HPSG - Revision history

Gert: /* Words */

2017-04-15T13:14:18Z

Words

← Older revision		Revision as of 13:14, 15 April 2017
Line 54:		Line 54:
	With this background about the 3 valence attributes, look at the lexical entries for ''Lilly'' and ''Fido'' again. You see that both expressions are marked as		With this background about the 3 valence attributes, look at the lexical entries for ''Lilly'' and ''Fido'' again. You see that both expressions are marked as

	SUBJ <>		SUBJ <><br>
	SPR <>		SPR <><br>
	COMPS <>.		COMPS <>.

Gert: /* Words */

2017-04-15T13:13:30Z

Words

← Older revision		Revision as of 13:13, 15 April 2017
Line 60:		Line 60:
	From this we conclude that both words do not need to be combined with a subject, a specifier, or any complements in order to function as the subject or object of a sentence. And this is correct, as the two sentences below illustrate:		From this we conclude that both words do not need to be combined with a subject, a specifier, or any complements in order to function as the subject or object of a sentence. And this is correct, as the two sentences below illustrate:

	(1) <span style="color: blue>Lilly snores.</span></br>		(1) <span style="color: blue>Lilly snores.</span><br>
	(2) <span style="color: blue>I like Fido.</span>		(2) <span style="color: blue>I like Fido.</span>

	In (1), the word ''Lilly'' serves as the subject of the verb ''snores'' and can do so all by itself. In (2), the word ''Fido'' is the direct object complement of the transitive verb ''likes'' and again it can do so all by itself. Compare this with what happens, when we substitute a common noun for the names ''Lilly'' and ''Fido'' in the sentences above:		In (1), the word ''Lilly'' serves as the subject of the verb ''snores'' and can do so all by itself. In (2), the word ''Fido'' is the direct object complement of the transitive verb ''likes'' and again it can do so all by itself. Compare this with what happens, when we substitute a common noun for the names ''Lilly'' and ''Fido'' in the sentences above:

	(3) <span style="color: blue>*Student snores.</span></br>		(3) <span style="color: blue>*Student snores.</span><br>
	(4) <span style="color: blue>*I like cat.</span>		(4) <span style="color: blue>*I like cat.</span>

	Both sentences become ungrammatical! The reason is simple: as already mentioned above, common nouns like ''student'' or ''cat'' are [SPR <D>], which means that they first need to combine with a determiner in order to serve as the subject or object of a sentence. Thus, (3) and (4) can be made grammatical by putting determiners in front of the two common nouns:		Both sentences become ungrammatical! The reason is simple: as already mentioned above, common nouns like ''student'' or ''cat'' are [SPR <D>], which means that they first need to combine with a determiner in order to serve as the subject or object of a sentence. Thus, (3) and (4) can be made grammatical by putting determiners in front of the two common nouns:

	(5) <span style="color: blue>'''The''' student snores.</span></br>		(5) <span style="color: blue>'''The''' student snores.</span><br>
	(6) <span style="color: blue>I like '''any''' cat.</span>		(6) <span style="color: blue>I like '''any''' cat.</span>

Gert: /* Words */

2017-04-15T13:13:02Z

Words

← Older revision		Revision as of 13:13, 15 April 2017
Line 60:		Line 60:
	From this we conclude that both words do not need to be combined with a subject, a specifier, or any complements in order to function as the subject or object of a sentence. And this is correct, as the two sentences below illustrate:		From this we conclude that both words do not need to be combined with a subject, a specifier, or any complements in order to function as the subject or object of a sentence. And this is correct, as the two sentences below illustrate:

	(1) <span style="color: blue>Lilly snores.</span>		(1) <span style="color: blue>Lilly snores.</span></br>
	(2) <span style="color: blue>I like Fido.</span>		(2) <span style="color: blue>I like Fido.</span>

	In (1), the word ''Lilly'' serves as the subject of the verb ''snores'' and can do so all by itself. In (2), the word ''Fido'' is the direct object complement of the transitive verb ''likes'' and again it can do so all by itself. Compare this with what happens, when we substitute a common noun for the names ''Lilly'' and ''Fido'' in the sentences above:		In (1), the word ''Lilly'' serves as the subject of the verb ''snores'' and can do so all by itself. In (2), the word ''Fido'' is the direct object complement of the transitive verb ''likes'' and again it can do so all by itself. Compare this with what happens, when we substitute a common noun for the names ''Lilly'' and ''Fido'' in the sentences above:

	(3) <span style="color: blue>*Student snores.</span>		(3) <span style="color: blue>*Student snores.</span></br>
	(4) <span style="color: blue>*I like cat.</span>		(4) <span style="color: blue>*I like cat.</span>

	Both sentences become ungrammatical! The reason is simple: as already mentioned above, common nouns like ''student'' or ''cat'' are [SPR <D>], which means that they first need to combine with a determiner in order to serve as the subject or object of a sentence. Thus, (3) and (4) can be made grammatical by putting determiners in front of the two common nouns:		Both sentences become ungrammatical! The reason is simple: as already mentioned above, common nouns like ''student'' or ''cat'' are [SPR <D>], which means that they first need to combine with a determiner in order to serve as the subject or object of a sentence. Thus, (3) and (4) can be made grammatical by putting determiners in front of the two common nouns:

