University of Southern Denmark
Portuguese ->  Comments   Visual Interactive Syntax Learning  
 
Portuguese VISLSentence AnalysisEdutainmentCorporaDictionariesMachine Translation

Printer-friendly version

  C6-8:    A instituição deste direito só será, contudo, efectivado depois do lançamento nacional do projecto dentro de pouco mais de um mês.
 

This sentence presents two possible alternative syntactic analyses, as it is ambiguous with regard to the adjunct adverbial "dentro de pouco mais de um mês" is concerned. It is a PP attachment ambiguity where  the parser would need to transcend the local context for safe disambiguation,  producing therefore just one of the possible analyses. The alternative analysis was added manually afterwards.
One of the analyses, as can be seen below (T1), considers the referred adjunct adverbial as being at the same constituency level as the predicate "efectivado".
 
 

(T1)    A instituição deste direito só será, contudo, efectivado depois do lançamento nacional do projecto dentro de pouco mais de um mês.

A1
STA:fcl
(...)

ADVL:adv só
P:vp-
=AUX:v-fin(FUT 3S IND) será
,
ADVL:adv(<kc>) contudo
,
-P:vp
=MV:v-pcp(M S) efectivado
ADVL:advp
=H:adv depois
=A<:pp
==H:prp(<sam->) de
==P<:np
===>N:art(<-sam> M S) o
===H:n(M S) lançamento
===N<:adj(M S) nacional
===N<:pp
====H:prp(<sam->) de
====P<:np
=====>N:art(<-sam> M S) o
=====H:n(M S) projecto
===N<:pp
=H:prp dentro_de
=P<:np
==>N:adjp
===>A:advp
====>A:pron-det(<quant> M S) pouco
====H:adv(<quant>) mais_de
===H:num(<card> M S) um
==H:n(M S) mês
.

In other words, the adverbial is attached to the predicate, which can be seen by placing the adverbial in another position, but at the same constituency level (bearing in mind that it is an ad- verbal constituent), as in (T2):

       (T2)    A instituicão deste direito só será, contudo, efectivado, dentro de pouco mais de um mês, depois do lançamento nacional do projecto.

                    TREE:

                    (...)

                    ADVL:adv só
        P:vp-
                    =AUX:v-fin(FUT 3S IND) será
                    ,
                    ADVL:adv(<kc>) contudo
                    ,
                    -P:vp
                    =MV:v-pcp(M S) efectivado
        ADVL:pp
                    =H:prp dentro_de
                    =P<:np
                    ==>N:adjp
                    ===>A:advp
                    ====>A:pron-det(<quant> M S) pouco
                    ====H:adv(<quant>) mais_de
                    ===H:num(<card> M S) um
                    ==H:n(M S) mês
                    ADVL:advp
                    =H:adv depois
                    =A<:pp
                    ==H:prp(<sam->) de
                    ==P<:np
                    ===>N:art(<-sam> M S) o
                    ===H:n(M S) lançamento
                    ===N<:adj(M S) nacional
                    ====H:prp(<sam->) de
                    ====P<:np
                    =====>N:art(<-sam> M S) o
                    =====H:n(M S) projecto
                    .
 

However, and this is where ambiguity rises,  the adverbial might have another reading if in the original position (T1). Semantically speaking, the adverbial could be attached to the adverbial immediately preceding it ("depois do lançamento nacional do projecto"), which syntactically would mean lowering the constituency level of the adverbial (T3).

(T3)    A instituição deste direito só será, contudo, efectivado depois do lançamento nacional do projecto dentro de pouco mais de um mês.

A2
STA:fcl
(...)

ADVL:adv só
P:vp-
=AUX: v-fin(FUT 3S IND) será
,
ADVL:adv(<kc>) contudo
,
-P:vp
=MV:v-pcp(M S) efectivado
ADVL:advp
=H:adv depois
=A<:pp
==H:prp(<sam->) de
==P<:np
===>N:art(<-sam> M S) o
===H:n(M S) lançamento
===N<:adj(M S) nacional
===N<:pp
====H:prp(<sam->) de
====P<:np
=====>N:art(<-sam> M S) o
=====H:n(M S) projecto
===N<:pp
====H:prp dentro_de
====P<:np
=====>N:adjp
======>A:advp
=======>A:pron-det(<quant> M S) pouco
=======H:adv(<quant>) mais_de
======H:num(<card> M S) um
=====H:n(M S) mês
.

In this case, the adverbial , however roughly, can be moved  to another position within the same constituency level (T4), but not to any other position in the sentence, as it is now a dependent of the previous adverbial.

(T4)    (?) A instituição deste direito só será, contudo, efectivado depois do lançamento nacional, dentro de pouco mais de um mês, do projecto.

            TREE:

                    (...)

                    ADVL:adv só
                    P:vp-
                    =AUX: v-fin(FUT 3S IND) será
                    ,
                    ADVL:adv(<kc>) contudo
                    ,
                    -P:vp
                    =MV:v-pcp(M S) efectivado
                    ADVL:advp
                    =H:adv depois
                    =A<:pp
                    ==H:prp(<sam->) de
                    ==P<:np
        ===N:art(<-sam> M S) o
               ===H:n(M S) lançamento
               ===N<:adj(M S) nacional
               ===N<:pp
                    ====H:prp dentro_de
                    ====P<:np
                    =====>N:adjp
                    ======>A:advp
                    =======>A:pron-det(<quant> M S) pouco
                    =======H:adv(<quant>) mais_de
                    ======H:num(<card> M S) um
                    =====H:n(M S) mês
        ===N<:pp
                    ====H:prp(<sam->) de
                    ====P<:np
                    =====>N:art(<-sam> M S) o
                    =====H:n(M S) projecto
                    .

Note that there is a lack of agreement between subject and verb. efectivado agrees with direito which is not the subject. This agreement error is more common when the distance between the subject and the verb is bigger, tending the verb, though, to agree with closest element.

C7-2:    Porque é a embraiagem que põe o motor em contacto com as rodas que geram o movimento, que para nós é o desenvolvimento.
 

One can identify four clauses in this sentence: a subordinate clause and three relative clauses headed by the relative pronoun que. Because they are relative, the pronoun refers to something that appeared previously in the sentence. There are several tests that can be applied in order to disambiguate what the pronoun refers to. A simple test is just to check verb-subject agreement. So, in the first relative clause

         que põe o motor a trabalhar   ,
 

the relative pronoun is the subject and it refers to the previous noun "embraiagem", bearing in mind the verb inflection (3 rd person singular).
A similar argument holds for the second relative clause:

         que geram o movimento    ,

where the verb inflection (3rd person plural) points unmistakably to "rodas".
If you now look at the third relative clause and apply the same agreement test, you may come up with the same result: que, being the subject which relates to "movimento" according to the verb inflection.

As it can be seen below (T1), the relative clauses are attached to the immediate preceding noun (@N<), heading the np's:

(T1)    Porque é a embraiagem que põe o motor em contacto com as rodas que geram o movimento, que para nós é o desenvolvimento.

A2
ADVL:fcl
SUB:conj-s Porque
P:v-fin(PR 3S IND) é
SC:np
=>N:art(F S) a
=H:n(F S) embraiagem
=N<:fcl
==SUBJ:pron-indp(<rel> M/F S/P) que
==P:v-fin(PR 3S IND) põe
==ACC:np
===>N:art(M S) o
===H:n(M S) motor
==ADV:pp
===H:prp em
===P<:np
====H:n(M S) contacto
====N<:p
=====H:prp com
=====P<:np
======>N:art(F P) as
======H:n(F P) rodas
======N<:fcl
=======SUBJ:pron-indp(<rel> M/F S/P) que
=======P:v-fin(PR 3P IND) geram
=======ACC:np
========>N:art(M S) o
========H:n(M S) movimento
========,
========N<PRED:fcl
=========SUBJ:pron-indp(<rel> M S) que
=========ADVL:pp
==========H:prp para
==========P<:pron-pers(M/F 1P NOM/PIV) nós
=========P:v-fin(PR 3S IND) é
=========SC:np
==========>N:art(M S) o
==========H:n(M S) desenvolvimento
.

However, an attachment ambiguity is observed in this sentence. Semantically speaking, it would make sense to consider that the relative pronoun, in the third relative clause could be referring not  to the whole content of the previous proposition, but to the focus of that period, an analysis which is also reinforced by the presence of the punctuation mark (comma), as can be seen in (S1). In this case, the realtive clause would constitute a comment on the previous proposition.

(S1)    (Porque é a embraiagem ((que põe o motor em contacto com as rodas) que geram o movimento)), que para nós é o desenvolvimento.

Furthermore, if the all period is to be pronominalised , the latter analysis becomes clearer (S2):

(S2)    Porque é a embraiagem que põe o motor em contacto com as rodas que geram o movimento, o que para nós é o desenvolvimento.

 The demonstrative pronoun (o) resumes "Porque é a embraiagem que põe o motor em contacto com as rodas que geram o movimento".
In fact, the parser more or less expects o=que in these cases, tagging it as sentence apposition @S<.

Applying this test to any of the two other relative clauses (S3) (S4), it will fail, leading to the conclusion that the first two relative pronouns refer to the previous noun.

(S3)    *Porque é a embraiagem o que põe o motor em contacto com as rodas...
(S4)    *Porque é a embraiagem que põe o motor em contacto com as rodas o que geram o movimento...

Instead of using the pronoun as resuming the all period, it is also possible to use any other word that somehow stands for it (N + que), for instance, in this case (S5), the noun "metáfora" amongst others:

(S5)    Porque é a embraiagem que põe o motor em contacto com as rodas que geram o movimento, metáfora (engrenagem/ situação/ coisa) que para nós é o desenvolvimento.

or a construction like N + DET + que:

 (S5')    Porque é a embraiagem que põe o motor em contacto com as rodas que geram o movimento, metáfora (engrenagem/ situação) essa que para nós é o desenvolvimento.

Note that in all cases (S2), (S5) and (S5') there is now a clear subject-verb agreement.
Therefore, and back to syntactic trees, the generated tree, in this case, would present the last relative clause placed at the first constituency level, as it can be seen in (T2):

(T2)    Porque é a embraiagem que põe o motor em contacto com as rodas que geram o movimento, que para nós é o desenvolvimento.

