[Lexicog] lexical relation

Wed Apr 29 21:03:27 UTC 2009

Thanks for your helpful response, Ron. A lot more than I bargained on, but it certainly helped lay out the issue(s)! I'm wondering how to apply it, though ... All I wanted was a utilitarian means of helping the reader get from A to B; i.e., "I know /sororo/ but what was the word for 'slippery' again?!" Perhaps this has a lot to do with intended audience--I would expect speakers of the language to know the 'pair' (and not even think English 'slippery' when associating them)--but language learners / non-language speaking users would not be expected to know the pair. But in my (limited) experience, even language speaker are keen to point out related words, and it's my feeling that this should be reflected somehow in the dictionary.

At one level, just what /kind/ of relationship the words have is a not as important as /that/ they are related--to go to the entry for /sororo/ and find reference to /madali/ (and vice versa) is probably more helpful than not cross-referencing them at all. But, as you pointed out, determining and defining the relationship can be tricky, and it should have fairly wide application to be considered useful. So I'm back to where I started--wanting to demonstrate the relationship between the words and not quite knowing how to represent it. I'm aware of your penchant for semantic domains ;-), but even that has its challenges (at least in a dictionary with 'traditional' alphabetical layout)--how does the reader connect /sororo/ with semantic domain: 3.4.5.6 (completely made up) to /madali/ s.d.: 3.4.5.6? S.d. 3.4.5.6 is probably opaque to most dictionary readers ... Suddenly what I thought was a simple question has brought me to dictionary design philosophy! :-0

For now I'll probably make some kind of 'catch-all' lexical relation for fringe relations like 'slip' / 'slippery'--just to link them--and move on. Perhaps I can come back to this some day when I have more time and energy to spend on it. :-/