	(5) <span style="color: blue>'''The''' student snores.</span>		(5) <span style="color: blue>'''The''' student snores.</span></br>
	(6) <span style="color: blue>I like '''any''' cat.</span>		(6) <span style="color: blue>I like '''any''' cat.</span>

Gert: /* Words */

2017-04-15T13:11:48Z

Words

← Older revision		Revision as of 13:11, 15 April 2017
Line 51:		Line 51:

	which means that they must combine with a determiner like ''a'' or ''the'' in order to function as a subject or an object.		which means that they must combine with a determiner like ''a'' or ''the'' in order to function as a subject or an object.

			With this background about the 3 valence attributes, look at the lexical entries for ''Lilly'' and ''Fido'' again. You see that both expressions are marked as

			SUBJ <>
			SPR <>
			COMPS <>.

			From this we conclude that both words do not need to be combined with a subject, a specifier, or any complements in order to function as the subject or object of a sentence. And this is correct, as the two sentences below illustrate:

			(1) <span style="color: blue>Lilly snores.</span>
			(2) <span style="color: blue>I like Fido.</span>

			In (1), the word ''Lilly'' serves as the subject of the verb ''snores'' and can do so all by itself. In (2), the word ''Fido'' is the direct object complement of the transitive verb ''likes'' and again it can do so all by itself. Compare this with what happens, when we substitute a common noun for the names ''Lilly'' and ''Fido'' in the sentences above:

			(3) <span style="color: blue>*Student snores.</span>
			(4) <span style="color: blue>*I like cat.</span>

			Both sentences become ungrammatical! The reason is simple: as already mentioned above, common nouns like ''student'' or ''cat'' are [SPR <D>], which means that they first need to combine with a determiner in order to serve as the subject or object of a sentence. Thus, (3) and (4) can be made grammatical by putting determiners in front of the two common nouns:

			(5) <span style="color: blue>'''The''' student snores.</span>
			(6) <span style="color: blue>I like '''any''' cat.</span>

Gert: /* Words */

2017-04-15T12:59:00Z

Words

← Older revision		Revision as of 12:59, 15 April 2017
Line 42:		Line 42:
	means that the expression does not need and, in fact, is not allowed to combine with any complement. That is correct for intransitive verbs like ''appear'', since you cannot say such things as <span style="color: blue>*Lilly appears the cake.</span>.		means that the expression does not need and, in fact, is not allowed to combine with any complement. That is correct for intransitive verbs like ''appear'', since you cannot say such things as <span style="color: blue>*Lilly appears the cake.</span>.

	~~Besides~~		As we will see, words cannot only select complements, but also subjects and specifiers. This is what the two attributes SUBJ and SPR are for. When we get to verbs, we will see that they are marked as

			SUBJ <NP>,

			i.e. they must have a subject NP and so-called common nouns (i.e. ''cat'' or ''student'') are listed as

			SPR <D>,

			which means that they must combine with a determiner like ''a'' or ''the'' in order to function as a subject or an object.

Gert: /* Words */

2017-04-15T12:53:51Z

Words

← Older revision		Revision as of 12:53, 15 April 2017
Line 40:		Line 40:
	COMPS <>		COMPS <>

	means that the expression does not need and, in fact, is not allowed to combine with any complement.		means that the expression does not need and, in fact, is not allowed to combine with any complement. That is correct for intransitive verbs like ''appear'', since you cannot say such things as <span style="color: blue>*Lilly appears the cake.</span>.

			Besides

Gert: /* Words */

2017-04-15T12:50:00Z

Words

← Older revision		Revision as of 12:50, 15 April 2017
Line 30:		Line 30:
	The second piece of information in our list of properties that speakers know about the words in their language is their part of speech. Traditional grammar categorizes both words as nouns. In our formal theory this is represented as the blue <span style="color: blue>''n''</span> at the end of the path SYNSEM\|LOC\|CAT\|HEAD. Do not worry for now about the meanings of the elements in the path, those will be explained in due course!		The second piece of information in our list of properties that speakers know about the words in their language is their part of speech. Traditional grammar categorizes both words as nouns. In our formal theory this is represented as the blue <span style="color: blue>''n''</span> at the end of the path SYNSEM\|LOC\|CAT\|HEAD. Do not worry for now about the meanings of the elements in the path, those will be explained in due course!

