Finnish Language

Focus marking is an important function of prosody in many languages. While many phonological accounts concentrate on fundamental frequency (F0), studies have established several additional cues to information structure. However, the relationship between these cues is rarely investigated. We simultaneously analyzed five prosodic cues to focus-F0 range, word duration, intensity, voice quality, the location of the F0 maximum, and the occurrence of pauses-in a set of 947 simple Subject Verb Object (SVO) sentences uttered by 17 native speakers of Finnish. Using random forest and generalized additive mixed modelling, we investigated the systematicity of prosodic focus marking, the importance of each cue as a predictor, and their functional shape. Results indicated a highly consistent differentiation between narrow focus and givenness, marked by at least F0 range, word duration, intensity, and the location of the F0 maximum, with F0 range being the most important predictor. No cue had a linear relationship with focus condition. To account for the simultaneous significance of several predictors, we argue that these findings support treating multiple prosodic cues to focus in Finnish as correlates of prosodic phrasing. Thus, we suggest that prosodic phrasing, having multiple functions, is also marked with multiple cues to enhance communicative efficiency.

Spanish and Finnish have a similar, syllable-timed, rhythm but very different stress and intonation patterns. Based on the "Frequency Code," whereby meanings of confidence and aggressiveness are associated with low pitch, and those of submission and smallness with high pitch, we establish a taxonomy of the functions and meanings of intonation with examples in Spanish and Finnish. These functions can be grammatical, attitudinal, pragmatic, and sociolinguistic. To teach intonation to Finnish L2 Spanish students of an advanced level, we raise their awareness through the comparison of the intonational patterns of the two languages and through inductive reasoning. Our proposal fosters a multimodal approach based on the imitation of native speech, on the visual representation of prosody through software like Praat, and on the use of gestures, all integrated into discourse-oriented activities. Due to its schematic and flexible nature, our proposal can be adapted by teachers to the necessities of their L2 Spanish students. RESUMEN El castellano y el finés poseen un ritmo similar, de tipo silábico, mientras que su acento y entonación son distintos. A partir del "Frequency Code", que establece que los significados de confianza y agresividad se asocian con un tono bajo y los de sumisión e insignificancia con uno alto, postulamos una clasificación de las funciones y significados de la entonación con ejemplos en castellano y finés. Las funciones pueden ser gramaticales, actitudinales, pragmáticas y sociolingüísticas. Para enseñar entonación a estudiantes finlandeses de ELE, se les puede hacer conscientes de lo que aprenden a partir de la comparación de los patrones entonativos de las dos lenguas y a través de un razonamiento inductivo. Nuestra propuesta promueve un acercamiento multimodal basado en la imitación del acento nativo, en la representación visual de la prosodia en software como Praat, y en el uso de gestos, todo ello integrado en actividades dirigidas a emular el discurso real. Por su naturaleza flexible y esquemática, los profesores pueden adaptar nuestra propuesta en función de las necesidades de sus estudiantes de español como L2.

Finnish path adpositions meaning ‘through’, ‘along’, ‘across’, ‘past’, ‘over’ and ‘via’, can be used both prepositionally and postpositionally and are thus representative of the typologically rare class of bipositions. In the literature, the alternation has been taken to reflect differences in style rather than meaning. The current study shows, however, that there are semantic differences between the variants that are related to different kinds of motion along the path: factive vs. fictive motion. The term fictive motion refers to the use of directional and dynamic elements to designate situations that are static in the extralinguistic reality (e.g. The highway goes from Paris to Berlin). I distinguish three types of paths; 1) paths of locomotion (actual motion), 2) paths of location (canonical fictive motion expressed by a motion verb), 3) frame-setting paths, which consist of two subtypes: a) paths of occurrence (presence of a multiplicity of entities along the path: There is snow all across the country) and b) paths of process (a process taking place all along the path: It rains across Europe). I argue that the postpositional use of Finnish bipositions is the unmarked way to indicate paths of  locomotion and location, whereas their prepositional use is the unmarked way of indicating frame-setting paths. This is because bipositions used as postpositions are more closely associated with the force-dynamic meaning of the verb, which they elaborate, while those used prepositionally set up an independent, setting-like path. This semantic independence typical of the bipositions used as prepositions is probably also the main reason for their predominance in the expression of time: temporal adverbials are in general more independent of the verb than spatial ones.