A1
ADVL:fcl
SUB:conj-s Porque
P:v-fin(PR 3S IND) é
SC:np
=>N:art(F S) a
=H:n(F S) embraiagem
=N<:fcl
==SUBJ:pron-indp(<rel> M/F S/P) que
==P:v-fin(PR 3S IND) põe
==ACC:np
===>N:art(M S) o
===H:n(M S) motor
==ADV:pp
===H:prp em
===P<:np
====H:n(M S) contacto
====N<:pp
=====H:prp com
=====P<:np
======>N:art(F P) as
======H:n(F P) rodas
======N<:fcl
=======SUBJ:pron-indp(<rel> M/F S/P) que
=======P:v-fin(PR 3P IND) geram
=======ACC:np
========>N:art(M S) o
========H:n(M S) movimento
,
S<:fcl
=SUBJ:pron-indp(<rel> M S) que
=ADVL:pp
==H:prp para
==P<:pron-pers(M/F 1P NOM/PIV) nós
=P:v-fin(PR 3S IND) é
=SC:np
==>N:art(M S) o
==H:n(M S) desenvolvimento
.
As an example, please have a look at sentence  C15-3  , where the relative pronoun refers to all of the previous period, by pronominal substitution.
 

C12-3:    Há quem defenda, no entanto, que se trata de um fax apócrifo, realmente escrito por o deputado José Magalhães, o qual teria, aliás, imitado a letra do ex-deputado António Barreto.
 

Before going into the syntax of the verb haver in this particular sentence, it is important to be aware of the forms it takes.  The verb haver has two different semantic forms which interferes with the verb inflection. Haver may semantically mean existir (exist) or it may work just as an auxiliary verb, that can be replaced by ter. Table 1, sums up the semantic form and its implications in the verb inflection:
 

Table 1: Verb haver, semantic form vs. verb inflection
Semantic form
Verb inflection
          Haver= existir        3rd person singular form (unique form, independently of the person, number), 
       varying according to different tenses:
 
    TENSE
 INDICATIVE
SUBJUNCTIVE
               Present Simple                     Há                   Haja
               Past Simple                     Houve                      -
               Imperfect                     Havia                   Houvesse
              "Mais que Perfeito"                     Houvera                      -
               Future Simple                     Haverá                    Houver
          Haver, auxiliary verb ( =ter)        Inflection bearing in mind person, number and tense (all forms present). 
      Always followed by the Past Participle of the main verb or que (in há que fazer)

Let's focus now on the verb haver, meaning existir. Consider the following examples:

(Ex. 1) Houve guerras mundiais no passado, mas agora paz.
(Ex. 2) *Haviam divergências quanto a ideologias entre os países da Europa, o que provocou uma enorme tensão diplomática.
 

In (Ex. 1),  houve, by its non-inflecting characteristics (houve: 3rd person singular; guerras: noun , feminine, plural), stands for existir. Furthermore, the replacement is possible, however the replacement does inflect, which is explained below by the syntactic differences between the constructions with haver and existir, respectively:

(Ex. 1') Existiam muitas guerras mundiais no passado, mas agora existe paz.

On the other hand, (Ex. 2) is not acceptable when the existir meaning of haver is concerned, as the verb presents inflection in person and number.  Haver, therefore, would be expected to be followed by the Past Participle of a main verb (Ex. 2')

(Ex. 2') Os países da Europa haviam divergido quanto a ideologias, o que provocou uma enorme tensão diplomática.

Here, the auxiliary verb replacement by existir is, then, innacurate (Ex. 2''), but works for ter (Ex. 2'''):

(Ex. 2'') *Os países da Europa existiam divergido quanto a ideologias, o que provocou uma enorme tensão diplomática.
(Ex. 2''')  Os países da Europa tinham divergido quanto a ideologias, o que provocou uma enorme tensão diplomática.

Applying these considerations to the initial sentence (C12-3), quem defenda, no entanto, que se trata de um fax apócrifo,......, and realising that the verb haver is not followed by the Past Participle of another verb, it must stand for the semantic meaning of existir:

(S1) Existe quem defenda, no entanto, que se trata de um fax apócrifo, realmente escrito por o deputado José Magalhães, o qual teria, aliás, imitado a letra do  ex-deputado     António Barreto.

Moving to the syntactic analysis, existe is the finite main verb, and the subject is the whole period "quem defenda, no entanto, que se trata de um fax apócrifo, ". However, the semantic substitution cannot be applied to the syntactic substitution. In other words,  the verb existir is intransitive, therefore, "quem defenda, no entanto, que se trata de um fax apócrifo, realmente escrito por o deputado José Magalhães, o qual teria, aliás, imitado a letra do ex-deputado António Barreto." has to be the subject.
The same is not true for haver. The verb is transitive, so it requires a direct object, but there is no subject in this type of construction (haver= existir).

A1
STA:fcl
P:v-fin(PR 3S IND) Há
ACC:fcl
=SUBJ:pron-indp(<rel> M/F S/P) quem
=P:v-fin(PR 3S SUBJ) defenda
=,
=ADVL:adv(<kc>) no_entanto
=,
=ACC:fcl
==SUB:conj-s que
==ACC:pron-pers(<refl> M/F 3S/P ACC) se
==P:v-fin(PR 3S IND) trata
==PIV:pp
===H:prp de
===P<:np
====>N:art(<arti> M S) um
====H:n(M S) fax
====N<:adj(M S) apócrifo
==,
==>A:adv realmente
==N<PRED:icl(<pcp>)
===P:v-pcp(<ap> M S) escrito
===PASS:pp(<postmod>)
====H:prp(<sam->) por
====P<:np
=====>N:art(<-sam> M S) o
=====H:n(M S) deputado
=====N<:prop(M S) José_Magalhães
=====,
=====N<PRED:fcl
==SUBJ:pron-indp(<rel> M S) o_qual
==P:vp-
===AUX:v-fin(COND 3S) teria
==,
==ADVL:adv(<kc>) aliás
==,
==-P:vp
===MV:v-pcp(M S) imitado
==ACC:np
===>N:art(F S) a
===H:n(F S) letra
===N<:pp
====H:prp(<sam->) de
====P<:np
=====>N:art(<-sam> M S) o
=====H:n(M S) ex-deputado
=====N<:prop(M S) António_Barreto
.

In other constructions, haver as an auxiliary (= ter), does possess a subject, because it is the main verb that governs the valency, not the auxiliary. In (Ex. 2'),

(Ex. 2') Os países da Europa haviam divergido quanto a ideologias, o que provocou uma enorme tensão diplomática    ,

 the subject  is the noun phrase, "Os países da Europa". The sentence does not present a direct object because the main verb divergir is not a transitive verb.  Here's another example where the main verb is transitive:

                a) Os países haviam quebrado o princípio básico da diplomacia.

         SUBJ:np ----   Os países
         FAUX      ----   haviam
         FMV       ----   quebrado
         ACC:np   ----   o princípio básico da diplomacia
 

A good test to check the valency of the verb haver, meaning existir, is pronoun substitution (PS). The direct object would make it possible to be replaced by a pronoun:

(Ex. 1) Houve guerras mundiais no passado, mas agora paz.
 PS:    Houve-as no passado, mas agora há-a.                        (as, standing for guerras mundiais;  a, standing for paz)

C12-3: Há quem defenda, no entanto, que se trata de um fax apócrifo, realmente escrito por o deputado José Magalhães, o qual teria, aliás, imitado a letra do ex-deputado António Barreto.
PS:       Há-os.            (os, standing for  "quem defenda, no entanto, que se trata de um fax apócrifo, realmente escrito por o
                                       deputado José Magalhães, o qual teria, aliás, imitado a letra do ex-deputado António Barreto.")
 

The lack of inflection itself, meaning no subject-verb agreement, might also indicate that there is no subject when haver semantically means existir.
Other examples in the sentence corpus:  C5-10

There are other constructions with the verb haver, where, despite the fact that it does not mean existir semantically, it is also impersonal (no subject) and it takes just a unique form (3rd person singular). That's the case when haver stands for time duration. See examples in  C9-1 and C21-1
 
 

C9-1:    Co-produção franco-egípcia, O Emigrante inspira-se na história de José, filho de Jacob, contando o percurso de Ram que, há 3000 anos, decide abandonar a sua terra árida para se instalar no Egipto dos faraós, centro da civilização.

AND

C21-1:    O russo será uma das seis línguas principais usadas por João Paulo II amanhã e depois, no Encontro Mundial da Juventude, no Santuário Mariano de Czestokowa, onde se prevê a presença de um milhão de jovens, o dobro dos que há dois anos se congregaram em Santiago de Compostela.

In this case, the verb haver has another semantic meaning, standing for time that has gone through. In such cases, the PALAVRAS, in terms of tagging found a solution:  Há dois anos, as in the above sentence is an ADVL, with há as the H ead of the adverbial. Typically, Adverbials are prepositional phrases, being the head, typically a preposition. Therefore, the option of considering há as a preposition was taken, bearing in mind that, functionally, it presents the same behaviour: invariable word, requiring a complement.
However, there is no tradition or of having a form of a verb (haver) classified as a preposition and so, other possibilities could be implemented:

(1) ADVERBIAL (finite verb as the head of the prepositional phrase)

ADVL:pp
=H:v-fin   
=P<:np
==>N:num    dois
==H:n    anos

(2) ADVERBIAL CLAUSE

ADVL:fcl
=MV:v-fin    há
=ACC:np
==>N:num    dois
==H:n    anos

(2) would be very hard on the parser but (1) it can figure out by ways of (1a), with a word based filtration in the end.

(1a) ADVERBIAL (verb word class filtered into a preposition)

ADVL:pp
=H:prp    
=P<:np
==>N:num    dois
==H:n    anos
 

Click here to the syntactic discussion on the verb haver as semantically meaning time gone through in constructions like: Há <time length> que.....
 
 

C5-10:    A proposta é muito bem vista, porque será mais vantajosa do que se houver só um pólo de interesse no local, afirmou o major Carlos Barbosa, das relações públicas da Força Aérea, admitindo que, com o parque temático, se o interesse for diversificado, toda a gente fica a ganhar.
 

If the basic principles of the syntax of the verb haver are followed, it is clear that houver (future simple, subjunctive) in this case is semantically similar to existir, therefore, it is impersonal (no subject) and transitive, which means that "só um pólo de interesse" would then be the direct object, as it can be seen from the constituent tree:
 

TREE:

A1
STA:fcl
SUBJ:np
=>N:art(F S) A
=H:n(F S) proposta
P:vp-
=AUX:v-fin(PR 3S IND) é
>A:advp
=>A:adv(<quant>) muito
=H:adv(<quant>) bem
-P:vp
=MV:v-pcp(F S) vista
,
ADVL:fcl
=SUB:conj-s porque
=P:v-fin(FUT 3S IND) será
=SC:ap
==>A:adv(<quant>) mais
==H:adj(F S) vantajosa
==KOMP<:acl
===COM:conj-s do_que
===AS<:fcl
===SUB:conj-s se
====P:v-fin(FUT 3S/P SUBJ) houver
====ACC:np
=====>N:adv só
=====>N:art(<arti> M S) um
=====H:n(M S) pólo
=====N<:pp
======H:prp de
======P<:n(M S) interesse
====ADVL:pp
=====H:prp(<sam->) em
=====P<:np
======>N:art(<-sam> M S) o
======H:n(M S) local
===,
(...)
 

