
(3) *I want you to hold!

In the terminology of Heidi Harley (English Words: A Linguistic Introduction), content listemes have an argument structure as part of their meaning. That is, part of the meaning of the listeme “hold” is that it establishes a relationship between two arguments: there must be a holder and a holdee . . . at least in the grammatical sense. (Can you tell me what term Harley would use for each of these arguments?). This is the case, even if the logic of the expression does not require them. Just as for the more contentful dimensions of words, we must also learn their argument structure, something that cannot always be predicted by semantic content. Consider the following sentences:

(4) Adam ate voraciously.
(5)*Adam devoured.

Why is (4) a grammatical sentence of English but not (5)? Note that we cannot appeal to meaning, as these two sentences mean essentially the same thing!

Lexical listemes may be associated with a range of different arguments. Consider the case of “like” below:

(6) Adam likes Eve.
(7) Adam likes going to school.
(8) Adam likes to go to school.

In (6) we see that like takes a so-called simple argument—a noun phrase. In (7) and (8), by contrast, like takes a propositional argument, a gerund in the case of (7) and an infinitive phrase in the case of (8). Again, to stress the distinction between structure and meaning, let’s look at the superficially antonymous set of sentences below:

(9) Adam dislikes Eve.
(10) Adam dislikes going to school.
(11) *Adam dislikes to go to school.

Note that there is no way to predict from the meaning of ‘dislikes’ that sentence (11) is ungrammatical. Its paraphrase is perfectly legal:

(12) Adam hates to go to school.

Hence we can see that argument structure is quite a separate thing from the lexical content of a listeme.

We have seen situations in this essay where meaning and structure clash. Harley points out, towards the end of our reading for this week, that when this happens, structure always ‘wins.’ Thus she points out that although ‘coffee’, for instance, is a mass noun (notice we quantify it with ‘much’ and not ‘many’), we may force a ‘count’ interpretation of this mass noun:

(13) I’ve had three coffees today. (‘coffees’ now means ‘cups of coffee’)

Adjectives have the structural diagnostic of permitting comparison. That is why a sentence like (14) below could be heard, even though the semantics seem to prohibit it:

(14) Today was the most perfect day I’ve had in a long time.

Here’s Noam Chomsky’s famous sentence:

(15) Colorless green ideas sleep furiously.

As a testament to the priority of structure over content, note just how well (15) works compared to the more meaningful but structurally impossible (16)

(16) *John sleep bed nights.


Can you think of some examples in your native language which illustrate the showdown between structure and meaning?

Border Patrol

One intention with last week’s “mílujem ťa” was to show how the notion of “listeme” can be rather more helpful in explaining language than the notion “word”: As I pointed out, the Slovak expression has two words, while the English expression has three, yet they mean the same thing. Thanks to Eva S, we can highlight further the problem with the notion “word,” she writes:

“But let`s complicate the task with a Hungarian word “Szeretlek” which means “I love you”. It is just one word in which everything is included as in “I love you”. In “szeretlek” there is included both the object and the subject.”

So whether we have one, two, or three words we nevertheless have the same meaning in each case, crucially because each of these expressions contains exactly the same three listemes: the content listeme {love}, and the two function listemes {first person, singular, subject} and {second person, singular, informal, object). I leave it to my students with a knowledge of Hungarian to show exactly how “Szeretlek” expresses these three listemes.

I think most readers could see the distinction between word and listeme involved in analyzing “mílujem ťa.” However, how to separate out the boundaries between the three listemes turned out to be considerably more contentious. Zdenka K thought that the scope of the content listeme should be a bit more expansive:

“As far as concerning the word MILUJEM TA, in my opinion the base of the verb is MILU and the person who wrote it is JEM.”

And Michaela L argued to push the boundary of the content listeme “to love” even farther rightward into the sentence:

“But in the word “Milujem”, I think that this ending “m” means that “I” love somebody. If there will be “miluješ” – it means that not “I” but “you” love somebody (ja milujem, ty miluješ, on miluje, my milujeme, vy milujete, oni milujú). I think that the root of the verb “milujem” is “miluje”. Because, “mil” is not a verb, we do not know from which word, verb it is derived.”