This study examines addressing choices by combining university students’ and teachers’ perspectives, concentrating on the French pronominal address forms (tu, T and vous, V). Our data consist of a triangulation of an oral discourse completion task (ODCT), retrospective interviews, and focus group discussions among informants consisting of L2 speakers of French and L1 speakers of Finnish and French. In our analysis, we examined the horizontal and vertical dimensions of address forms on production data and relational work on metapragmatic data. Student data showed that advanced L2 speakers of French used address forms in a more L1-like manner than beginner and intermediate L2 speakers, although all L2-speaking groups used the T forms less often compared to speakers of L1 French. Teacher data revealed that address choices were subject to a discursive dispute.

This paper looks at negation in Finnish dialects from a typological perspective. The focus is on standard negation, i.e. the negation of declarative verbal main clauses. The dialectal variation Finnish shows in its negative construction is examined in the light of current typological knowledge of the expression of negation. Developmental trends connected to the micro-typological variation are also discussed, Finnish dialects are compared with related and neighbouring languages, and relevant theoretical and methodological issues relating to the meeting point of typology and dialectology are addressed.

This work is a cognitive semantic study on existential constructions in Finnish, the main argument being that existentials code an idiosyncratic, external perspective on the semantic relationships they designate. Unlike
nonexistential (transitive and intransitive) sentences, existentials do not assign the status of the semantic starting point to a single participant (typically the referent of the subject), that then steers our perspective toward the other relations designated in the sentence. In existentials, the
perspective on the situation remains external to activities and events that take place in the location or other dimension within which the existential relationship is established. Existentials also suppress individual activities and foreground a holistic viewpoint over the event. It is theoretically important
that the central differences between existential and nonexistential sentences are not found in the ‘‘objective’’ semantic content of the sentence construction but rather in the conceptualizer’s subjective way of approaching
the situation.

This study belongs to the field of folk linguistics, the study of non-linguist attitudes towards and perceptions of language . Here, non-linguist refers to language users who are not language researchers or otherwise trained in linguistics. The folk linguistic approach targets the perceptions, attitudes, and observations of language users rather than or compared to actual usage. Indeed, although folk descriptions can be highly accurate, sociolinguistic studies have shown that language users can at times be quite unsuccessful describers of their actual language use (see, e.g., . Both types of information are, however, of academic interest, and attitudes towards language can provide insight into social attitudes and human behaviour in general. Swearing is a phenomenon that inspires a wealth of folk linguistic theories, and one in which the discrepancy between actual usage and perceived usage often emerges. An investigation into perceptions of such a social norm-governed activity can yield interesting results. On the other hand, as swearing is still a relatively under-explored phenomenon, a folk linguistic study of swearing can identify aspects worthy of further academic investigation.

This thesis explores the inventory of syntactic features that drive Finnish word order. It focusses on several non-finite constructions that manifest categorially inconsistent morpho-syntactic properties. The central assumption underlying this research is that such incongruities result from alternative combinations of cross-linguistically common syntactic features rather than from the presence of rare and exceptional functional categories. My main proposal is that lexical and functional syntactic features may combine to produce hybrid forms. I examine the nature of such feature complexes in different clause types in Finnish: main, finite and non-finite embedded and relative, and infinitival clauses.

This paper examines the possessive system in Finnish. Finnish has morphologically bound possessive suffixes which are hosted by the possessed nouns. These suffixes interact with syntactically independent genitive pronouns and lexical nouns in a complex way. The analysis presented here is based on previous Lexical-FunctionalGrammar (LFG) analyses of subject and object `pro-drop' phenomena. The bound morphemes have for theory-internal reasons previously been assumed to correspond to dual functions: a pronominal function and an agreement function. The Finnish data provide new, empirical evidence for this dual function.