C52-1:  Há muito tempo que tenho uma estranha relação afectuosa com esta ilha.

The syntactic analysis of sentences of this structure: Há <time length> que...., involves complex issues.  The first one, connected to the morphologic category of "há" was discussed previously (C9-1 and C21-1).
The other question is related to the analysis of the (hypothetical) clause: que tenho uma estranha relação afectuosa com esta ilha.
Let's start by looking at the parser's output of this sentence:

[há] <*> PRP @ADVL                 [haver] <*> <vUK> <ink> <fmc> V PR 3S IND VFIN @FMV

muito=tempo [muito=tempo] <am> <dur> <temp> N M S @P<

que [que] <rel> <que-n> SPEC M/F S/P @ACC> @#FS-N<

tenho [ter] <vt> <ink> V PR 1S IND VFIN @FMV

uma [um] <arti> DET F S @>N

estranha [estranho] <jh> ADJ F S @>N

relação [relação] <+entre> N F S @<ACC

estreita [estreito] <n> ADJ F S @N<

com [com] PRP @<ADVL

esta [este] <dem> DET F S @>N

ilha [ilha] <top> N F S @P<
.

Both prepositional reading and verbal reading of the verb haver are considered. However, the sentence analysis is based on the prepositional reading (muito tempo: @P< but there's no @<ACC alternative).
The real problem arises when it comes to analyse "que" as a relative pronoun initiating a relative clause.

Let's then look at the possibilities for the analysis of "que":

1. relative pronoun:

    If so, then it may refer to the previous noun (as considered by the parser) or noun phrase or sentence (@S<)- sentence apposition.
However, see the agrammaticality of the following sentences:

 (1a) *O que tenho uma estranha relação afectuosa com esta ilha? Muito tempo.
 (1b) * Há muito tempo. Muito tempo tenho uma estranha relação afectuosa com esta ilha.

2. Subordinating conjunction: "que "

The subordinating conjunctions initialise a suborninated clause, which must have a syntactic function in relation to the main clause, like "Ela disse que tinha uma relação estreita com a ilha", que tinha uma relação estreita com a ilha is the complement of the verb dizer, main verb of the main clause Ela disse.
Bearing in mind the syntax of the verb haver (C12-3) and the uniqueness principle of the direct object:

    2.1. "Há muito tempo" would be the main clause (and the verbal reading of haver the only possible reading);
    2.2. "que tenho uma estranha relação afectuosa com esta ilha", a subordinated clause.

[haver]<*> <vUK> <ink> <fmc> V PR 3S IND VFIN @FMV

muito=tempo [muito=tempo] @<ACC

que [que]    KS @SUB  @#FS-< ?

tenho [ter] <vt> <ink> V PR 1S IND VFIN @FMV

uma [um] <arti> DET F S @>N

estranha [estranho] <jh> ADJ F S @>N

relação [relação] <+entre> N F S @<ACC

estreita [estreito] <n> ADJ F S @N<

com [com] PRP @<ADVL

esta [este] <dem> DET F S @>N

ilha [ilha] <top> N F S @P<
.

Which external function would the subordinate clause hold? It cannot be subject (@SUBJ) as haver is an impersonal verb:

(2.2.a) *Que tenho uma estranha relação afectuosa com esta ilha há muito tempo.

The argument that the verb obligatorily selects is the direct object (@ACC) which follows the uniqueness principle (a sentence can only present one direct object). That leads to hypothesise about which the direct object of the main verb is: muito tempo or the "subordinate" clause, and it is not clear. Pronominalising the direct object(s):

(2.2.b) (?) Há-o que tenho uma estranha relação afectuosa com esta ilha.
(2.2.c) (?) Há muito tempo isto   but    * Há isto muito tempo, (isto standing for the "que-clause")

which indicates that "muito tempo" alone cannot function as an @ADVL (like other time adverbials like ontem, hoje, and so on) and, therefore, cannot be separated from the verb haver. This might be a good argument to consider muito tempo the argument selected by haver.

Furthermore, if the "que-clause" was an object clause, being "que" the conjuction, then "que" could not be omitted which does not apply to this particular example:

(2.2.d) Tenho uma estranha relação afectuosa com esta ilha há muito tempo.

If "há muito tempo" is mantained in the beginning of the sentence, agrammaticality arises:

(2.2.e) * Há muito tempo tenho uma estranha relação afectuosa com esta ilha

but

(2.2.f) Há muito tempo tinha / tive uma relação afectuosa com esta ilha.

These changes in the verb carry a difference in semantics. Apparently, "que" does not held a relevant syntactic function to the sentence as a whole, being a dummy operator, like in the following sentences:

(S1) Que tenhas um bom-dia!
(S2) Desde 1990 que estava na mesa a reformulação das "secretas". (source CETEMPúblico)

In (S2), que is not obligatory:  Desde 1990 estava na mesa a reformulação das "secretas".
 

Despite that fact, "que" seems to have an interference with the meaning.  Let's examine some examples:

(Q1)  De facto, há muitos anos que há quem proponha a utilização de pequenas bombas nucleares em certas grandes obras de engenharia como terraplanagens de montanhas ou abertura de albufeiras.(source CETEMPúblico)

(Q2)  Isso poderia revestir-se de particular gravidade se o ataque ocorresse dentro de Angola, onde há poucos dias Ø  dois padres e uma freira foram mortos num ataque de «bandidos» . (source CETEMPúblico)

The difference between the two sentences is in the time that is referred. That is, while in (Q1), there is a clear continuity, in (Q2) that continuity does not exist for two reasons:
        1. the verb tense is fixed:  *...., onde há poucos dias dois padres e uma freira são mortos
                                                        (?) ..., onde há poucos dias dois padres e uma freira eram mortos (possible, if describing the act of murdering)

           2. The verb itself (matar) does not allow that continuity.

Therefore, in the (Q1) the time reference is a period of time (from a long time ago until now, and possibly is not finished), while in (Q2) the time reference is a specific moment in time (some days ago).
The above can be schematised in the following diagrams:

(Q1)
                                                                 ano x-n         "Há muitos anos que..."                             ano x
                                ---------------|---------------------------|------------>
                                                                   "proposta"                                                                     "momento da ilocução"           t
 

(Q2)

                                                                    dia x-n                                                                                dia x
                                ---------------|---------------------------|--------------->
                                                             "....foram mortos"                                                          "momento da ilocução"               t
 
 

The fact that "que" can be  a mark for two different time references is corroborated by applying the questions "Quando?" (When?) and "Desde quando?" (since when?):

(Q'1) Desde quando há quem proponha a utilização de bombas nucleares.....?
            Há muito tempo que há quem proponha.....

         but

            *Quando há quem proponha a utilização de bombas nucleares ...?
            *Há muito tempo que há quem proponha....

(Q'2) Quando foram mortos dois padres e uma freira?
            Há poucos dias foram mortos....

         but

            *Desde quando foram mortos dois padres e uma freira?
            *Há poucos dias foram mortos.

Furthermore, if we invert the order, one can realise that the meaning does not change in (Q2) but it might become ambiguous in (Q2):

(Q''1) Há quem proponha a utilização de bombas nucleares há muito tempo.

The verb in the present tense makes the non-continuity reading impossible. But if the tense is changed the semantic ambiguity arises:

(T1) Há muito tempo que havia quem propusesse a utilização de bombas nucleares... (continuity in a period of time, clearly)

(T2) Havia quem propusesse a utilização de bombas nucleares (...) há muito tempo (unclear: both questions "Quando?" and "Desde quando?" can be applied).

(Q''2) ....dois padres e uma freira foram mortos há poucos dias. (no change in meaning).
 

Summing up, although there is a clear difference in meaning by the use or not of the "que", it doesn't seem to have any specific function syntatically, though. Therefore, one could argue that its use indicates that somehow there is a topic construction- the @ADVL "Há muito tempo " is the known information and the rest of the sentence the new information-  and, for that reason it ought to be tagged as @TOP.
 
 

C15-3:    Sou um leitor assíduo e atento do Público, desde o primeiro número, e não poucas vezes tenho manifestado opiniões nas suas páginas, o que me leva agora a emitir o meu pensamento sobre o jornal, sobre quem o faz e sobre os ditos comunicados.
 

At first this sentence may also look ambiguous. It is in a strictly syntactic surface analysis point of view. This means that the period before the relative clause is a coordenation of two clauses, which makes it difficult to know exactly what the demonstrative pronoun "o" refers to; it could  be referring to the focuses of all the previous period (ser um leitor assíduo e atento + não poucas vezes ter manifestado opiniões) or just to the copulative coordinated clause ("e não poucas vezes tenho manifestado opiniões nas suas páginas": focus- não poucas vezes ter manisfestado opiniões). But as it was referred above, the ambiguity is strictly syntactic, because it is possible to disambiguate it, if the analysis is conjoined with semantics.

It would seem awckward to consider the demonstrative pronoun as referring to the focus of the copulative coordinated clause only. In a semantic point of view, it is difficult to accept that the fact that this reader had written several times to the newspaper before would constitute the reason why he is writing now. This is more clearly seen if the coordinated clause is omitted (S1):

(S1) Não poucas vezes tenho manifestado opiniões nas suas páginas, o que me leva agora a emitir o meu pensamento sobre o jornal, sobre quem o faz e sobre os ditos
         comunicados.

It works syntactically, it is a perfectly well formed sentence, but in terms of meaning some reservations must be made.
Furthermore, the punctuation (commas) seem to indicate that ", e não poucas vezes tenho manifestado opiniões nas suas páginas," is an accessory information, a comment on the coordinated clause, which could lead to the reading that the pronoun refers to the focuses of all the period (both coordinated clauses' focuses).