So we have as possible candidates for the content listeme we associate with the meaning love: “mil-“, “milu-“, or “miluje.” One of Michaela’s problems with the first two possibilities is that they do not stand for something on their own; that is, they cannot be used as separate phonological words in a sentence. This, however, should not rule out their status as listemes, which are defined as he smallest units of meaning in a language, and not as something that must stand alone. When it cannot possibly stand alone, we have the terminology “bound listeme” for describing it. Note that we would have a problem in considering “miluje” to be a listeme in that there are not one but two meaningful units, the contentful notion involving love, and the functional notion involving a third person singular subject: “miluje” means ‘he or she loves”.

Deciding on a content listeme between “mil-“ and “milu-“ is a bit trickier, although a beautiful posting by Lenka V helps point the way. Notice that all the words in Lenka’s list share two things: in terms of meaning, a positive, affective denotation; in terms of form, the sounds constituting “mil-“:

“In the classroom last week you asked about the meaning of ‘mil-‘, and then about what the love is. That discussion motivated me to do a little research: Immediately after the lesson I went to the library and took an old Slovak dictionary (Slovník slovenského jazyka II. Vydavateľstvo SAV. Bratislava 1960). I found all the words beginning with ‘mil-‘, meaning something positive. I came to a list of 30 items as following:
milenec, milenecký
milkovať sa, miliskovať sa
milký, milkovný
milohlasný, milohlasý
milosťpán, milosťpani, milosťslečna

All these words have a listeme indicating positive feelings (or love) of somebody towards another person or thing, or pleasant, agreeable perceptions. The underlined items are strange to my computer and probably to a modern Slovak reader too, because they were created by some Slovak writers like Andrej Sládkovič (milokrásny, milovonný, milošialený) or P. O. Hviezdoslav (milký, milkovný, milohlasný). I did not understand the words put in italics. Just to make it clear: the term ‘milostnica’, introduced by Margita Figuli (also a Slovak writer), was used to denote a beloved woman or a courtesan; and Milostenka is the name of a goddess of attractiveness, Grace (according to Sládkovič). Another surprise for me was the fact that I have never heard about ‘milota’ denoting a genus of gramineous plants (milota veľkoklásková, milota chlpatá). No idea what it is, however, interrresssting… Some proper nouns should belong to the list above too: Milada, Milan, Milena, Milica, Miloslav, Miloslava, Miloš, Milota (again!). I am sure that they were created through the application of our “positive listeme” as well as all the mentioned nouns.”

To capture the overlap in meaning in the list that Lenka V provides, we must narrow down to the listeme “mil-,“ the only surface phonological part that all the words share. Lenka’s posting is an excellent model of the way we can circumstantially detect bound listemes, those that cannot stand alone as a phonological word.

Now Mirka O’s question can be sorted out:

“I just wonder about how did you came up with the idea that the first part “MIL” of the word MILUJEM means something laden ‘love’. When I think about “MIL-“, I rather connect it with another Slovak word “mily” (kind).”

We can trace the difference between “mil-” and “mily” not to the meaning of “mil-“, but rather to the contribution made by the functional listeme “-y,” which may have the role of somewhat softening the intensity of “mil-“.

But this last claim points to another “boundary issue” regarding in particular content listemes: their meaning tends to be a little slippery. As Petronela L observed . . .

“I would like to add my personal experience with the expression ´milujem ťa´, or, more precisely, with its french equivalent ´je t´aime´. When I was in France, I met a boy and after some time, we became really good friends. As the time went on, the expression ´je t´aime´started to appear in his messages and e-mails and I became confused about what it could mean, so, after some hesitations, I simply asked him. Much to my surprise, it turned out to be just a casual way to finish one´s message or e-mail! I made a fool of myself, but the idea of saying ´milujem ťa´to any of my friends appears really strange to me! This proves that even such a simple expression like ´milujem ťa´ doesn´t necessarily have to have one clear content and clear meaning. In one language, it is used to express feeling of love, whereas in another, you can use it when adressing even to your neighbor!”

That fluidity of meaning in content lexemes is not just across languages, but occurs language internally, as Alena B observed . . .

“Many words and expressions are becoming less and less intensive. Even their true and original meaning is sometimes arguable. They are being used so often and in so many different situations that they start to lack the power they once had. They are more and more often being taken for grounded. Milujem ta, for example. It used to be a verbal proof of one’s strong and true affection, emotional warmth towards some other person (your “definition” is nice as well); a testimony of profound feeling that one could rely on, believe in, consider to be of great importance…
Today, Ja milujem otca, matku, brata, sestru;
Also Ja milujem priatela, manzela, milenca; – boyfriend, husband, lover
But I also milujem vychadzky do parku – walking in the park
A ja proste milujem taliansku kuchinu – Italian kitchen
A ja ozaj milujem Channel 5…
Ja milujem on every step I make, in every situation, what ever comes into my way.
I am rather disappointed and sad to see/hear that even the expressions such is this are becoming usual, universal and worn out.”

