In this paper we address the main issues and results of a research thesis (Romani, 2020) dedicated to the annotation of metonymies in T-PAS, a corpus-based digital repository of Italian verbal patterns (Ježek et al., 2014). The annotation was performed on the corpus instances of a selected list of 30 verbs and was aimed at both implementing the resource with metonymic patterns and identifying and creating a map of the metonymic relations that occur in the verbal patterns. The annotated corpus data (consisting of 1218 corpus instances), the patterns, and the relations can be useful for NLP tasks such as metonymy recognition.1
universitadipavia.it
Metonymy is a language phenomenon for which
one referent is used to denote another referent
associated with it
The annotation of metonymies was performed
on the corpus instances of a list of 30 verbs
contained in T-PAS
The research had several aims. First, we were interested in studying qualitatively the phenomenon in and through the corpus instances and in implementing the annotation tool of the resource with a specific feature that allowed us to encode metonymic arguments in the verbal patterns. For the latter purpose, we collaborated with the Faculty of Informatics at Masaryk University of Brno (CZ): they gave us the technical support for the implementation of the annotation tool.
Second, our intention was to conceive a general theoretical framework to represent the metonymies found through the qualitative corpus analysis, by designing a map of metonymies and by drafting a list of the metonymic relations that occur in the verbal patterns (see Section 4).
The paper is organized as follows. In Section 2 we present related studies. In Section 3 we describe the methodology we followed in annotating the corpus instances for metonymies, together with a brief introduction to T-PAS. In Section 4 we present the results of our annotation: a generalization of the metonymic relations found, and a map which visually highlights the semantic and cognitive connections between the semantic types. Further developments of the project are described in Section 5; our intention is to enrich the number of analysed verbs and eventually add new types of metonymic relations. 2
Corpus-based studies on metonymy are often
intended for NLP tasks.
Corpus-based studies on metonymies do not
necessarily address NLP tasks. An attempt to
implement corpus-based resources to display
metonymies is described in
T-PAS is the corpus-based resource used in this
research. It consists of a repository of Typed
Predicate Argument Structures (T-PAS) (
Typed predicate argument structures (or t-pass) are patterns which display the syntactic and semantic properties of verbs: for each meaning of a verb a specific t-pas is provided. Verb sense is determined by the arguments it combines with (subject, object, etc.), which are defined through a specific Semantic Type.3
T-pass are corpus-derived: patterns were
acquired through the manual clustering and
annotation of corpus instances for each verb following the
CPA procedure
In Table 1, we show the number of instances annotated for each of the 30 verbs. Overall, the dataset consists of 1218 annotated instances. The number of instances from the random sample can vary from a verb to another, because verbs have different frequencies in the corpus and metonymic phenomena can be more or less pervasive according to the verb under examination. Some cases (e.g. divorare – ‘to devour’ – in Table 1) did not provide any metonymic instance at all (for an explanation and further discussion on this point, see Romani, 2020).4
The annotation procedure was conducted manually by one single annotator (the first author) and, so far, it was not possible to evaluate our annotation procedure as we focused on the qualitative analysis and the retrieval of the relations: it is our intent for the future, as it is essential for further progresses in the research.
Currently, the adopted annotation scheme did not provide ambiguous cases, as metonymies were usually clear-cut and the shift of referent from the source to the target semantic type easily identifia4 In some cases, additional instances were included, if the number of metonymic instances provided by the sample was not sufficient to exemplify a specific relation. Instances with arguments and semantic types analogous to the ones already tagged were selected. To do so, we exploited other Sketch Engine functions (see Romani, 2020 for further details). n. of annotated instances verbs n. of annotated instances 39 parcheggiare 27 raggiungere 21 recarsi 0 rimbombare 24 sentire 66 sorseggiare 16 udire 39 venire 84 versare 13 visitare ble. This may differ from metaphors, for example, where the shift between literal and non-literal meaning may be perceived as more gradual. However, further investigation needs to be done through the annotation of a higher number of instances (expanding the list of verbs) and the comparison and the evaluation of the annotation results of more than one annotator. 4
The overall aim of the research was to give a theoretical account of the metonymic relations found through the corpus analysis and pattern annotation. Therefore, the main results of the study are a list of metonymic relations between the target and the metonymic (source) semantic type (Table 2, Appendix) and a map where the target semantic types are connected to the metonymic types, and the relation between the two is expressed (Figure 3).