But there is a stronger argument that is related to the presence of the adverbials "desde o primeiro número" and "agora".  Somehow they are connected in order to establish a meaning. In other words, the reader has payed attention to this newspaper since it was first published (ADVL: desde o primeiro número) and this fact, together with the frequent writing, makes him legitimate to express his opinions in the present moment (ADVL: agora).  If the copulative coordinated clause is bracketted for a moment, this combined adverbial reading is made clearer (S2):

(S2) Sou um leitor assíduo e atento do Público desde o primeiro número,  (...) o que me leva agora a emitir o meu pensamento sobre o jornal, sobre quem o faz e sobre os ditos comunicados.       (= desde sempre)
 

This is a good example of how a surface-syntactic ally ambiguous sentence can be disambiguated by loking at the semantics. So, combining the two points of view, it is suggested that the reading that considers that the pronoun replaces all the period is preferred. Syntactically, the constituent tree would present the relative clause at the same constituency level as the compound unit level (which comprehends both coordinate and copulative coordinate clauses), as it can be seen in (T1):

(T1)

A1
STA:cu
CJT:fcl
=P:v-fin(PR 1S IND) Sou
=SC:np
==>N:art(<arti> M S) um
==H:n(M S) leitor
==N<:cu
===CJT:adj(M S) assíduo
===CO:conj-c(<co-postnom>) e
===CJT:adj(M S) atento
===N<:pp
===H:prp(<sam->) de
===P<:np
====>N:art(<-sam> M S) o
====H:prop(M S) Público
=,
=ADVL:pp
==H:prp desde
==P<:np
===>N:art(M S) o
===>N:adj(<NUM-ord> M S) primeiro
===H:n(M S) número
,
CO:conj-c(<co-vfin>) e
CJT:fcl
=ADVL:np
==>N:adv não
==>N:pron-det(<quant> F P) poucas
==H:n(F P) vezes
=P:vp
==AUX:v-fin(PR 1S IND) tenho
==MV:V PCP manifestado
===ACC:n(F P) opiniões
===ADVL:pp
===H:prp(<sam->) em
===P<:np
====>N:art(<-sam> F P) as
====>N:pron-det(<poss 3S/P> F P) suas
====H:n(F P) páginas
,
S<:fcl
=SUBJ:pron-indp(<rel> M/F S/P) o_que
=ACC:pron-pers(M/F 1S ACC) me
=P:vp-
==AUX:v-fin(PR 3S IND) leva
=ADVL:adv(<kc>) agora
=PRT-AUX<:prp a a
=-P:vp
==MV:v-inf emitir
=ACC:np
==>N:art(M S) o
==>N:pron-det(<poss 1S> M S) meu
==H:n(M S) pensamento
==ADVL:cu
==CJT:pp
===H:prp sobre
===P<:np
====>N:art(M S) o
====H:n(M S) jornal
==,
==CJT:pp
===H:prp sobre
===P<:fcl
====SUBJ:pron-indp(<rel> M/F S/P) quem
====ACC:pron-pers(M 3S ACC) o
====P:v-fin(PR 3S IND) faz
==CO:conj-c(<co-prparg>) e
==CJT:pp
===H:prp sobre
===P<:np
====>N:art(M P) o
====>N:adj(M P) ditos
====H:n(M P) comunicados
.

C 1-3:   É uma das mais antigas discotecas do Algarve, situada em Albufeira, que continua a manter os traços decorativos e as clientelas de sempre.

Let's focus only in the relative clause "que continua a manter os traços decorativos e as clientelas de sempre."  The automatic syntactic analysis is presented in (T1):

(T1)

(...)
===,
===N<:fcl
====SUBJ:pron-indp(<rel> M/F S/P) que
====P:vp
=====AUX:v-fin(PR 3S IND) continua
=====PRT-AUX<:prp a
=====MV:v-inf manter
====ACC:cu
=====CJT:np
======>N:art(M P) os
======H:n(M P) traços
====OC:adj(M P) decorativos
=====CO:conj-c(<co-acc>) e
=====CJT:np
======>N:art(F P) as
======H:n(F P) clientelas
======N<:pp
=======H:prp de
=======P<:adv sempre
.

 The finite main verb manter is a transitive verb, which means that it requires a complement- the direct object. As it can be seen from the above tree, the direct object is a compound unit (coordinated direct objects). There is no question about "as clientelas de sempre" being direct object (S1):

(S1)    ..., que continua a manter as clientelas de sempre.

which successfully responds to the pronoun replacement test (S2):

(S2)    ..., que continua a mantê-las.                  (las, standing for as clientelas de sempre)
 

The question that is to be highlighted here is what really constitutes the first coordinated direct object, that is, having the adjective decorativos as belonging to the noun phrase, or just the noun phrase os traços.
Taking the automatic syntactic analysis of the sentence as as a starting point, the noun phrase os traços is the direct object while the adjective decorativos is considered to be the Object Complement.  Applying the pronoun replacement test, we would then have (S3):

(S3)    ..., que continua a mantê-los decorativos.        (los, standing for the noun phrase os traços)

The pronoun replacement of the direct object, in this case, seems grammatically awckward, which may lead to the conclusion that the adjective in question should not have a clause level function but a group level function instead. In other words, the adjective should be a post modifier of the previous noun, traços (@N<), as it just adds the information of decorative to the notion of traços.

(S4)    *...,que continua a manter decorativos os traços....

The alternative tree should, then, be (T2):

(T2)

(...)
===,
===N<:fcl
====SUBJ:pron-indp(<rel> M/F S/P) que
====P:vp
=====AUX:v-fin(PR 3S IND) continua
=====PRT-AUX<:prp a
=====MV:v-inf manter
====ACC:cu
=====CJT:np
======>N:art(M P) os
======H:n(M P) traços
======N<:adj(M P) decorativos
=====CO:conj-c(<co-acc>) e
=====CJT:np
======>N:art(F P) as
======H:n(F P) clientelas
======N<:pp
=======H:prp de
=======P<:adv sempre
.

Now, the pronoun replacement test is successful when applied to the new noun phrase "os traços decorativos" (S4):

(S4)    ..., que continua a mantê-los.                             (los, standing for os traços decorativos)
 

It is possible to distinguish  nominal complements from adnominal adjuncts by coordinating the suspected object complement with a word or group that features the number and gender agreement in relation to the direct object. (S5) is an example that reinforces the idea that the adjective decorativos is not to be regarded as the object complement:

(S5)   * ..., que continua a manter os traços decorativos e vivos.

(S5')  *..., que continua a mantê-los decorativos e vivos.

On the other hand, if only vivos possesses the clause level function (OC) - and the adjective vivos the group level function, and no coordination between them is, then, possible-  the sentence would be:

(S6)    ..., que continua a manter os traços decorativos vivos.

or in the pronominalised version:

(S6')    ..., que continua a mantê-los vivos.

In this case, if you invert the direct object with the object complement,  it is quite clear that vivos is not as decorativos an adnominal adjunct, having a clause level function

(S6'')   ..., que continua a manter vivos os traços decorativos.
 

More generally, there are some transitive verbs in Portuguese that require object complement, so that the information conveyed is complete. Such verbs are:

a)    nomear / eleger         O povo nomeou / elegeu Jorge Sampaio Presidente da República.

b)    ter     Tenho os pés frios.
                               Tinha a mãe doente.

c)    encontrar (=considerar/ julgar) Encontrei-a triste.

         but  in  Encontrei uma pedra brilhante      , the adjective is just a adnominal adjunct, when encontrar means finding something unknown or  that had been lost)
         (*Encontrei-a brilhante)

d)    considerar    Considero o André inteligente.

e)    reconhecer                Reconheço-o bom profissional.
 

Let's go back to the examples in a) and e). The same sentences could be expressed like:

a')    O povo nomeou/ elegeu Jorge Sampaio como Presidente da República.
e')    Reconheço-o como bom profissional. (C19-1)

And replacing the verb reconhecer in the previous sentence by the verb ter:

e'')    Tenho-o por bom profissional.

Although the object complement is preceded the preposition por  and the adverb como, the function remains the same.
Therefore, as to form, the object complement may be:

            - a noun;
            - an adjective (this is the case where the distinction between adjuncts and object complements is harder, because, in Portuguese, adjectives are mainly postposed)

which in some cases can be preceded by the adverb como or the preposition por, as referred above.

Taking the example e''), one can see that the object complement takes the form of a prepositional phrase,  being the preposition por the head. If one thinks of other prepositional phrases that can replace adjectives, for instance, would the pp still have the same fuction of object complement?
Take the following examples:

(P1)    Encontrei-o bem-humorado.

There's no question whether the adjective bem-humorado is the object complement. The pronoun replacement wipes away any doubts.
However, the adjective can be replaced by a prepositional phrase:

(P1')    Encontrei-o de bom humor.

The head of the prepositional phrase is the preposition de (not por, as the case above considered). Which should then be the syntactic function of the pp at the sentence level?
It can be argued that because it is a prepositional phrase, the syntactic analysis would point at an adverbial. However, semantically, despite the fact that it is a prepositional phrase, it still describes the way he was, with reference to time (Encontrei de bom humor o João). From that point of view, it should be considered as an object complement.
(It is a similar case to the case of prepositional phrases being subject complements:  C1-4 ).
 

Another example where it is clear the difference between adjunct and object complements can be found in C17-3 . Compare both sentences (C1-3 and C17-3) in terms of object complement analysis and adnominal adjunct analysis.
 

C1-4:    É um pouco a versão de uma espécie de outro lado da noite, a meio caminho entre os devaneios de uma fauna
periférica,  seja de Lisboa, Londres, Dublin ou Faro e Portimão, e a postura circunspecta dos fiéis da casa, que de ela esperam a música geracionista dos 60 ou dos 70.

This is not actually a case of a object complement being a pp. It is a case of a subject complement but the principles are basically the same.
In the finite clause "seja de Lisboa, Dublin. Faro e Portimão" (predicative of the previous noun phrase: fauna periférica), the copulative verb requires a complement, even though the subject is not expressed. The only possible complement is actually a coordination of prepositional phrases. They would be considered adverbials in its pure syntactic form, but despite the fact that they represent a place,  they predicate the subject, providing valency mediated information [ser], and therefore, must be, in fact, considered the subject complement (in a compound unit).

C17-3:    Este golpe de Estado deixa céptica a maior parte grandes mestres de xadrez (cerca de 300 em todo o mundo) , que esperam ver a situação clarificada, independentemente da parte que acabe vencedora.
 

Applying all the described tests leading to an object complement analysis, both the adjective céptica and the past participle clarificada are object complements.
The adjective in the subordinate clause, céptica, is immediately identified, syntactically, as an object complement, first by its position  (leading to a clause level fuction), second by pronominal replacement (Este golpe de Estado deixa-a céptica), third by coordination (Este golpe de Estado deixa céptica e pessimista a maior parte grandes mestres de xadrez).

As to clarificada,  the past participle of the verb clarificar, it wouldn't make much sense, if it was considered an adnominal np, to put it in a pre-positional place:
*..., que esperam ver a clarificada situação.