	Next on our list comes '''valence''', i.e. the representation of the knowledge speakers have about what other kinds of constituents the word needs to combine with. You will remember from traditional grammar the distinction between '''intransitive''' and '''transitive''' verbs. These are just names for those verbs, respectively, which do not need a direct object (i.e. the verbs ''appear'' and ''cough'') and those which do (like '''~~have~~''' ~~and~~ '''~~trust~~''').		Next on our list comes '''valence''', i.e. the representation of the knowledge speakers have about what other kinds of constituents the word needs to combine with. You will remember from traditional grammar the distinction between '''intransitive''' and '''transitive''' verbs. These are just names for those verbs, respectively, which do not need a direct object (i.e. the verbs ''appear'' and ''cough'') and those which do (like ''have'' and ''trust''). Objects are called '''complements''' in our theory; so, to express that the verb ''have'' needs a direct object, its representation would contain the line

			COMPS <NP>,

			which translated into people speech means that the expression needs one and only one complement and that the part of speech of this complement needs to be NP.

			Correspondingly,

			COMPS <>

			means that the expression does not need and, in fact, is not allowed to combine with any complement.

Gert: /* Words */

2017-04-15T12:37:21Z

Words

← Older revision		Revision as of 12:37, 15 April 2017
Line 29:		Line 29:

	The second piece of information in our list of properties that speakers know about the words in their language is their part of speech. Traditional grammar categorizes both words as nouns. In our formal theory this is represented as the blue <span style="color: blue>''n''</span> at the end of the path SYNSEM\|LOC\|CAT\|HEAD. Do not worry for now about the meanings of the elements in the path, those will be explained in due course!		The second piece of information in our list of properties that speakers know about the words in their language is their part of speech. Traditional grammar categorizes both words as nouns. In our formal theory this is represented as the blue <span style="color: blue>''n''</span> at the end of the path SYNSEM\|LOC\|CAT\|HEAD. Do not worry for now about the meanings of the elements in the path, those will be explained in due course!

			Next on our list comes '''valence''', i.e. the representation of the knowledge speakers have about what other kinds of constituents the word needs to combine with. You will remember from traditional grammar the distinction between '''intransitive''' and '''transitive''' verbs. These are just names for those verbs, respectively, which do not need a direct object (i.e. the verbs ''appear'' and ''cough'') and those which do (like '''have''' and '''trust''').

Gert: /* Words */

2017-04-15T12:28:00Z

Words

← Older revision		Revision as of 12:28, 15 April 2017
Line 28:		Line 28:
	Returning to our two words above, we have already found 3 things that they have in common and one where they differ: (i) both are marked as ''word'', (ii) both have a PHON attribute, expressing that they have a phonology, and (iii) in both cases the value of this attribute is a list of "segments", in truth letters. The difference that we found is that the PHON value is <Lilly> in one case and <Fido> in the other.		Returning to our two words above, we have already found 3 things that they have in common and one where they differ: (i) both are marked as ''word'', (ii) both have a PHON attribute, expressing that they have a phonology, and (iii) in both cases the value of this attribute is a list of "segments", in truth letters. The difference that we found is that the PHON value is <Lilly> in one case and <Fido> in the other.

	The second piece of information in our list of properties that speakers know about the words in their language is their part of speech. Traditional grammar categorizes both words as nouns. In our formal theory this is represented as the blue <span style="color: blue>''n''</span> at the end of the path SYNSEM\|LOC\|CAT\|HEAD. Do not worry for now about the meanings of the elements in the path, those will be explained in due course.		The second piece of information in our list of properties that speakers know about the words in their language is their part of speech. Traditional grammar categorizes both words as nouns. In our formal theory this is represented as the blue <span style="color: blue>''n''</span> at the end of the path SYNSEM\|LOC\|CAT\|HEAD. Do not worry for now about the meanings of the elements in the path, those will be explained in due course!

Gert: /* Words */

2017-04-15T12:27:12Z

Words

← Older revision		Revision as of 12:27, 15 April 2017
Line 28:		Line 28:
	Returning to our two words above, we have already found 3 things that they have in common and one where they differ: (i) both are marked as ''word'', (ii) both have a PHON attribute, expressing that they have a phonology, and (iii) in both cases the value of this attribute is a list of "segments", in truth letters. The difference that we found is that the PHON value is <Lilly> in one case and <Fido> in the other.		Returning to our two words above, we have already found 3 things that they have in common and one where they differ: (i) both are marked as ''word'', (ii) both have a PHON attribute, expressing that they have a phonology, and (iii) in both cases the value of this attribute is a list of "segments", in truth letters. The difference that we found is that the PHON value is <Lilly> in one case and <Fido> in the other.

	The second piece of information in our list of properties that speakers know about the words in their language is their part of speech. Traditional grammar categorizes both words as nouns. In our formal theory this is represented as the ~~green~~ <span style="color: blue>n</span> at the end of the path SYNSEM\|LOC\|CAT\|HEAD. Do not worry for now about the meanings of the elements in the path, those will be explained in due course.		The second piece of information in our list of properties that speakers know about the words in their language is their part of speech. Traditional grammar categorizes both words as nouns. In our formal theory this is represented as the blue <span style="color: blue>''n''</span> at the end of the path SYNSEM\|LOC\|CAT\|HEAD. Do not worry for now about the meanings of the elements in the path, those will be explained in due course.