The second column in Table 2 (Appendix) reports the 37 relations we identified and encoded (the relations are grouped according to their target type, following this order: [Human], [Location], [Document], [Beverage], [Vehicle], [Sound]). The relation is a short description that illustrates how the metonymic semantic type is connected to the target semantic type; for example (see the highlighted line of the table), [Container] is the metonymic semantic type (first column) and ‘contains’ is the relation (second column) which links [Container] to the target semantic type [Beverage] (third column).5 An instance for this is: ‘we went out to drink a glass’ (glass ‘contains’ something to drink). The fourth column contains an instance in Italian 5 Highlighted in bold are the metonymic semantic types that are also target types (for example, [Sound] is the metonymic semantic type in “[Sound] is emitted by [Human]”, but also the target semantic type in “[Medium] produces [Sound]”). from ItWaC reduced corpus, for each relation found. For each instance, the metonymic argument (exemplifying the source-metonymic semantic type) is highlighted in bold, and the verb triggering the metonymy is in italics.
As a second step, we attempted to draw a map of the metonymic relations, by connecting the target semantic types to their metonymic arguments. In Figure 3, each target semantic type is at the centre of a circular area (target type area), highlighted in bold; in each area the metonymic types related to the target semantic type are included; for each target semantic type, a different colour is given. In most of the cases, they intersect with each other, showing how semantic types can refer to different areas. For example, [Sound] and [Human] share various semantic types (e.g. [Machine], [Musical Instrument], [Medium], as visible in Figure 3) as they can be used both to refer to [Sound] and to [Human] (for clarifying examples, see Table 2 in the Appendix). As mentioned, we included metonymic semantic types in the areas of the map. In our representation, metonymic and target semantic types are connected to each other through arrows, on which the relation is notated. The direction of the arrow traces the direction of the metonymic shift: from the metonymic semantic type to the target semantic type (e.g., [Container] → [Beverage]).
Our results show that metonymic semantic types are fluid; target types can also be metonymic types, in certain contexts, as previously mentioned. For example, [Human] is the target type for [Document] (as [Document] is written or composed by [Human] as an author; e.g. ‘this book tells about II World War’) but also [Document] is the target type for [Human] (as [Human] writes or composes [Document]; e.g. ‘I am reading Shakespeare’).
The structure of the map we conceived draws attention to two main aspects. First, it depicts the complexity of the metonymic relations between semantic types and highlights how metonymy is not a unidirectional phenomenon but, conversely, it is fluid and changeable. Second, from a cognitive point of view, [Human] is at the centre of most of the relations and each target type area is connected to it by multiple relations. In particular, in our data, [Human] is deeply connected and involved within the [Sound] area (for more details, see Romani, 2020).
Finally, for what concerns the limited sample of verbs under investigation, it is interesting to notice that even if there are various source types, the potential target semantic types are only six. We may argue that there is a limited number of target types that attract different source types, in particular regarding [Human] and [Sound], which have the highest number of relations (see Table 2). Further investigation on this point is necessary, together with the extension of the number of examined verbs and instances.
In this paper, we approached the study of
metonymy from a corpus-based perspective. The
research was conducted on a selected list of
verbs, taken from a background study
We also attempted to make some theoretical generalizations based on the metonymic relations we found through the corpus analysis. We therefore created a list of metonymic relations and we designed a map in which the relations are connected to the semantic types they involve. Both the map and the list depict the complexity and variety of the phenomenon, in terms of number of possible metonymic relations and of the semantic types interested.
In future perspectives, we intend to enrich the
map and the list with new relations by extending
the number of verbs investigated and to evaluate
the annotation procedure. For future annotations,
we will provide the current version of the list and
of the map on the online public version of T-PAS
(upcoming). We are also interested in comparing
our results with those in
In line with previous studies (Section 2), we believe that the annotated corpus data, as well as the relations in Table 2, could be useful for automatic detection of metonymies. To our knowledge, little work has been done on this for Italian language: it would be therefore intriguing to test our data in NLP tasks. Task 7: Argument Selection and Coercion. In Proceedings of the 5th International Workshop on Semantic Evaluation (ACL-2010). Uppsala, Sweden. 27-32.
Romani, E. (2020). Searching for Metonymies in Natural Language Texts. A Corpus-based Study on a Resource for Italian Verbs. BA Thesis, University of Pavia, Pavia, Italy.