On the other hand if we :
1)    make it directly referred to the verb: ..., que esperam ver clarificada a situação;
2)    pronominalise it: ..., que esperam vê-la clarificada.
3)    coordination:..., que esperam vê-la clarificada e resolvida.

it seems obvious that 'object complement' is the appropriate reading.

Apparently, the use of past participles, with adjectival value and in postposed position, seems to indicate that its function is object complement. One important condition for that is the definiteness of the @<ACC before: 'uma situação clarificada' vs. 'a situação clarificada'. The past participle adjectival value seems to play a weaker role in qualifying the noun directly than the adjectives themselves. For instance, the situation would be a whole lot different if instead of clarificada  we had clarificadora (S1):

(S1)    ..., que esperam ver a situação clarificadora.

In this case, we would have not a object complement as the syntactic function of clarificadora, but a adnominal adjunct.
The problem, however, remains when there is not two forms for the same adjective, but a single one. See example (S2):

(S2)    ..., que esperam ver a  situação contestada.

What is the syntactic function of the adjective / past participle (verb contestar) contestada in this case?

Let's go through the tests:

1)    Position:

(T1)    ..., que esperam ver contestada a situação

2)    Pronoun replacement:

(T2)    ..., que esperam vê-la contestada.

3)    Coordination:

(T3)    ..., que esperam ver a situação contestada e revogada.

So,  apparently we are facing an object complement. However it doesn't seem ackward at all to have:

(T1')    ...., que esperam ver a contestada situação.
(T2')    ..., que esperam vê-la.   (la, standing for situação contestada)
(T3')    ..., que esperam ver a situação contestada revogada.                              or              ..., que esperam vê-la revogada (la, standing for situação contestada)

And from this point of view, contestada could actually be an adnominal adjunct too, a fact that constitutes a syntactic function ambiguity.

There are other cases of syntactic ambiguity between object complements and adnominal adjuncts, derived from the semantic meaning of the verb. For instance, the verb achar and manter are good examples:

(A1)    Achei o livro interessante.

where the verb achar means to consider  then interessante would have the syntactic function of object complement.
However if the verb achar means to find (something that was lost or something that is unknown):

(A1')    Achei um / o livro interessante.

interessante is the adnominal adjunct of the noun livro.

(you can look both tree analyses in the Portuguese closed corpus)

There is a similar case with the verb manter. The syntactic reading will be different according to the semantics of the verb, that is, whether it has the meaning of to keep or to conserve:

(A2)    João mantém o seu cristal brilhante.

This sentence could either mean that a shiny crystal is still kept by João, being brilhante, in this case, the adnominal adjunct, or that João conserves his crystal shiny, by doing something to it (wash, polish, etc). In the latter case, brilhante is the object complement.  Applying the described tests, this ambiguity is clear.
 
 

C3-4:     Estamos a dotar os computadores de um novo sentido disse Steve d' Averio, director de marketing para a Europa da Logitech.
 

This sentence presents a subject-verb inversion in "disse Steve d'Averio". It is quite obvious that the proper noun here is the subject of the clause, given the agreement between subject and the verb disse (3rd person singular). If the verb referred, as it could, to the 1st person singualr, then Steve d'Averio would have to fill in the places either of direct object or, in this case, more probable, indirect object. If that was the case, then Steve d'Averio would have to be the complement of a preposition (disse a Steve d'Averio).
As this is not the case, Steve d'Averio is the subject, which is inverted with the verb. This is one of the contexts where inversion is allowed, that is in constructions with verbs like dizer, sugerir, perguntar, protestar and other similar verbs,  where they end statements in direct speech or are inserted in direct speech statements:

Estamos a dotar os computadores de um novo sentido,- direct speech       ended by   disse Steve d'Averio,.....

Please cf. sentences C2-2, C5-10, C6-6,  C12-2 for other examples.

There are, though, other contexts, where grammatical inversion is allowed, as in the following sentences:

a)     C6-7 for subject-subject complement inversion;
b)     C6-6  and C19-1 for inversion in case of pronominal passivisation;
c)     C11-2  for inversion in the context of adverbial subordinate clauses;
d)     C13-6  for subject-ADVL inversion.
 
 

C6-7:     E tantos foram os candidatos que o período destinado a testar a aplicação do RMG acabaria por ceder lugar a um processo efectivo de financiamento.

If one is to analyse the sentence,  bearing in mind the direct word order in Portuguese, S V O, and considering the presence of a copula verb (ser), the following analysis would be reached, concerning subject and subject complement distinction:

STA:fcl
=CO:conj-c      E
=SUBJ:pron-det(<quant> M P)       tantos
=MV:v-fin(<ink> PS/MQP 3P IND)  foram
=SC:np
==>N:art(<artd> M P)    os
==H:n(M P)      candidatos
=SUB:conj-s     que

(...)

However, one cannot rely only on the position to analyse a sentence, as the direct order is not fixed, for example, subject-verb inversion C3-4 .
If we invert the order, the sentence is still perfectly alright:

                    E os candidatos foram tantos que o período destinado...

If one is to rely only on the position, now candidatos would then be the subject and tantos the subject complement.
In both cases, there is verb-(pseudo)subject agreement.

Let's look at all the context Tantos foram os candidatos : there's a copula verb and neihgbouring it the quantifier tantos. A determiner  article, os, follows the copula verb ser.
If candidatos is the subject complement, then it is possible to mantain its position near the verb and transfer the assumed subject to a post subject complement position as described in (S1):

(S1)    (?) E foram os candidatos tantos...

The sentence sounds ackward. However, if one considers candidatos as the subject and tantos de subject complement, in inverted order, the above construction is possible (S1'):

(S1')    E foram tantos os candidatos...

Concluding, this is a case where there's a subject-subject complement inversion, corresponding to a focus predicative construction. You can find a similar construction in sentence C7-1.

The correct tree would then be:

STA:fcl
=CO:conj-c      E
=SC:pron-det(<quant> M P)       tantos
=MV:v-fin(<ink> PS/MQP 3P IND)  foram
=SUBJ:np
==>N:art(<artd> M P)    os
==H:n(M P)      candidatos
=SUB:conj-s     que

(...)

The same analysis would apply if in the place of the quantifier tantos there were other quantifiers like poucos, outros or muitos.

C6-6:    Para tal, referiu aquele responsável, foram montadas mais estruturas, mais zonas estão abrangidas e, por isso, mais pessoas se podem candidatar.

If not by position and order, the inversion in the case of pronominal passivisation can be immediately spotted by checking verb-subject agreement.
Foram montadas must refer to a feminine plural subject. The only noun phrase that matches the description is mais estruturas.
Therefore, the subject in the passive construction is mais estruturas.

There is another inversion in the sentence, regarding subject-verb inversion referiu aquele responsável that is treated in C3-4
 

C19-1:    Durante a análise do relatório de actividades passadas, foram identificadas como principais insuficiências a ausência de uma orientação nacional junto dos quadros técnicos, e o fraco recrutamento e pouca contribuição na área da Ciência e Tecnologia.
 

This sentence presents two features that had been dealt with before:

1)    Inversion in case of pronominal passivisation:

The verb ser (foram) followed by a past participle, leading to an auxiliary reading of the verb ser, points to a passive voice sentence. The subject, bearing in mind subject-verb agreement, would then be the coordinated np's: a ausência de uma orientação nacional junto dos quadros técnicos , o fraco recrutamento and  pouca contribuição na área da Ciência e Tecnologia.
It's the same case as in sentence C6-6, referred above.
 

2)    Subject complement started by an adverb:

    como principais insuficiências predicates the subject identified before (1)). Even not being an np, it adds valency mediated information on the subject.Therefore, the averbal clause has a subject complement function (clause level function). Note that the valency of 'identificar' in the active voice would call for an object and an object complement, but passivisation has turned the object into a subject, and - consequently - the object complement into a subject complement.
 

C11-2:    Seguro, eficiente, decidido-- tivesse o Penafiel outro Vasco e provavelmente o resultado teria sido outro.

Considering the analysis in the vertical tree:

A1
STA:cu
CJT:fcl
=PRED:cu
==CJT:adj(M S) Seguro
==,
==CJT:adj(M/F S) eficiente
==,
==CJT:v-pcp(M S) decidido
=--
=P:v-fin(IMPF 3S SUBJ) tivesse
=SUBJ:np
==>N:art(M S) o
==H:prop(M S) Penafiel
=ACC:np
==>N:pron-det(<diff> M S) outro
==H:prop(M S) Vasco
CO:conj-c(<co-vfin> <co-fmc>) e
CJT:fcl
=ADVL:adv provavelmente
=SUBJ:np
==>N:art(M S) o
==H:n(M S) resultado
=P:vp
==AUX:v-fin(COND 3S) teria
==MV:V PCP sido
=SC:pron-det(<diff> M S) outro
.

there are two coordinated clauses in this sentence: tivesse o Penafiel outro Vasco and o resultado teria sido outro, linked by the copulative coordenating conjunction e.
In the first conjoined clause tivesse o Penafiel outro Vasco, there is a grammatical inversion (subject-verb) due to the context in which it appears.
Semantically, the phrase tivesse o Penafiel could be faithfully replaced by a subordinate conditional clause (S1):

(S1)    ....., se o Penafiel tivesse outro Vasco, o resultado seria outro.

There are some differences in the form as it can clearly be seen; in the case of an explicit adverbial subordinate clause, there is no subject-verb inversion and, because  a subordinate clause is now present, the coordination disappears.
The semantic value of hipothetical conditional expressed by the Imperfect tense of the Subjunctive remains in both sentences.
The above considerations account for the value of the coordenating conjunction e. Although syntactically there's no question as to its being a conjuction, the semantic value is not of aditive function, but it has the value of consequence / conclusion, which does, in fact, fulfil the conditional aspect of the whole sentence.
 

Apart from the grammatical inversion, the sentence also presents a stylistic inversion to what the subject-adnominal adjunct is concerned. Its value is to give emphasis to the subject's characteristics which might lead to a different result.
 
 
 

C13-6: Muito bem estavam também os três médios-- Maminov, Drozdov e Kharlachev-- , desempenhando Kosolapov as funções de pivot do ataque.

The two highlighted phrases (green) correspond to two different cases of inversion.

1)    Muito bem estavam também os três médios...

Analysing the internal structure of the clause, a copula verb, estavam, is immediately spotted, which will lead to a search for the subject and the subject complement.
The subject is clearly the np, os três médios. Let's consider muito bem. The generated tree is the following:

A1
STA:fcl
ADV:advp
=>A:adv(<quant>) Muito
=H:adv(<quant>) bem
P:v-fin(IMPF 3P IND) estavam
SUBJ:np
=>N:adv também
=>N:art(M P) os
=>N:num(<card> M P) três
=H:n(M P) médios
--

In orange, we have muito bem not functioning as a subject complement, but as an adverbial phrase. However, it is a subject attached adverbial, that is, it is an argument of the subject.
However, some grammarians would consider muito bem a subject complement, as in the similar case:

(S1)    Maria sente-se bem.
 