Paul
--- In lexicographylist at yahoogroups.com, "Ronald Moe" <ron_moe at ...> wrote:
>
> FLEx allows you to create new lexical relations. However this does not solve
> your problem. The big problem with lexical relations is that they are
> over-generalizations based on similarities in pairs (or sets) of words. For
> instance the part-whole relation can be applied to things as diverse as
> human body parts, parts of a car, and perhaps even parts of an event. But
> human body parts are not easily divided from one another and therefore have
> indeterminate (fuzzy) boundaries. The parts of a car can generally be
> disassembled and therefore have a very different relationship to the other
> parts surrounding them. Parts also can have functions. So the engine of a
> car has a different relation to the whole than does the paint. To say that
> the relationship of 'engine' to 'car' is part:whole is a gross
> over-simplification. That isn't to say that lexical relations aren't useful.
> But you have to be very careful to understand the semantics of each word and
> the precise relation between them. It is nice to be able to say that the
> agent:activity relation is exemplified by the pairs builder:build,
> writer:write, architect:design. But when you start getting specific about
> the precise relation, you find that a builder may not actually build
> anything. He may just supervise the carpenters and other subcontractors who
> actually do the building. A writer may use a laptop. So the precise sense of
> 'write' may be 'to author a document'.
> 
>  
> 
> When we look closely at pairs such as slip:slippery, we have to define each
> word. It helps to start with the theoretical notion of a scenario. With
> 'slip' we need to put the word in a frame such as "John slipped and fell on
> the wet sidewalk." First we have a surface on which John was walking. (He
> might have just been standing, but we'll assume walking.) Second we need
> some lubricant, in this case water. Third we assume he was on his feet and
> that the traction between his feet (presumably shoes) and the sidewalk was
> not sufficient to keep his feet where he planted them.
> 
>  
> 
> Next we look at 'slippery'. Take the sentence, "Watch out, the sidewalk is
> slippery." Here we have the same surface as in the previous example
> sentence. We have the same lubricant. We could have also chosen the example
> sentence, "This oil is slippery," in which we would have had a very
> different frame. Or how about the sentence, "My mother is getting feeble and
> has trouble holding onto slippery glasses when she is doing the dishes." So
> 'slippery' has a variety of frames and therefore a general meaning. But even
> if we limit the meaning to "the sidewalk is slippery", we have some serious
> difficulties in describing the relation between 'slip' and 'slippery'. Let's
> define both, but limit the sense to the sentences "John slipped and fell."
> and ".the sidewalk is slippery."
> 
>  
> 
> slip v. 1) to accidentally slide a short distance quickly because the
> surface on which you are standing is lubricated by some liquid or substance
> such as sand, especially when sliding in this way causes you to lose your
> balance and fall.
> 
>  
> 
> slippery adj. 1) if a surface such as a sidewalk is slippery, it does not
> provide much traction because it has a wet or oily substance on it that
> makes it easy for an object to slide on it.
> 
>  
> 
> Now we can start looking at the actual lexical relation that exists between
> 'slip' and 'slippery'. The practice of indicating a lexical relation by
> putting the two words together with a colon between them (slip:slippery) is
> misleading. Lexical relations are between the meanings of two lexemes, not
> between their forms. So to be more precise, the lexical relation between
> these meanings is actually:
> 
>  
> 
> to accidentally slide a short distance quickly because the surface on which
> you are standing is lubricated by some liquid or substance such as sand,
> especially when sliding in this way causes you to lose your balance and
> fall:if a surface such as a sidewalk is slippery, it does not provide much
> traction because it has a wet or oily substance on it that makes it easy for
> an object to slide on it
> 
>  
> 
> Once you recognize this little fact, it becomes obvious why lexical
> relations are illusions. Try to find two other words that have the same
> relation as that expressed in the previous paragraph. It turns out that the
> lexical relation between any two words is absolutely unique, because the
> meanings of the two words are unique.
> 
>  
> 
> So why do we even bother talking about lexical relations? It is because our
> minds have the amazing ability to generalize and see patterns in the
> confusing mass of details around us. The notion of 'part of (sth)' is a
> generality that we can apply to "My finger is a part of my body," and "The
> engine is part of the car." So for us to talk about a lexical relation, we
> need a generalized concept that we can apply to numerous pairs (or sets) of
> words. What generalized concept can be applied to the meanings of 'slip' and
> 'slippery'? It isn't event:location. It isn't event:cause. The best I can
> come up with is:
> 
>  
> 
> event:causative property of object associated with event
> 
>  
> 
> Can we think of other examples? How about cut:sharp, crush:heavy? But
> thinking of examples like this requires a lot analytical thought. The
> relationship is too complex and not frequent enough. For a lexical relation
> to be useful it has to be fairly simple and it has to be common (there have
> to be lots of pairs of words that are related in a similar way).
> 
>  
> 
> I prefer to deal with semantic domains. Many domains refer to a scenario. A
> culture/language conceptualizes many scenarios. Each scenario has a set of
> lexemes that refer to or describe various aspects of the scenario. So one
> way to look at lexicography is to identify each scenario and describe how
> each lexeme associated with the scenario relates to the whole. Step
> 1-identify a scenario. Step 2-collect all the lexemes that belong to the
> scenario. Step 3-describe the scenario. Step 4-describe how each lexeme
> relates to the overall scenario and to the other lexemes in the scenario. If
> you look at the FrameNet website, you will see one approach to what I am
> talking about.
> 
>  
> 
> Ron Moe
> 
>  
> 
>   _____  
> 
> From: lexicographylist at yahoogroups.com
> [mailto:lexicographylist at yahoogroups.com] On Behalf Of lengosi
> Sent: Monday, April 27, 2009 10:22 PM
> To: lexicographylist at yahoogroups.com
> Subject: [Lexicog] lexical relation
> 
>  
> 
> 
> 
> 
> 
> 
> What kind of lexical relation would be recommended for the following?
> 
> /madali/ Vsta 's.t. is slippery'
> /sororo/ Vi 's.o. slips'
> 
> I'm using FLEx which provides "paired" sets of:
> 
> instrument / action
> process / result
> undergoer / verb
> 
> but none of these quite seem to fit. There is the rather generic Counterpart
> relation, but this seems more specific than that. Any ideas would be
> welcome. Thanks,
> 
> Paul
> 
> 
> 
> No virus found in this incoming message.
> Checked by AVG - www.avg.com
> Version: 8.0.238 / Virus Database: 270.12.4/2082 - Release Date: 04/27/09
> 06:19:00
>

------------------------------------

Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/lexicographylist/

<*> Your email settings:
    Individual Email | Traditional

<*> To change settings online go to:
    http://groups.yahoo.com/group/lexicographylist/join
    (Yahoo! ID required)

<*> To change settings via email:
    mailto:lexicographylist-digest at yahoogroups.com 
    mailto:lexicographylist-fullfeatured at yahoogroups.com

<*> To unsubscribe from this group, send an email to:
    lexicographylist-unsubscribe at yahoogroups.com

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/