"Wurruk is wurruk if ye're paid to do it an' it's pleasure if ye pay to be allowed to do it." (Finley Peter Dunn)
This is not the kind of article I used to write. This is not a conventional, scientific piece of writing carried out in cold blood, with no emotions, highly respecting each and every one of those sacred tenets we preach about to our students and our better selves. It does not follow many, if any, of the golden rules of how a “B1” [addition: nowadays A1] article is expected to be written.
Rather, this is a rhizome of some themes that continue to intrigue and puzzle me. It is fragmentary, as life is. It is eclectic, as science is. It is personal, as it should be. It highlights bits and pieces — bits and bytes — about what is going on in my mind, with what ontologies, epistemologies and axiologies my conscious and cognisant mind is preoccupied at any moment. Although a dozen other rhizomatic issues had to be omitted, I hope Annikki that you will enjoy this rhizome, because it is what it should be, a rhizomatic spectre of ideational insights, mental representations of Popper’s Worlds 1, 2 and 3, fuzzy images, soul-forbidding frustrations, empowering nanoseconds of joy, and fluid perceptible foresights of artefacts of the third kind. In a word, this is a kaleidoscope of where we came from, where we are, and, perhaps, in the best of worlds, where we might be heading.

Named Entity Recognition (NER), search, classification and tagging of names and name like frequent informational elements in texts, has become a standard information extraction procedure for textual data. NER has been applied to many types of texts and different types of entities: newspapers, fiction, historical records, persons, locations, chemical compounds, protein families, animals etc. In general a NER system's performance is genre and domain dependent and also used entity categories vary (Nadeau and Sekine, 2007). The most general set of named entities is usually some version of three partite categorization of locations, persons and organizations. In this paper we report first large scale trials and evaluation of NER with data out of a digitized Finnish historical newspaper collection Digi. Experiments, results and discussion of this research serve development of the Web collection of historical Finnish newspapers. Digi collection contains 1,960,921 pages of newspaper material from years 1771–1910 both in Finnish and Swedish. We use only material of Finnish documents in our evaluation. The OCRed newspaper collection has lots of OCR errors ; its estimated word level correctness is about 70–75 % (Kettunen and Pääkkönen, 2016). Our principal NER tagger is a rule-based tagger of Finnish, FiNER, provided by the FIN-CLARIN consortium. We show also results of limited category semantic tagging with tools of the Semantic Computing Research Group (SeCo) of the Aalto University. Three other tools are also evaluated briefly. This research reports first published large scale results of NER in a historical Finnish OCRed newspaper collection. Results of the research supplement NER

We compare different LSTMs and transformer models in terms of their effectiveness in normalizing dialectal Finnish into the normative standard Finnish. As dialect is the common way of communication for people online in Finnish, such a normalization is a necessary step to improve the accuracy of the existing Finnish NLP tools that are tailored for norma-tive Finnish text. We work on a corpus consisting of dialectal data from 23 distinct Finnish dialect varieties. The best functioning BRNN approach lowers the initial word error rate of the corpus from 52.89 to 5.73.

This chapter offers an overview of alliteration in Finnic cultures. The focus is on oral rather than written traditions and particularly on two poetic systems which are generally considered to have roots in the vernacular culture going back more than 1000 years.

Our aim in this paper is to explore ways of modeling the distribution of pause durations in conversation using oscillator models (Wilson, Wilson 2006), and to consider how these models might be integrated into our Coupled Oscillator Model of speech timing (COM (O’Dell, Lennes, Werner, Nieminen 2007; O’Dell, Lennes, Nieminen 2008; O’Dell, Nieminen 2009)).

The present study examined a variety of acoustic correlates to the stop length contrast in Finnish beyond the duration of the consonant itself. Of interest were the durations of surrounding vowels, the duration of voice onset time (VOT), and the amplitude of the release burst and the following vowel. Results indicated that for geminate stops, VOT is shorter and the amplitude of both the following vowel and the release burst are higher than for singleton stops. Further, long vowels preceding geminate stops are shorter than those preceding singleton stops, although no difference was found for short vowels. Post-consonantal vowel duration does not vary as a function of consonant length, but is affected by the length of the first-syllable vowel. These results agree with data from other languages in some respects, but not in others. It is proposed that this discrepancy arises from the fact that Finnish, despite being stress-or syllable-timed, also has moralike length features.