 

2)    As to desempenhando Kosolapov, this is a case of a gerund clause, which corresponds to  a simultaneou action in relation to the previous clause Muito bem estavam também os três médios-- Maminov, Drozdov e Kharlachev--. The subject is clearly Kosolopov, being  as funções de pivot do ataque the direct object (pronoun replacement: Kosolopov desempenhando-as).
The gerund form could also be a compound form, with the auxiliary ter, taking the gerund form and desempenhar the past participle, which in this case would indicate not simultaneous actions but anteriority in relation to the previous clause (S2):

(S2)    Muito bem estavam também os três médios-- Maminov, Drozdov e Kharlachev-- , tendo Kosolapov desempenhado as funções de pivot do ataque.                                                                 (GER)          (SUBJ)

In both forms (simple or compound gerund) of the same construction, when the subject is expressed it always comes after the gerund form of the verb (auxiliary or main verb).
 
 
 

C12-2:    A frase que o senhor primeiro-ministro leu, não foi ele-fui eu.
 

The issue that is here handled is the analysis of the personal pronouns "ele" and "eu" On the one hand, one might consider that the possible analysis for the sequence "..., não foi ele- fui eu " is to consider both personal pronouns as subjects for the following reasons:

1. the nominative case might indicate the subject reading;
2. verb-subject agreement;

which would then lead to the conclusion that there is a simple S-V inversion.

However, "ser", being a copula verb, selects obligatory a SC as an argument, which is absent in the above sentence, regarding the subject reading of "ele" and "eu".
Therefore, the @SC is the obligatory argument of "ser", not the @SUBJ.

The agreement feature is a kind of "mental interference" from ordinary S-V-SC- agreement, like for instance, in cases of plural inflection where the @SC has agreement with the verb:

(a1) Isto são flores.

Note the V-SC agreement and the agrammaticality of the S-V agreement:

(a2) * Isto é flores.
 
 

C2-4:    Encomendem-nos a projectistas de fora porque, se as obras vierem a ser financiadas, eles até saem de graça, já que, nesse caso, os fundos comunitários pagam os projectos, o mesmo não acontecendo quando eles são feitos por os GAT, dado serem organismos do Estado.
 

Let's focus on the adverbial clause "..., dado serem organismos do Estado".

The problem is to analyse dado- is it a past participle or a preposition?
Let's look at some clearer examples:

1) If it is to be a past participle, then it should have the same features as any other past participle:

        - number and gender inflection in agreement with the subject.

            (a1) Dado o carácter insólito da situação, rir é a melhor solução!

            (a2) Dadas as circunstâncias insólitas, rir é a melhor solução!
 

Accomplishing the PCP's features, the past participle reading  seems fairly stable.
However, some examples with dado, as an invariable word were encoutered while searching in the corpus:

CORPUS NATURA/PÚBLICO v.2.1
(b1)  par 47159: , o que se compreende dado os grandes problemas de alcoolismo que aí
(b2)  par 69333: portuguesa», o que, dado os «níveis de analfabetismo», não será

CORPUS NATURA/DIÁRIO DO MINHO anotado v.1.0
(b4)  par 55766: a criação de micro-empresas, dado as dificuldades de absorção de mão-de-obra
(b5)  par 67165: ser encarados com optimismo, dado as novas perspectivas que se abrem com a

CORPUS CETEMPúblico:
(b5) Ext 21132 (nd, 98b): A Grécia foi o único país a levantar algumas reservas quanto ao ataque militar aos sérvios, dado as relações próximas dos dois povos, embora tenha acabado por anuir .
(b6) Ext 37565 (soc, 98a): O tratamento dos resíduos será feito, de acordo com o protocolo, «em conformidade com as exigências de segurança, saúde pública e protecção do ambiente» e tendo em conta as «exigências de segurança acrescidas, dado as características e especificidades dos medicamentos» .
(b7) Ext 879 (soc, 96a): Estas foram iniciadas há meses, depois de se considerarem esgotadas as hipóteses de venda por arrematação judicial, dado os valores oferecidos -- num máximo de 186 mil contos -- estarem muito aquém da avaliação .

The invariable use of the word dado in these cases might have to do with two facts:

a) semantical proximity with the preposition "visto":

        Ex1: Visto o carácter insólito da situação, rir é a melhor solução!

         Visto as circunstâncias insólitas, rir é a melhor solução!

 The above cases should be separated from the case where "visto" (seen)  is actually the past participle of the verb "ver"
(to see):

         Vistas bem as coisas, o melhor é rir.

b) semantic association with the preposition: dado que (or visto que, uma vez que)

     Let's go back to the initial sentence: ..., dado serem organismos do Estado.

In this case, the subject, unlike the previous examples, is a clause: ..., dado (os GAT) serem organismos do Estado. Therefore, it is harder to check the morphological form of dado: preposition or past participle of the verb "dar", because the agreement is singular.

The clause could be transformed, substituting dado by the preposition dado que :

    ..., dado que são organismos do Estado

However, the replacement of dado by a preposition (dado que) and still maintaining the meaning is not a sufficient condition to consider dado as belonging to the same morphological category.

As a conclusion, both past participle and preposition readings are acceptable, bearing in mind the agreement condition. In case of underspecified male singular  (M S), the past participle is the preferred reading.
 

C179-7: Claro que preferia correr em condições idênticas às que encontramos em Portugal, mas, não sendo possível, paciência."
 

The aim of this comment is to discuss two issues:

1. how ellipses are generally treated;
2. what is the form / function of paciência  in the sentence.
 

1. The above sentence presents one ellipsis and possibly a second one (cf. 2 a)).

The first ellipsis to take into account is present in:

Claro que preferia correr em condições idênticas....

The analyis given by the parser was the following:

Claro    [claro] ADV@ADVL
que    [que] KS @SUB @#FS-A<
preferia (...)
correr    (...)
 

Another possibility is to reconstruct a possible ellipsis that is present here. And in this case, Claro would be, instead of an @ADVL, the subject complement (@SC) of the elliptic verb ser :

(É) claro que preferia correr.....

And consequently, the clause initiated by the subordinate conjunction would be the subject (@SUBJ):

Claro    [claro] ADJ M S @SC
que    [que] KS @SUB @#FS-<SUBJ
preferia (...)
correr    (...)

The question here is to determine the morphological class of the word. As claro may be an adverb or an adjective, then the syntactic analysis may depend on that.

Briefly, ellipses are treated either:

1.1. keeping them visible by attibuting them the syntactic function they would hold if they were embedded in a clause that is elliptic;

1.2. or, especially in cases involving determiners, the determiners become the HEAD of the phrase, which is a far better solution than to have empty heads.

Sentence C170-12 is a very good example of the choices taken when dealing with ellipses.
 
 

2. Let's concentrate on the period: não sendo possível, paciência.

Three possibilities were taken into account:

a) paciência as an averbal clause

This possibility would mean that there is an ellipsis present in the period:

...não sendo possível, (há que haver) paciência

                  or

...não sendo possível, (temos de ter) paciência

This case would fit perfectly in the whole sentence itself, since we would then have  a compound unit with two clauses (the first conjoint being a finite clause, and the second conjoint an averbal clause). Syntactically, and bearing in mind the reconstruction of the ellipsis, paciência would be the direct object of the elliptic verb: ter/ haver.
However, the ellipsis is not immediately reconstructed by the reader, in part because there is not a unique possibility (as it can be seen above) and also because there are no co-textual elements, previously in the sentence, that may help that reconstruction.
 

Furthermore, in similar cases like azar, the reconstruction of the hypothetic ellipsis would be different:
 

Ex: Não conseguimos financiamento, azar .( ~ foi azar)
                                                                               ( ~ tivemos azar)
                                                                               ( ~ por azar)
 

Let's see some examples taken from the CETEMPúblico corpus:
 

Ext 5806 (opi, 98b): Na Alemanha, onde a OCDE estima para este ano um crescimento de 2,7 por cento, o desemprego praticamente não diminui, nem se prevê que nos tempos mais próximos -- azar de Kohl...  (source: CETEMPúblico)
 

Ext 18293 (soc, 95b): Por isso, é natural, nas letras rap, ouvirem-se histórias como a daquele negro que se «fez à life» e -- azar -- foi «de cana» .  (source: CETEMPúblico)
 

Ext 268141 (soc, 95a): Trata-se de um empresário cinquentão que , azar, teve de andar uns tempos com o carro da filha . (source: CETEMPúblico)

Ext 1076000 (des, 96a): «Se não quiserem ver-me jogar, azar. (source: CETEMPúblico)
 

b) Noun

Because of the lack of regularity seen while trying to reconstruct the ellipsis, and in this case, having already the syntactic function (conjoint), the option taken was to consider paciência a noun.
However, this case might not work when the sentence is not a compound unit. Let's take the last example in a):

Ext 1076000 (des, 96a): «Se não quiserem ver-me jogar, azar. (source: CETEMPúblico)

Here we have an adverbial clause (conditional clause), therefore we would expect to have a main clause. Instead, and not considering azar as part of an elliptic clause, we have a noun and determining its syntactic function is a problem.

c) Interjection

And finally, in terms of form, paciência or similar words (azar, for instance) could be considered as being interjections. However the same problem as to determining the syntactic function would remain the same, if the sentence is not a compound unit.
However, many interjections belong to other morphological classes and are used in an "interjectional way" (Bick, 2000).
This is also the case of paciência, a noun possibly used in this context in an "interjectional way". The same happening to azar. So, the original word class should be mantained.
 
 
 

C170-12: «A Marca de Fogo» foi filmado em 1914 em simultâneo com «The Golden Chance, o primeiro de dia e o segundo de noite.

Clearly, one can spot two ellipses in the sentence coordinated. However this is a distinct case from the case in sentence C179-7 (....mas, não sendo possível, paciência), as the ellipsis can easily be resconstructed by the co-textual reference that occurs previously in the sentence: foi filmado.

Due to this fact, there aren't more than one, a unique possibility of reconstructing the ellipsis that should be:

A Marca de Fogo» foi filmado em 1914 em simultâneo com «The Golden Chance, o primeiro foi filmado de dia e o segundo foi filmado de noite.

Therefore, the ellipsis is kept visible, conferring the constituents the syntactic analysis they would held if there was not an elliptic predicator, as the case of the pp's (de dia / de noite) being @ADVL.