Thus boundaries, both in terms of phonological form and semantic content prove to be continual points of interest. It nevertheless becomes quite clear when borders are illegitimately crossed, as Renata M shows us in this final anecdote:

“I add a special kind of explanation of the utterance “Milujem”: it has nothing to do with love… just I was playing with the verb milujem and I create a sentence Milu(as a slovak biscuit)jem (eat)=I eat Milu:-)”

Function and Meaning (Continued)

Like many places in Slovakia, Nitra has its share of ‘public art,’ expressed in all the advertisements, instructions, graffiti, and ubiquitous tagging of buildings, trains, bridges, fences, in short, anything that represents a potential canvas, something that can be seen by the random passers-by. So I wasn’t so surprised when a sidewalk in our fine city’s Stare Mesto neighborhood greeted me last week with the banner “Milujem ťa.” Of course, its full meaning is as inscrutable to me as the tags surrounding me as I walk through any underpass.

I look closely at the brownish paint—or is it dried blood—rather boldly spread out in front of my feet under the broad afternoon babie leto sunshine. Sidewalks rarely communicate so engagingly. A pleasant collection of lexical content and functional instructions which I can unpack as follows:

“Milujem t’a,” that is to say: ‘I have a certain feeling for you, an awareness amounting to approval, appreciation, and approbation, something stronger than mere liking, more kindred in intensity to hate, yet with positive polarity.’

Let’s investigate: there’s “t’a” . . . the single solitary walker, reader, familiar to the writer, or so the writer indicates, functioning as object of the writer’s love? Could it be me? If not, then who?

Continuing to decode, next there’s “-ujem” . . . . the single, solitary writer, agent, lover? Functioning as the speaker subject of the expression. Definitions so far are hard to come by: I’m left with the circularity of “I” and “you,” but I don’t know who.

But, aah, there’s that first part: “Mil-” from “milovat’” the content laden ‘love’ with all its entailments and entanglements. With ‘Mil’ we know that the speaker is human, has feelings, and knows what it is like to experience these feelings deeply, their presence and absence. “Mil-“ concept or atom, that which is composed of basic parts, or that which stands on its own, yet is connected with so much else? . . . such are the questions that run through the linguist’s mind . . .

Function and Content

Last week we spent some time thinking about the little Slovak word “no,” which turns out to have a very far-reaching and subtle range of significance. Words can be broken down into what linguists call “function words” and “content words.” Now this is a simplification, but the significance of function words is to organize the flow of information of the expressions in which they occur. Consider the following sentence pair:

(i) John was fond of Mary
(ii) John was found by Mary

If we try to explain the significance of the words “of” and “by,” we might have some problems, especially if we think only of dictionary definitions. It would be far easier to reflect on how they function rather than what they mean, and much more productive, practically speaking! In the first sentence, “of” functions to link “Mary” to “fond” as its object. That is, the relationship between “fond” and “Mary” and “John” is identical to what we have in the following sentence:

(iii) John likes Mary

In comparison, “by” in (ii) tells us that “Mary” is linked to “found” as its subject, and that “John” should be interpreted as the object of the action of the verb, just as in (iv):

(iv) Mary found John

Returning to Slovak “no,” we can see that it works on a variety of functional planes. Signifying either affirmation or negation, it has an almost mathematical formulation: it is an ‘operator,’ specifically, a “polarity item” (+/-) which tells how the associated expression is to be taken. We also saw Slovak “no” signify something close to ‘but,’ another organizational concept, linking two grammatically identical but semantically variant expressions into a compound structure.

Other instances of Slovak “no,” where the English translation would be something like “well,” showcase the “modal function” of the word, where ‘modality’ can be summarized as reflecting the speakers’ feelings about the associated expression.

Contrasting with function words, content words carry the lion’s share of the specific information in expressions in which they occur. But what exactly is their “content”? As we attribute definitions to the content words of our language, how do we account for relationships that exist among them? Why are “hot” and “cold” opposites, but not “lollipop” and “diagonal”? How are “cause” and “coerce” related? And do “boy” and “lad” really mean the same thing?