We present an open source Python library to automatically produce syntactically correct Finnish sentences when only lemmas and their relations are provided. The tool resolves automatically morphosyntax in the sentence such as agreement and government rules and uses Omorfi to produce the correct morphological forms. In this paper, we discuss how case government can be learned automatically from a corpus and incorporated as a part of the natural language generation tool. We also present how agreement rules are modelled in the system and discuss the use cases of the tool such as its initial use as part of a computational creativity system, called Poem Machine. Tiivistelmä Tässä artikkelissa esittelemme avoimen lähdekoodin Python-kirjaston kielio-pillisten lauseiden automaattista tuottamista varten suomen kielelle. Kieliopilli-set rakenteet pystytään tuottamaan pelkkien lemmojen ja niiden välisten suh-teiden avulla. Työkalu ratkoo vaadittavan morfosyntaktiset vaatimukset kuten kongruenssin ja rektion automaattisesti ja tuottaa morfologisesti oikean muodon Omorfin avulla. Esittelemme tavan, jolla verbien rektiot voidaan poimia auto-maattisesti korpuksesta ja yhdistää osaksi NLG-järjestelmää. Esittelemme, miten kongruenssi on mallinnettu osana järjestelmää ja kuvaamme työkalun alkuperäi-sen käyttötarkoituksen osana laskennallisesti luovaa Runokone-järjestelmää.

This paper introduces the second version of SemFi, a semantic database for Finnish with syntactic relations. The previous version of SemFi has been used in poem generation, and thus it has application area in NLG applications. In addition to extending SemFi, this paper describes and evaluates its translation into four endangered Uralic languages , Skolt Sami, Erzya, Moksha and Komi-Zyrian, all of which are greatly under-resourced. The translated dataset is known as SemUr.

The study concerns itself with Old Helsinki Slang (OHS), a peculiar speech variety (spoken approximately between 1890 and 1950), mixing Finnish morpho-syntax and Swedish vocabulary. It is claimed that it resembles intertwined mixed languages. The study focuses on the lexical, phonological, morphological and syntactical features which are most likely contact- based. Besides the vocabulary, the impact of Swedish is clearly seen in phonology, while in morpho-syntax differences from Finnish are minor. In the conclusion, OHS is compared with Media Lengua and Ma’á. Although OHS was never an independent or an established system, it has a particular value as an example of language mixing in an exceptionally well-documented urban environment, even though the process was interrupted, and OHS gradually merged with Finnish.

Infinitives do not receive much attention in the linguistic literature, although there are many interesting questions that concern them. Some questions are: What is the exact difference between an infinitive and a participle? What syntactic category do infinitives belong to? They have both nominal and verbal features, but do not behave exactly like either nouns or verbs. If they do not belong to anyone category, how do we label them in a syntactic tree model? Is it at all important to label them?

It is well-known that in the Finnic languages there is group of Uralic word-roots which appear to have undergone an unexplained vowel shift in the first and second syllables: e.g., Finnish sarvi : sarve- ‘antler’ (< Proto-Uralic *śorwa) and talvi : talve- ‘winter’ (< Proto-Uralic *tälwä). These words have been referred to as ‘secondary e-stems’, as the shape of their cognates outside Finnic suggests a proto-form with the stem vowel *a or *ä. This paper proposes a solution which provides a regular phonological account of the development of this class of word-roots. The solution involves the revision of certain aspects in the theory of historical vocalism of Saami and Mordvin.

We present a creative poem generator for the morphologically rich Finnish language. Our method falls into the master-apprentice paradigm, where a computationally creative genetic algorithm teaches a BRNN model to generate poetry. We model several parts of poetic aesthetics in the fitness function of the genetic algorithm, such as sonic features, semantic coherence, imagery and metaphor. Furthermore , we justify the creativity of our method based on the FACE theory on computational creativity and take additional care in evaluating our system by automatic metrics for concepts together with human evaluation for aesthetics, framing and expressions.

