36.2278, Disc: Could I have your opinion and advice on my hypothesis for “Preposition stranding” in Brazilian Portuguese?

[The LINGUIST List]

LINGUIST List: Vol-36-2278. Tue Jul 29 2025. ISSN: 1069 - 4875.

Subject: 36.2278, Disc: Could I have your opinion and advice on my hypothesis for “Preposition stranding” in Brazilian Portuguese?

Moderator: Steven Moran (linguist at linguistlist.org)
Managing Editor: Valeriia Vyshnevetska
Team: Helen Aristar-Dry, Mara Baccaro, Daniel Swanson
Jobs: jobs at linguistlist.org | Conferences: callconf at linguistlist.org | Pubs: pubs at linguistlist.org

Homepage: http://linguistlist.org

Editor for this issue: Daniel Swanson <daniel at linguistlist.org>

================================================================


Date: 29-Jul-2025
From: Gabriel Malheiros [gabrielteodoro_94 at hotmail.com]
Subject: Could I have your opinion and advice on my hypothesis for “Preposition stranding” in Brazilian Portuguese?


Does Brazilian Portuguese allow P-stranding?A syntactic and lexical
argument for P-orphaning
Hi everyone,
I’m preparing to present a version of my MA thesis this October at the
Konstanz Linguistics Conference and would really appreciate your
thoughts and feedback before then. The topic concerns the syntax of
Brazilian Portuguese (BP) prepositions that appear without adjacent
complements — often analyzed as P-stranding, but I argue these are
better analyzed as P-orphaning, involving a lexical preposition
selecting a null resumptive pronoun (pro), rather than movement plus
trace. I don’t feel prepared to present yet, so I would like to know
what you think and if my hypothesis has any value.
Core Hypothesis
Brazilian Portuguese does not have productive P-stranding as found in
English. Instead, apparent “stranding” cases involve lexical
prepositions like sem (‘without’), contra (‘against’), and sobre
(‘about’) that select a null resumptive pronoun (pro) inside the PP.
Functional prepositions like de, em, com, para never appear stranded.
Main Question
Why can prepositions like sem, contra, and sobre appear
stranded/orphaned, while others like de and em cannot?
Summary of the Hypothesis
* Brazilian Portuguese does not allow true P-stranding.
* What looks like P-stranding is actually P-orphaning: a lexical
preposition selects a silent resumptive pronoun (pro) instead of a
moved DP.
* This only happens with lexical prepositions (like sem, contra,
sobre) and not with functional prepositions (like de, em, com, para).
Lexical vs. Functional Prepositions
Property        Lexical Prepositions (sem, contra, sobre)
Functional Prepositions (de, em, com, para)
Appear stranded/orphaned?       Yes     No
Contract with determiners?      No      Yes (e.g., dele = de + ele)
Semantic weight Full    Bleached/functional
Survive in islands?     Yes     No (movement blocked)
Can appear orphaned in discourse        Yes     No
Cannot be deleted with null pronouns    Yes     No
Examples:
* Lexical Ps stranded/orphaned:
1. A vida me tirou pessoas que eu achava que nunca viveria sem.
“Life
took from me people I thought I’d never live without.”
2. Tem dois homens que eu não aguento que falem sobre.
“There are two
men I can’t stand people talking about.”
3. Quem ele fez campanha contra?
“Who did he campaign against?”
(Possible but marked)
* Functional Ps never stranded:
1. *O lugar que eu moro em 
“The place I live in” (Ungrammatical)
2. *A pessoa que eu falei com 
“The person I spoke with”
(Ungrammatical)
Structural Environments: Movement vs. Dislocation
The bulk of real-life BP examples come from topicalized, coordinated,
or pragmatically marked structures, not classic A′-movement
environments like wh-questions or relatives. Examples:
1. Você é igual meu celular: não vivo sem, mas odeio quando está na
mão dos outros.
“You’re like my phone: I can’t live without (it), but
I hate when others are holding it.”
2. Esse julgamento, eu sou contra.
“That verdict — I’m against (it).”
3. Técnicas de comunicação, prefiro nem falar sobre.
“Communication
techniques — I’d rather not talk about (them).”
These resemble left dislocation rather than wh-movement.
Weak Evidence in Wh-Contexts
Although some authors claim BP allows P-stranding in questions, these
are rare and often feel marginal to native speakers:
1. O que você não vive sem?
“What can’t you live without?” ()
2. Quem ele fez campanha contra?
“Who did he campaign against?”
(Possible but marked)
Data from corpora (e.g., Adams 2023) shows that interrogative cases
with stranded sobre are extremely rare — in over 5,000 tokens of
sobre, only 2 were in wh-questions.
Null Resumptives and Island Effects
A crucial piece of evidence is that BP constructions survive in
islands, where movement is blocked — but orphaning with pro is
allowed:
1. Coisas que se eu falar sobre, eu sou cancelado.
“Things that, if I
talk about, I’ll get canceled.”
2. Essa é a única parte da física que eu conheço alguém que teve aula
sobre.
“This is the only part of physics I know someone had a class
about.”
Compare to English, where a pronoun is required:
1. Things that if I talk about __, I’ll be canceled. ✖
→ Must say:
“...talk about them.”
This points to the presence of a base-generated resumptive pro, not a
gap from movement.
Semantic and Morphosyntactic Support
* Lexical prepositions in BP (sem, contra, sobre) have semantic weight
and never contract with determiners.
* Functional prepositions (de, em, com) contract and never appear