At the top of this essay is a photo of a t-shirt that I’ve been wearing around Slovakia. What is it that caused someone—the creator of this t-shirt—to select these words in particular? Why do Slovaks often laugh at me when they notice what is printed there? What can we glean about the nature of ‘meaning’ by thinking about other words that would be appropriate for this shirt. I suspect that these might be candidates: “burčiak,” “žinčica” . . . . Can you think of others?

Finding a Simple Place to Start

Let’s think about what it means for a word to “mean” something. Learning Slovak, I’ve had some surprises with the word “no.” Part of the problem is that we have the word in English, too, where it means the opposite of “yes,” signifying something I might summarize as ‘refusal’ or ‘denial’; in Slovak, however, I struggle with the fact that “no” can perversely mean “yes,” (agreement, acceptance, approbation) even though there is a very similar word—“nie”—which means, well, ‘no.’ So, after climbing for several hours up above Strbske Pleso in the High Tatras with one of my ‘uncles’ (husband of my grandmother’s sister’s great-grandaughter) I ask:

i) Mozeme ist dolu?

And, to my consternation, he answers:

ii) No . . . . paci sa.

Only after much subsequent conversation do I recognize that he is not insisting that we climb to the very top of Mount Krivan, but that he is actually agreeable to whatever maniac idea I may suggest.

I’m told that such a “no” derives from “ano”—a shortened form of ‘yes,’ most commonly used in informal settings.

So you see that word meaning can consist of what we generally call a ‘definition,’ (in this sense “no” and “ano” mean the same thing) but, in addition, there is an associated contextual requirement. That is, part of a word’s meaning appears to include a requirement on exactly what context in which it may be used. “Ano” (almost) always means ‘yes’; though “no” only means ‘yes’ in certain circumstances.

So far, we are in the normal everyday territory of ‘denotation’ and ‘connotation.’ But words commonly have many denotations. My Kratky Slovnik Slovenskeho Jazyka says that “no” can also mean ‘but’:

iii) skromne, no dobre jedlo.
iv) Vykloni sa, no pre hmlu nevidi.
v) Usiluje sa, no nestaci.

It also appears to have a signification similar to English ‘well’, indicating heightened intensity:

vi) Zda sa, ze maju radi zmrzlinu Slovaci. No, pravdaze!

Or lack of intensity, communicating resignation:

vii) Musim ist do Rusku. No, zbohom.

These examples show that we can fine-tune the notion of ‘context’: there are social contexts in which language is used, and in informal contexts “no” can mean ‘yes’, but there are also grammatical, or more specifically, syntactic contexts that serve to tease out further dimensions of meaning. So in (iii-v) above ‘no’ occurs in situations where it is clearly connecting neighboring words or phrases. Its ‘meaning’ clearly makes a functional contribution: here it indicates that neighboring syntactic elements are related by means of contrast. Under the circumstances, ‘no’ seems to have the same meaning as ‘ale’.

Perhaps ‘no’ owes the lion’s share of its popularity to its function as an indicator of speakers’ disposition towards the comments they are making. People always have some attitude regarding what they are hearing and saying. And ‘no’ can be used to communicate that what you have just said is very obvious (e.g. vi), or conversely completely surprising:

viii) No, tu ho mas!

It can communicate resignation verging on defeat (e.g. vii), or conversely hearty exuberance:

ix) No! suri ho otec!

And the affective dimension of ‘no’ can combine with the denotative dimension:

x) Si spokojny? No!

And in the case of (x) we get a convergence of denotative, affective, and contextual cues clustering on “no”, here indicating hearty agreement underscoring an informal context.

One of the maddening—or fascinating—features of language is that words can often indicate themselves . . . or their opposites, a point repeatedly illustrated in the examples above. An adequate ‘theory’ of lexical meaning needs to explain the frequency of this apparently surprising phenomenon, and might suggest that a word’s meaning is constructed by bringing together more primary units of meaning . After all, for something to be something else’s opposite, the pair must be alike in every respect except for one, with regard to which there is a binary opposition: ‘no’ can be the opposite of ‘yes’ because they are alike in every way . . . . except one (consider ‘boy’ and its opposite ‘girl’ both human, non-adult, living creatures—alike in every way, except one). So the meaning of a word taps into ‘denotation,’ ‘connotation,’ social context, grammatical function, and the nexus of other words which make up the dictionary of the language in which the word occurs. . . at which point you may ask:

xi) ‘Uz sme skoncili?’

To which the only appropriate answer right here and now could only be . . . ‘no!”