This article examines the semantic factors motivating the use of directional (‘to’) local cases in the morphological marking of the locative argument of the verb jäädä ‘remain’ in Finnish. It is argued that this use of the directional cases is motivated by
the dynamic conceptualization of an apparently static situation. In this dynamic conceptualization, the actual course of events where the entity remains in a location is contrasted with a projected course of events whereby the entity leaves the location. The event of ‘remaining’ thus constitutes a change from the projected course of events into the actual one, and it is this dynamic meaning that motivates the directional case marking. The paper also addresses the more theoretical question of whether conceptualization can motivate the use of grammatical markers with other Finnish verbs.

We present a novel approach for generating poetry automatically for the morphologically rich Finnish language by using a genetic algorithm. The approach improves the state of the art of the previous Finnish poem generators by introducing a higher degree of freedom in terms of structural creativity. Our approach is evaluated and described within the paradigm of computational creativity, where the fitness functions of the genetic algorithm are assimilated with the notion of aesthetics. The output is considered to be a poem 81.5% of the time by human evaluators.

Named Entity Recognition (NER), search, classification and tagging of names and name like informational elements in texts, has become a standard information extraction procedure for textual data during the last two decades. NER has been applied to many types of texts and different types of entities:

The basic semantic function of locational elements with the syntactic status of clause modifier is typically assumed to be the establishing of a (spatial, temporal, or other) setting, within which entities interact and processes unfold. Less attention has been paid to structures where several settings are connected to the same predicate, and to the mutual semantic relation
of the settings in these structures. In this article evidence is produced for the argument that these constructions (called multiple locational specifications) reveal a scope relation between the settings such that a semantically higher domain dominates a lower one. For instance, in the sentence On
Monday nights, Ollie played chess with Arthur in London, time dominates space by confining the spatial relation between Ollie and London to Monday nights only. The opposite scope relation can be achieved by creating a special interpretation of the lower domain. For instance, space can dominate
time only if it acquires a temporal sense itself. Such a reading foregrounds the duration of the presence of an entity in the space. Syntactically, this effect is created by word order; e.g. In London, Ollie played chess with Arthur on Monday nights evokes a temporal reading for in London (‘when in London’), enabling it to include on Monday nights within its scope. It
is argued in this paper that temporal and other special interpretations are quite productive and possible for many different kinds of domain.

We present Poem Machine, an interactive online tool for co-authoring Finnish poetry with a computationally creative agent. Poem Machine can produce poetry of its own and assist the user in authoring poems. The main target group for the system is primary school children, and its use as a part of teaching is currently under study.

This article illustrates the grammatical and error annotations of a morphologically rich learner language with the help of the International Corpus of Learner Finnish (ICLFI). It especially focuses on problems and solutions in morphological and error annotation, both of which are challenging due to the rich morphological structure of the target language. The article also introduces existing Finno-Ugric learner data and their annotation schemes, and compares those with the ones used in ICLFI annotations. Learner data variables, taxonomy, and principles in grammatical and error annotation are also discussed with the help of the ICLFI in the present article.

Subject constituent of the existential sentence in the writing of advanced learners of Finnish (article is in Finnish)

The subject constituent of the existential sentence in Finnish differs from the prototypical subject: there is no congruence between this constituent and the predicate verb, and it can take either the nominative or partitive case.
In the article I discuss the use and variation of the subject constituents of existential sentences. The article is based on the analysis of 2575 sentences from exam essays by advanced learners of Finnish. The data is part of the Advanced Finnish Learners’ Corpus (LAS2) at the University of Turku. The reference material is 1310 sentences of exam essays by native speakers of Finnish.
In the material the nominative variant of the subject constituent of existential sentences occurs more frequently in the learner language than in the reference material. This overuse is partly due to the unidiomatic casemarking of the subject constituents; it may also be due to an unidiomatic analogy between the subject constituents of existential sentences and other subject constituents. The appearance of two possible forms, the nominative and the partitive, may create a new situation of free variation in the casemarking of the subject constituent of existential sentences. Longitudinal observation indicates three consecutive phases: 1) idiomatic casemarking with an overuse of nominative subjects, 2) free variation between nominative and partitive subject constituents, and 3) casemarking similar to the target language. Some potential co-occurrences were also found, such as unidiomatic casemarking together with an unidiomatically congruent verb or subject qualifier.
Based on these findings, analogy may also cause unidiomatic variation in the interlanguage. Thus more thorough corpus research of the tendencies observed in this article may reveal aspects that are out of reach without extensive corpus analysis.