The same could be applied to primeiro and segundo, considering that they stand for o primeiro filme and o segundo filme. But in this case, if the adjectives were given the @>N tag, then there would be an empty head of the np (@SUBJ).  Therefore the adjective assumes that syntactic function.

In the CG format:

A=Marca=de=Fogo  [A=Marca=de=Fogo]  PROP M S @SUBJ>
$"»
foi  [ser] <fmc> V PS 3S IND VFIN @FAUX
filmado  [filmar] V PCP M S @IMV @#ICL-AUX<
em  [em] PRP @<ADVL
1914  [1914] <card> NUM M S @P<
em  [em] PRP @<ADVL
simultâneo  [simultâneo] ADJ M S @P<
com  [com] PRP @N< @<ADVL
$"«
The=Golden=Chance  [The=Golden=Chance]  PROP M S @P<
$,
o  [o] <artd> DET M S @>N
primeiro  [primeiro] <NUM-ord> ADJ M S @SUBJ>
de=dia  [de=dia] PP @<ADVL
e  [e] <co-subj> KC @CO
o  [o] <artd> DET M S @>N
segundo  [segundo] <NUM-ord> ADJ M S @SUBJ>
de=noite  [de=noite] PP @<ADVL
$.

And the graphic tree
 
 
 

C165-6: As duas primeiras jovens fraquejam e Miguel vai logo buscar outra.
 
 

How to determine the morphologic tag of the word jovens?

Apparently, we might be tempted to consider it immediately a noun for the following reasons:

1. it can be modified by adjectives and numerals (duas ; primeiras)
2. it admits determiners (in this case the definite article, "a")
3.it is surely the Head of the noun phrase (np), being Subject (@SUBJ) its syntactic function.

However, the same word can also belong to the morphological word class of the adjectives.
Focusing on the following example:
 

(1) As duas primeiras moças jovens fraquejam.

jovens is undoubtedly an adjective (fulfiling the adjective-noun agreement in gender and number).

The reason why there can be a word belonging to different morphological classes has to do, in this case (adjective and noun) to the fact that some sufixes like [nte] produces adjectives and nouns (ex: doente, resistente).
On the other hand, should we consider this kind of words (doente, velho, vivo, morto, pobre, rico,...), belonging to both morphological classes?
It seems acceptable that when the form adjective exists, so does the form noun:

(2) jovem [adj]     ---------- jovem [n]

 but the reverse is not true:

      *mulher[adj]    ---------- mulher [n]
 

However, there are many cases where this one to one relation does not occur. For instance, take the case of quente:

(3) A adesão da Câmara Municipal à comunidade portuária de Aveiro foi um dos pontos mais quentes da noite (...) -
 

No question as to consider quente an adjective. However, it does not fit into a nominal reading (like, for instance, its antonym frio would):

(4) *O quente  é pouco tolerado por muita gente. (should be o calor)
 

And even frio, with both nominal and adjectival readings, would not completely fulfil the semantics either.

(5) O frio é pouco tolerado por muita gente.

(6) O chá está frio.

Frio in (5) is an abstraction, a concept, which appears to be a different case from, for instance, jovem. Jovem, as a nominal reading, would not mean the concept of juventude, but rather a person. Therefore one is tempted to consider that there is, in this case, an implied noun-head.

So, when the nominalised adjective implies a noun-head, the tag <n> was added. On the other hand,  the adjectives that in the nominal form are abstract concepts (not implying, therefore, a noun-head), are lexicalised as nouns.
 

Syntactically, the introduction of the tag <n> will help to determine the Head of the noun phrase, especially in case that two adjectives occur, as in the following example:

(3) O velho simpático é feliz.

In this case velho would have the secondary tag <n>, which would be sufficient to consider it the HEAD of the np and simpático the post modifier (DN).

Another example can be found in sentence C142-1.
 

C165-5: «Quantas Laura Pausini estão aqui esta noite?", pergunta.
 

In terms of tree representation and CG format, these type of sentence had to be manually revised because if it seems obvious for any human that the clause in quotation marks is the direct object of the main clause: pergunta , it is not that obvious for the parser since the subordinating conjunction que is not present.

In CG  format, the sentence would be:

Quantas  [quanto]  <interr> <quant> DET F P @>N
Laura=Pausini  [Laura=Pausini] PROP F S/P @SUBJ>
estão  [estar] <fmc> V PR 3P IND VFIN @FMV @#FS-ACC>
aqui  [aqui] ADV @<ADV
esta  [este] <dem> DET F S @>N
noite  [noite] N F S @<ADVL
$?
$"
$,
pergunta  [perguntar] <fmc> V PR 3S IND VFIN @FMV
$.

The external function of the clause (@#FS-ACC>)is indicated in the main verb of the accusative clause (estão), which would originate the graphic tree.

See sentence C142-1 for a similar case.
 
 

C142-1: «Eles fizeram da nossa aldeia um cemitério», contou uma velha.

This sentence represented in CG format is the following:

Eles  [ele]  PERS M 3P NOM @SUBJ>
fizeram  [fazer] <fmc> V PS/MQP 3P IND VFIN @FMV @#FS-ACC>
de  [de] <sam-> PRP @<PIV
a  [a] <-sam> <artd> DET F S @>N
nossa  [nosso] <poss 1P> DET F S @>N
aldeia  [aldeia] N F S @P<
um  [um] <arti> DET M S @>N
cemitério  [cemitério] N M S @<ACC

$,
contou  [contar] <fmc> V PS 3S IND VFIN @FMV
uma  [um] <arti> DET F S @>N
velha  [velho] <n>ADJ F S @<SUBJ
 

Two things worth noting:

1) velha maintained the morphological tag ADJ, but the tag <n> was added to indicate that the adjective is used as a noun, as was explained in C165-6.

2) contou is in fact the finite main verb, and being an elocutionary verb, it requires a direct object, in theis case, a clause. Since the subordinating conjunction que is absent, the external function is tagged in the finite main verb of the accusative clause (fizeram). For further details go to C165-5.

C172-2Penso que o fundamental é que «Where In The World é o primeiro álbum mesmo «da banda», é só isso.

This sentence presents some problems regarding the syntactic analysis, especially with the period "..., é só isso."
Three possibilities were considered:

1) Coordination at the clause level

Assuming that the comma would work as a coordinating conjunction, then we would have a compound unit with two conjoints being finite clauses.
However, if this option is acceptable strictly in syntactic terms, it clearly does not satisfy semantics.

2) Coordination at the subclause level

This possibility would coordinate two accusative clauses. Something similar to:

(2.1) Penso que o fundamental é que "Where in the world é o primeiro álbum mesmo "da banda", (penso que) é só isso.

The direct object would be a compound unit, with two conjoints as finite clauses.

3) Finite subclause attached to the whole previous period

The above sentence is similar to:

(3.1) Penso que o fundamental é só isso, é que Where in the world é o primeiro álbum mesmo "da banda"

So, the demonstrative pronoun isso would refer to the whole period previous in the sentence, almost acquiring a relative meaning.
Because there isn't still the option of having a clause predication (hypothetically @#FS-S<PRED), the solution would be the following:
 

Penso  [pensar] <fmc> V PR 1S IND VFIN @FMV
que  [que] KS @SUB @#FS-<ACC
o  [o] <artd> DET M S @>N
fundamental  [fundamental] ADJ M S @SUBJ>
é  [ser] V PR 3S IND VFIN @FMV
que  [que] KS @SUB @#FS-<SC
$"«
Where=In=The=World  [Where=In=The=World]  PROP M S @SUBJ>
é  [ser] V PR 3S IND VFIN @FMV
o  [o] <artd> DET M S @>N
primeiro  [primeiro] <NUM-ord> ADJ M S @>N
álbum  [álbum] N M S @<SC
mesmo  [mesmo] <ident> DET M S @N<
$"«
de  [de]  <sam-> PRP @N<
a  [a] <-sam> <artd> DET F S @>N
banda  [banda]  N F S @P<
$"»
$,
é  [ser] <fmc> V PR 3S IND VFIN @FMV @#FS-S<
só  [só] ADV @>N
isso  [isso] <dem> SPEC M S @<SC
$.

4) Chunks of discourse

According to the criteria for sentence separation that was established, the comma is never a sentence separation marker. Therefore the sentence was not divided although it is clear that there are two different, separate chunks of discourse, despite the anaphoric reference (isso). A sentence separation mark, like the full stop or semi-colon could actually replace the comma in this context.
The analysis found for this sentence was considering the top node as a compund unit, but instead of conjoints (CJT) one would have statements wahich are finite clauses (STA:fcl):

A1
STA:cu
<sic>R. -- </sic>
STA:fcl
=P:v-fin(PR 1S IND) Penso
=ACC:fcl
==SUB:conj-s que
==SUBJ:np
===>N:art(M S) o
===H:adj(M S) fundamental
==P:v-fin(PR 3S IND) é
=SC:fcl
==SUB:conj-s que
=="«
==SUBJ:prop(M S) Where_In_The_World
==P:v-fin(PR 3S IND) é
==SC:np
===>N:art(M S) o
===>N:adj(<NUM-ord> M S) primeiro
===H:n(M S) álbum
===N<:pron-det(<ident> M S) mesmo
==="«
===N<:pp
====H:prp(<sam->) de
====P<:np
=====>N:art(<-sam> F S) a
=====H:n(F S) banda
=="»
,
STA:fcl
=P:v-fin(PR 3S IND) é
=>N:adv só
=SC:pron-indp(<dem> M S) isso
.

The same case applies for sentence C176-2.
 

C176-5: Agora, o financiamento do projecto foi muito complicado, tentei na Suécia e mais tarde consegui na Alemanha--

There are two possibilities to analyse this sentence:

1) Consider it a compound unit, under which we would have three conjoints, all of them finite clauses: o financiamento do projecto foi muito complicado; tentei na Suécia; mais tarde consegui na Alemanha.

However, semantically, this analysis would change the original meaning of the sentence, because in fact, replacing the first comma by the coordinating conjunction e isn't possible.