Appendix metonymic (source) semantic type [Vehicle]
Ricordo la telefonata che mi raggiunse la mattina presto nella mia abitazione di Milano, la corsa in ufficio, il viaggio dell' indomani nei luoghi della catastrofe […] L' uomo viene raggiunto da cinque proiettili e muore mentre viene trasportato in ospedale.
Una campana annuncia l'inizio della messa.
L'altoparlante annunciava l'arrivo di un treno.
Oltre al Flauto d'oro, lo zufolo pastorale annuncia ed accompagna Papageno.
Alcuni studiosi accusavano la psicologia di naturalismo, altri di non essere una scienza naturale.
L'iniziativa consiste nella possibilità per gli anziani di contattare un numero messo a disposizione gratuitamente dal Comune, […] che attiverà uno degli oltre mille volontari. La frase venne interrotta dal suono di sirene, quelle della Polizia.
Una sera, mentre si sta recando ad una cena dove dovrà tenere un discorso, Henry riceve l'invito a presentarsi al commissariato.
Giovanni Paolo II ha visitato il Parlamento italiano, su invito dei Presidenti della Camera dei Deputati. […] raggiungiamo piazza Pio IX dove sorge la Pinacoteca Ambrosiana, entriamo per visitare le opere di Caravaggio, Leonardo e Botticelli.
La mia peste la sento tre volte al giorno, anche se non vuole venire al telefono a parlarmi […] Ho letto Dante, Moravia, Calvino.
Vi raccomandiamo, prima di procedere nella consultazione, di leggere le avvertenze.
Consiglio di leggere senza paraocchi ideologici questa intervista del prof. Dallapiccola sulla diagnosi preimpianto. Al pub Orange Paolo aveva bevuto un bicchiere di troppo e alcuni clienti […] hanno chiesto l'intervento dei carabinieri [Quantity] [Business Enterprise] [Human] [Fantasy Character] [Event] [Machine] [Weapon] [Sound Maker] [Musical Instrument] [Human] [Weather Event] [Part of Language] [Narrative] [Speech Act] [Event] is a portion of produces drives or travels on drives or travels on happens through activated by [Human], produces activated by [Human], produces activated by [Human], produces played by [Human], produces produces, emits produces pronounced by [Human], produces told by [Human] through [Part of Language], produces told by [Human] through [Part of Language], produces involves [Medium]
produces [TV Program]
emits [Beverage] [Beverage] [Vehicle] [Vehicle] [Vehicle] [Sound] [Sound] [Sound] [Sound] [Sound] [Sound] [Sound] [Sound] [Sound] [Sound] [Sound] [Sound] perché venisse allontanato.
Occorre portarsi le sedie e il fuoco, e mettere ciascuno due soldi, se si vuole bere un goccio.
Anche noi della Nazionale beviamo Uliveto! Il presidente del Consiglio è atterrato a mezzogiorno sul campo sportivo di Sant'Agnello a Sorrento.
Una coppia di alieni atterra sulla Terra, precisamente in una campagna.
Alle 16:50 è atterrato il volo speciale Parigi-Beirut della linea di bandiera libanese.
Ma molti non hanno voluto ascoltare la sirena d' allarme e sono rimasti nelle loro abitazioni […] I fucili echeggiano in lontananza mentre tutto intorno continua a muoversi e girare.
Le campane non risuoneranno i rintocchi della morte, ma echeggeranno a festa per celebrare la Vita.
Le trombe non si udivano più, ma dalla parte della vallata si udivano ad intervalli dei lontani fragori.
Se io fossi una persona che non ha mai ascoltato Patty Smith […] magari mi passerebbe anche la voglia di andarla a scoprire.
Ascolta la pioggia, se hai sonno ti tengo con me.
Avete ascoltato tutti le parole di Romano: sono sicuro che tanti tra noi pensano che le sue idee siano una buona base per governare il Paese.
Per ascoltare un racconto, una storia, occorre restare in silenzio.
Low Key udì a stento la domanda di Eric mentre tornava a concentrarsi sul presente.
In una grotta dedicata alla Madonna di Lourdes è possibile, oltre che ascoltare la Santa Messa la domenica, celebrare matrimoni […] Roberto Landi sta seduto dentro il camper e ascolta la televisione.
L'autista stava ascoltando un notiziario della Bbc su quanto è accaduto qualche giorno fa a Madaen.