Using etymological methods, the present study has identified five Sinitic and Uralic shared etymologies. These five etymologies form a rhyme correspondence. This regular sound change validates the etymological connection between Sinitic and Uralic. The Finnic term for 'sky' is among these five etymologies. It is demonstrated that this word root should be aboriginal in Sino-Uralic languages.

This essay sets out an approach to parallelism in verbal art as a semiotic phenomenon that can operate at multiple orders (or levels) of signification. It examines parallelism in the sounds through which words are communicated, in language communicated by those sounds, in symbols or minimal units of narration communicated through language, and then in more complex units of narration communicated through those symbols or units. Attention is given to how these different levels of parallelism interrelate and may diverge, while revealing that parallelism at all of these levels reflects a coherent semiotic phenomenon.

"Parallelism in Verbal Art and Performance: An Introduction" (in collaboration with Lotte Tarkka), "Parallelism and Orders of Signification (Parallelism Dynamics I)" and "Multimedial Parallelism in Ritual Performance (Parallelism Dynamics II)" can be read together as a monograph-length study.

The article discusses various syntactic and semantic properties of the so-called 5th infinitive in contemporary written Finnish. The non-finite verb form in question consists of the suffix -maisilla-/-mäisillä- followed by a possessive suffix indicating the subject of the verb. The semantic content of the 5th infinitive is that of the proximative aspect; in other words, it is a grammatical means of referring to ”a temporal phase located just before the initial boundary of the situation described by the main verb” (= ’be about to V, be on the verge of V-ing’; see e.g. Heine 1994, 36; Kuteva 2001, 92).

In previous descriptions of the 5th infinitive, it has been almost unanimously described as occurring only in connection with the verb olla ’be’, and the resulting proximative predicates have been vaguely characterised, for instance, as ”periphrastic types of constructions” or ”complex verbal constructions”. Drawing on data mainly provided by the corpus of 180 million word tokens (Language Bank of Finland, accessible at http://www.csc.fi/kielipankki/), I present a number of examples showing that the 5th infinitive is used not only to form a periphrastic proximative, but in about six percent of occurrences the 5th infinitive forms are used as more or less optional adverbial modifiers (converbs) of other verbs.

When scrutinising various functions of the 5th infinitive, it can be seen to be related to various adverbial verb forms in the language. Special attention in the article is given to the close relationship between the 5th infinitive and the progressive verb form in -massa/-mässä (”the 3rd infinitive inessive” in traditional Finnish grammar). Both of them are typically used to form periphrastic aspectual constructions with olla ’be’ as the main verb. In addition, they can also be used to modify many other verbs, they can be temporalised, have subjects of their own and may co-occur with various other adverbial modifiers and other aspectual verb forms, but they still retain their inherent aspectual meanings.

At the end of the article, the so-called 5th infinitive is defined as an adverbial, proximative verb form; in other words, in spite of occasional claims to the contrary, there is no plausible reason to regard proximative forms as adverbs. Even though characterised as an adverbial verb form, the 5th infinitive differs from many other adverbial verb forms (such as the so-called first infinitive translative -takse-/-täkse- [+ possessive suffix] ’in order to V’ or the second infinitive inessive -tessa/-tessä ’when/while V-ing’) in that it does not address any specific interpropositional relationship in itself and its use is not restricted to optional adverbial modification. Instead, the 5th infinitive, along with certain other non-finites, has much in common with adverbial local case forms whose optionality/obligatoriness is largely determined by the overall context.