2) Consider an Adverbial clause as a compound unit:

If the comma was to be replaced by a conjunction, and still the meaning of the sentence mantained, then the subordinating conjunction porque would fit. Therefore, the sentence in CG format is:
 

Agora  [agora] <kc> ADV @ADVL
$,
o  [o] <artd> DET M S @>N
financiamento  [financiamento] N M S @SUBJ>
de  [de] <sam-> PRP @N<
o  [o] <-sam> <artd> DET M S @>N
projecto  [projecto] N M S @P<
foi  [ser] <fmc> V PS 3S IND VFIN @FMV
muito  [muito] <quant> ADV @>A
complicado  [complicar] V PCP M S @<SC
$,
tentei  [tentar] <fmc> V PS 1S IND VFIN @FMV @#FS-<ADVL
em  [em] <sam-> PRP @<ADVL
a  [a] <-sam> <artd> DET F S @>N
Suécia  [Suécia] PROP F S @P<
e  [e] <co-vfin> <co-fmc> KC @CO
mais  [mais] <quant> <KOMP> ADV @ADVL> @>A
tarde  [tarde] ADV @ADVL>
consegui  [conseguir] <fmc> V PS 1S IND VFIN @FMV
em  [em] <sam-> PRP @<ADVL
a  [a] <-sam> <artd> DET F S @>N
Alemanha  [Alemanha] PROP F S @P<
$--

and the graphic tree presents the Adverbial clause as a compound unit.
 
 

C163-2 :Armindo Cordeiro, da Câmara Municipal de Lisboa, afirma ainda não ter sido aprovado nenhum, mas que tal virá a
                 acontecer a breve trecho, cabendo a responsabilidade ao Departamento de Espaços Verves, já que o projecto
                 apresentado por a Junta de Freguesia foi recusado por não reunir os requisitos de qualidade necessários.
 

The original sentence, clearly due to the author's mispelling, was:

(...), já que o projecto apresentado poe a Junta de Freguesia foi recusado por não reunir os requisitos de qualidade necessários.
 

The criterion established in principle was that errors in the original corpus should not be corrected. However, whenever these type of mistakes influence the syntactic analysis, especially when making it non possible, such as in the above case, then the mistake is corrected and the correction signaled with the tag <new>:

já=que  [já=que] KS @SUB @#FS-<ADVL
o  [o] <artd> DET M S @>N
projecto  [projecto] N M S @SUBJ>
apresentado  [apresentar] V PCP M S @IMV @#ICL-N<
por  [por] <sam-> PRP @<PASS
a  [a] <-sam> <artd> DET F S @>N
Junta=de=Freguesia  [Junta=de=Freguesia] PROP F S @P<
foi  [ser] V PS 3S IND VFIN @FAUX
recusado  [recusar] V PCP M S @IMV @#ICL-AUX<
por  [por] <new> PRP @<ADVL
não  [não] ADV @ADVL>
reunir  [reunir] V INF @IMV @#ICL-P<
os  [o] <artd> DET M P @>N
requisitos  [requisito] N M P @<ACC
de  [de] PRP @N<
qualidade  [qualidade] N F S @P<
necessários  [necessário] ADJ M P @N<
$.

In the following sentence (C149-8), however, and because a possible analysis was encountered, the mistake was not corrected but signaled with the tag <nil>, meaning that it is a mistake and it should not have been there in the first place:

Após  [após] PRP @ADVL>
uma  [um] <arti> DET F S @>N
perseguição  [perseguição] N F S @P<
de  [de] PRP @N<
mil  [mil] <card> NUM M P @>N
quilómetros  [quilómetro] N M P @P<
através  [através] ADV @ADVL>
de  [de] <sam-> PRP @A<
o  [o] <-sam> <artd> DET M S @>N
país  [país] N M S @P<
$,
a  [a] <artd> DET F S @>N
polícia  [polícia] N F S @SUBJ>
a  [a] <nil> PERS F 3S ACC @ACC>
distribuíu-  [distribuir] <hyfen> <fmc> V PS 3S IND VFIN @FMV
se  [se] PERS F 3S ACC @<ACC
por  [por] <sam-> PRP @<PIV
a  [a] <-sam> <artd> DET F S @>N
autoestrada  [autoestrada] N F S @P<
entre  [entre] PRP @<ADVL
Hofheim  [Hofheim] PROP M/F S @P<
e  [e] <co-prparg> KC @CO
Wiesbade  [Wiesbade] PROP M/F S @P<
$.
 

Another possibility would be analysing polícia not as @SUBJ> but @TOP and 'a' as @ACC> and consequently 'se' the @<SUBJ:

(...)
a  [a] <artd> DET F S @>N
polícia  [polícia] N F S @TOP
a  [a] <nil> PERS F 3S ACC @ACC>
distribuíu-  [distribuir] <hyfen> <fmc> V PS 3S IND VFIN @FMV
se  [se] PERS F 3S ACC @<SUBJ

(...)

C159-5: Devido a a acção de Ames, explicou o actual director da CIA, foi muito mais difícil para os EUA compreender o que se passava na URSS durante aquele período crítico, porque ele denunciou aos soviéticos muitos agentes que trabalhavam para os
serviços americanos.

This sentence presents the main clause (explicou o actual director da CIA) between commas, embedded in the clause which is the direct object (@ACC) of the verb 'explicar'.
More frequently, as similar cases in the CETEMPúblico corpus, this sentence would occur in the following form:

(1) Devido à acção de Ames, foi muito mais difícil para os EUA compreender o que se passava na URSS durante aquele período crítico, porque ele denunciou aos soviéticos muitos agentes que trabalhavam para os serviços americanos, explicou o actual director da CIA.

with no subordinating conjunction present. These cases were described in C142-1 and C165-5.

The same principle, as to the absent conjunction, applies to this sentence:

Devido=a  [devido=a] <sam-> PRP @ADVL>
a  [a] <-sam> <artd> DET F S @>N
acção  [acção] N F S @P<
de  [de] PRP @N<
Ames  [Ames] PROP M/F S @P<
$,
explicou  [explicar] <fmc> V PS 3S IND VFIN @FMV
o  [o] <artd> DET M S @>N
actual  [actual] ADJ M S @>N
director  [director] N M S @<SUBJ
de  [de] <sam-> PRP @N<
a  [a] <-sam> <artd> DET F S @>N
CIA  [CIA] <prop> N M S @P<
$,
foi  [ser] <fmc> V PS 3S IND VFIN @FMV @#FS-<ACC
muito  [muito] <quant> ADV @>A
mais  [mais] <quant> <KOMP> ADV @>A
difícil  [difícil] ADJ M S @<SC
para  [para] PRP @ADVL
os  [o] <artd> DET M P @>N
EUA  [EUA] <prop> N F P @P<
compreender  [compreender] V INF 3S @IMV @#ICL-<SUBJ
o=que  [o=que] <interr> SPEC M/F S @SUBJ> @#FS-<ACC
se  [se] <refl> PERS M 3S ACC @ACC>
passava  [passar] V IMPF 3S IND VFIN @FMV
em  [em] <sam-> PRP @<ADVL
a  [a] <-sam> <artd> DET F S @>N
URSS  [URSS] PROP F S @P<
durante  [durante] PRP @<ADVL
aquele  [aquele] <dem> DET M S @>N
período  [período] N M S @P<
crítico  [crítico] ADJ M S @N<
$,
(...)
 

However, the question of the main clause being embedded in the adverbial clause was tackled in the tree format by considering a discontinuous finite clause:

A1
STA:fcl
ACC:fcl-
=ADVL:pp
==H:prp(<sam->) Devido_a
==P<:np
===>N:art(<-sam> F S) a
===H:n(F S) acção
===N<:pp
====H:prp de
====P<:prop(M/F S) Ames
,
P:v-fin(PS 3S IND) explicou
SUBJ:np
=>N:art(M S) o
=>N:adj(M S) actual
=H:n(M S) director
=N<:pp
==H:prp(<sam->) de
==P<:np
===>N:art(<-sam> F S) a
===H:n(<prop> M S) CIA
,
-ACC:fcl
=P:v-fin(PS 3S IND) foi
=>A:advp
==>A:adv(<quant>) muito
==H:adv(<quant>) mais
=SC:adj(M S) difícil
=ADVL:pp
==H:prp para
==P<:np
===>N:art(M P) os
===H:n(<prop> F P) EUA
=SUBJ:icl
==P:v-inf(3S) compreender
==ACC:fcl
===SUBJ:pron-indp(<interr> M/F S) o_que
===ACC:pron-pers(<refl> M 3S ACC) se
===P:v-fin(IMPF 3S IND) passava
===ADVL:pp
====H:prp(<sam->) em
====P<:np
=====>N:art(<-sam> F S) a
=====H:prop(F S) URSS
===ADVL:pp
====H:prp durante
====P<:np
=====>N:pron-det(<dem> M S) aquele
=====H:n(M S) período
=====N<:adj(M S) crítico
===,
(...)

The descontinuity is clearly seen in the graphic tree.
 

C178-3: Os intérpretes são João Natividade (movimento), Luís Madureira (voz), Olga Pratts (piano), Pedro Wallenstein
                  (contrabaixo), António de Sousa Dias (percussão) e Clemente Cuba (desenho de luzes)

This sentence presents an interesting  peculiarity: although the structure of the items enumerated is the same:

                                            X (x') , Y (y') , Z (z') , ......n(n')      ,

the nature of x', y' and z' are different, fact that raises the problem of how to tag theses elements:

1) @APP
2) @N<PRED
3) @NPHR
4) different tags according to the nature of each n'
 

Neither of them seem to satisfy completely the cases. On the one hand, @APP, would not certainly be the case as there is no identification between n and n' (except probably in Luís Madureira (voz)). If the n' elements say something about n, it is not really the pure case of a postnominal nexus predicative (for instance, Olga Pratts is not a (kind of)piano) is the voice- so it would work perfectly in this case). On the other hand, the use of the tag @NPHR would not show the relation between n and n'. And finally, there could be different tags for each case / nature, but the evaluation of the nature would very much depend on a personal point of view and use of language:

(1) Os intérpretes são João Natividade (no movimento) @N<, Luís Madureira (voz) @APP, Olga Pratts (ao piano)@N<, Pedro Wallenstein (no contrabaixo) @N<, António de Sousa Dias (na percussão) @N< e Clemente Cuba (responsável pelo desenho de luzes) @N<PRED.

But this analysis would not be, at all, consistent.

Another possibility would be introducing a new tag, but for the moment that would have to be introduced fully manually and again it would still not be consistent or sistematic, as the parser would face, for instance, two nouns (like Olga Pratts (piano)) and it would be impossible for it to determine how the n' would semantically match n. For instance there could be the case of having, instead of the proper noun Olga Pratts (piano), Hammon (piano) which in that case piano would be an @APP.

Therefore, these cases are for the moment being dealt as @N<PRED which was the closest and relatively satisfactory tag to apply to such cases.
 
 
 
 
 
 
 
 
 
 
 

 


In order to continue using the Java applets, see troubleshooting tips and Download Java.
On Windows use Internet Explorer 11. macOS no longer supports Java applets.
The Chrome extension CheerpJ Applet Runner may work for some use-cases.


Copyright 1996-2022 | Report a Problem / Contact Us | Printable Version