without overt arguments.
* Omitting lexical Ps leads to ungrammaticality or semantic reversal:
1. A vida me tirou pessoas que eu não consigo viver sem.
→ Dropping
sem reverses meaning: “I can’t live with [them].”
2. O político que eu fiz campanha contra.
→ Dropping contra suggests
support, not opposition.
This semantic load aligns with orphaning, where the preposition still
selects an (invisible) argument.
Crosslinguistic Parallels
BP mirrors orphaning patterns in French:
* Elle est pas mal, ta vie. Tu devrais écrire un livre dessus.
“Your
life’s not bad. You should write a book about it.” (Poplack et al.,
2011; Authier, 2014)
Like in BP, only lexical prepositions are orphaned in these French
varieties. Functional prepositions don’t appear alone.
Syntactic Proposal (Based on Kato & Nunes 2009)
Relatives in BP can be derived without A′-movement. Instead, a
topic-DP is base-generated, and a resumptive pro appears in the PP.
For example:
1. O livro que eu falei sobre
“The book I talked about”
Syntactic structure:
csharp
Copy code
[DP o livroi [CP quek [LD tk [IP eu falei [PP sobre proi]]]]]
No movement from inside PP → no P-stranding.
Functional Prepositions and the [P+D] Hypothesis
Following Salles (1997, 2001), functional prepositions in BP form a
complex syntactic head [P+D] with the DP they select. This makes them
inseparable at PF — either both are pronounced or both are deleted (as
in resumptive relatives and cut relatives).
For example:
(1) Este é o livro que o aluno precisa Ø / dele.
“This is the book
that the student needs Ø / of-it.”
Here, the functional preposition de forms a head with the pronoun
(dele = de + ele), which must be deleted as a unit if the pronoun is
null. Hence, de cannot appear “stranded.”
By contrast, lexical prepositions like sobre or sem do not form a
[P+D] head. They stand alone and can select a silent pro without
requiring contraction or deletion.
(2) Este é o artigo que a gente debateu sobre Ø / ele.
“This is the
article that we debated about Ø / it.”
(3) É esse tipo de gente que eu me refiro Ø / a.
“It’s the kind of
people (to) whom I’m referring Ø / to.”
Functional Ps like a or de contract with pronouns. Lexical Ps like
sobre don’t — they are not deleted with the pronoun and often must be
pronounced, which would not happen if they were stranded.
Syntactic Derivations (Simplified)
Functional P:
(4) O aluno precisa dele. → [P+D] = dele
“The student needs of-him.”
Lexical P:
(5) A gente debateu sobre Ø.
“We debated about Ø.”
Lexical Ps select a null pro and are not deleted with it. This
difference aligns with the Proper Head Movement Generalization (Baker
2003) and the Head Movement Constraint: functional heads can
incorporate into other functional heads, but not into lexical ones.
Hence, D can’t move into P when P is lexical.
Empirical Data
* Sem, contra, and sobre often occur without an overt complement.
* De, em, com, etc., are never “stranded.”
* Omitting lexical Ps leads to ungrammaticality or a change in
meaning:
(6) A única coisa que eu tô sem é dinheiro.
“The only thing I’m
without is money.”
(Deleting sem would reverse the meaning.)
(7) Esse é o time que ele torce contra.
“This is the team he roots
against.”
(Without contra, it would mean “he supports the team.”)
(8) Odeio quando eu tô falando de um assunto que eu li sobre.
“I hate
when I’m talking about a topic I read about.”
Crosslinguistic Comparison: English vs BP
English:
(9) This is the book that I talked about.
Movement from PP is
possible. No resumptives allowed.
Brazilian Portuguese:
(10) Esse é o livro que eu falei sobre. - No movement. Preposition
selects pro.  - This is the book that I talked about (it)
Final Generalization
A preposition α can surface without an overt complement only if α is
lexical.
This explains why sem, sobre, contra are found “orphaned,” while de,
em, com are not — despite all being prepositions. This has
implications for the typology of P-stranding and resumptivity in
Romance languages.
Questions for Feedback
1. Would you consider this a Romance-style case of P-orphaning rather
than stranding?
2. Does the [P+D] hypothesis seem sufficient to explain why some Ps
are deleted with their complements and others not?
3. How could I refine the argument or clarify the syntactic
derivations?
Thanks very much for your time and comments!
Best regards,

Gabriel Malheiros

University of Brasília

Linguistic Field(s): General Linguistics
                     Syntax

Subject Language(s): Portuguese (por)

Language Family(ies): Romance



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

********************** LINGUIST List Support ***********************
Please consider donating to the Linguist List, a U.S. 501(c)(3) not for profit organization:

https://www.paypal.com/donate/?hosted_button_id=87C2AXTVC4PP8

LINGUIST List is supported by the following publishers:

Bloomsbury Publishing http://www.bloomsbury.com/uk/

Cascadilla Press http://www.cascadilla.com/

John Benjamins http://www.benjamins.com/

Language Science Press http://langsci-press.org

MIT Press http://mitpress.mit.edu/

Multilingual Matters http://www.multilingual-matters.com/

Narr Francke Attempto Verlag GmbH + Co. KG http://www.narr.de/

Netherlands Graduate School of Linguistics / Landelijke (LOT) http://www.lotpublications.nl/


----------------------------------------------------------
LINGUIST List: Vol-36-2278
----------------------------------------------------------