At first sight, the intonation systems of French and Finnish – which are typologically distant languages – seem completely different. This paper aims, however, at showing that the discourse-structuring role of the utterance-final pitch rises in spontaneous spoken Finnish is in many respects reminiscent of the fi nal pitch rises that constitute an essential component of the French intonation system (Morel & Danon-Boileau, 1998). Indeed, in both cases, utterance-final pitch rises occurring inside a multi-unit turn seem to “recategorise the rheme as a preamble” for what will follow (Morel & Danon-Boileau, 1998). ‘Preamble’ and ‘rheme’ are the two main constituents of the so-called ‘oral paragraph’, which is the basic structural unit of French discourse according to their model. In addition to comparing the realization of the so-called ‘recategorization phenomenon’ in French and in Finnish, this paper aims at describing the structure of the ‘oral paragraph’ of Finnish within Morel and Danon-Boileau’s framework.

Keywords: oral paragraph, recategorization, rheme, intonation, pitch rise, F0, spoken Finnish, Finnish intonation, prosody, structuration of discourse

The aim of this article is to examine communicative actions of nine Swedish-Finnish bilingual children enrolled in Swedish-medium early childhood education and care in Finland. We discuss the interplay between declared monolingual language policy, and the ideology of supporting the language background of the bilingual child. As a result we found a dual practiced language policy. With the agency and actions of bilingual children we could identify both monolingual policy being practiced, but also a bilingual policy being constructed at the same time. Neither of the practiced dual policies seemed to be explicitly resisted either by the children or the pedagogical practitioners. The communicative actions of the nine children showed that they are individuals and they all constructed their bilingual policies in varied ways, some of the children even, quite clearly, preferring the monolingual policy in their communicative actions. 1. Inledning Syftet med denna artikel är att undersöka tvåspråkiga barns språkliga agens så som den visar sig via barnens kommunikativa handlingar i svenskspråkig småbarnspedagogisk verksamhet i tre olika kommuner i Finland. Vi under-söker hur tvåspråkiga barns handlingar och agens gestaltar sig i utvalda verksamhetssituationer (fri lek ensam, fri lek med kompisar samt lärarledd verksamhet) samt vad dessa handlingar i sin tur berättar om den utövade språkpolicyn på daghemmet. Vi tittar således på den policy och praxis som formar det individuella barnets möjligheter att agera på sina respektive språk – svenska och finska – och ser barnens kommunikativa handlingar som ett slags fönster in i det mera abstrakta begreppet agens (se nedan för vidare diskussion och definitioner). Barnen, de individuella agenterna, deltar alla i svenskspråkig småbarns-pedagogisk verksamhet på daghem. Trots denna gemensamma nämnare är daghemmen olika och har varierande praxisstrukturer bl.a. som ett resultat av balansen mellan svenska och finska i det omgivande samhället. Detta ger i sin tur barnen olika ramar för att utöva sin agens. Genom att studera varia-tionsvidden i barnens handlingar och agens kan vi lära oss något om för-handlingsutrymmet i den uttalade språkpolicyn, dvs. vilka typer av språk

The present production study shows that information structure is prosodically expressed in several parallel ways in Finnish. In addition to producing words in narrow focus with (1) a larger duration, (2) higher f0 peaks and greater falls, and (3) an earlier and more consistent peak alignment, participants frequently marked these words using also (4) pauses and (5) non-modal voice quality. Furthermore, the means of prosodic marking of information structure interact with sentence position in shaping the prosodic form of the word.

Please cite this paper as: Kittilä, Seppo. 2014. The (non-)prototypicality of Direction: the (allative and illative) case(s) of Finnish. In Silvia Luraghi & Heiko Narrog (eds), Perspectives on semantic roles, 151-179. Amsterdam: Benjamins. If you want to have an offprint, send me an e-mail at seppo.kittila@helsinki.fi.

The present paper proposes an alternative approach to the aspects of word formation and lexical processes by the example of the Finnish causative verbs derived with the suffix (U)ttA. The causative verb behaviour is argued to be affected by a relational network of both prototypical and construction-specific structures. The prototype structures defined by the derivatives‟ salient features in form of the conceptual structure and linking regularities are used to describe the general behaviour of the causative derivatives and to contrast to the idiosyncratic patterns occurring in connection with these verbs. The constructions discussed in this paper are the emotive causative construction and the middle construction.