Ironi som forskningsobjekt. Är det möjligt att forska i det som finns mellan raderna?

Abstract (Irony as object of research: Is it possible to explore what is between the lines?): The main concern of this article is the interpretation of irony: how is it brought about and how can it be investigated? The method applied is based on authentic texts and their elicited interpretations − a method referred to in this article response analysis. Interpretation of irony in the approach taken is seen as being crucially dependent on the notion of coherence. A text is perceived as being coherent if it (a) makes sense and if it(b) hangs together. Incoherent texts can result in an ironic interpretation; however, the incoherence must also be perceived as being intentional, and intentionality in turn is a sign of the edge of the ironist. Ironic interpretation is defined as a combination of five factors: (1) an ironic edge that (2) reflects the intention of the ironist, and (3) has a target and (4) a victim too. Essential to irony is its fifth factor, the fact that one or more of these four factors must be inferred from co(n)text. This definition of irony is crucial in distinguishing irony from non-irony, and it also helps to discern the differences as well as the similarities between irony and related phenomena.

Otoksen edustavuuden mittaus R-indikaattorin avulla Suomen uhritutkimuspilotissa

Tutkielmassa sovelletaan aineiston edustavuutta mittaavaa laatuindikaattoria Suomen uhritutkimuspilottiin tilanteessa, jossa ilmenee vastauskatoa. Vastauskato on kasvava ongelma tilastotutkimuksissa: jos tutkimukseen osallistuneet eivät edusta otosjoukkoa tutkittavan asian suhteen, voi vastauskadosta aiheutuva harha olla estimoiduissa tunnusluvuissa hyvinkin suuri. Tutkimuksissa näkee usein julkaistavan vastausasteen ikään kuin se kertoisi aukottomasti tutkimuksen laadusta. Pelkkä korkea vastausaste ei kuitenkaan välttämättä takaa estimaattien harhattomuutta, sillä se ei kerro mitään vastanneiden ja vastaamattomien eroista tutkittavan asian suhteen. Tarvitaan siis muita mittareita, joilla vastanneiden laatua voitaisiin paremmin arvioida, ja R-indikaattori tarjoaa yhden vaihtoehdon. R-indikaattori mittaa otosalkioiden vastausalttiuksien välistä vaihtelua. R-indikaattorin estimoiminen edellyttää siis vastausalttiuksien estimointia, mikä puolestaan edellyttää apumuuttujien olemassaoloa kaikille otosalkioille. Vastausalttiuksien estimoimiseen käytettiin linkkifunktiona sekä logistista mallia että ja Särndalin ja Lundströmin (2008) vastausvaikutusten mallia. Vastauskäyttäytymiseen vaikuttavan apumuuttujajoukon valinta tehtiin alan kirjallisuuteen perustuen (Groves & Couper 1998). Koska R-indikaattorin estimaattori on satunnaismuuttuja, täytyi sille estimoida varianssi ja mahdollinen harha (Shlomo ym. 2009). Estimoinnissa käytettiin Bootstrap-pseudotoistomenetelmää, jossa alkuperäisestä aineistosta poimitaan niin kutsuttuja pseudo-otoksia, joiden avulla R-indikaattorin estimaattorille voidaan laskea keskivirhe. Suomen uhritutkimuspilotti koostui kolmesta eri tiedonkeruumenetelmällä poimitusta otoksesta: CAPI-, CATI- CAVVIotoksesta. Vastausasteet vaihtelivat aineistoissa paljon, mutta R-indikaattorin estimaatit olivat kaikille aineistoille liki samat. Suurempi vastausaste ei siis merkinnyt parempaa edustavuutta. Lisäksi CAVVI-aineistossa muistutusviestein ja -kirjein suoritettu vastausasteen kasvattaminen huononsi edustavuutta R-indikaattorin näkökulmasta. Mielivaltainen vastausasteen kasvattaminen ei siis ole välttämättä perusteltua. R-indikaattorin estimaattorin ominaisuuksien osalta empiiriset tulokset vahvistivat RISQ-projektin aiempia tutkimustuloksia. Estimaattorin arvo oli sitä pienempi mitä enemmän vastausalttiuden mallissa oli selittäjiä, koska tällöin vastausalttiuksien varianssi kasvoi (Schouten ym. 2009). Otoskoko vaikutti merkittävästi varianssin suuruuteen: mitä pienempi otoskoko oli, sitä leveämmät olivat luottamusvälit ja sitä vaikeampi oli tehdä johtopäätöksiä edustavuudesta.

Proofs, Paradoxes, and Probabilities : The Logical Turn of Philosophy in Finland

This monograph describes the emergence of independent research on logic in Finland. The emphasis is placed on three well-known students of Eino Kaila: Georg Henrik von Wright (1916-2003), Erik Stenius (1911-1990), and Oiva Ketonen (1913-2000), and their research between the early 1930s and the early 1950s. The early academic work of these scholars laid the foundations for today's strong tradition in logic in Finland and also became internationally recognized. However, due attention has not been given to these works later, nor have they been comprehensively presented together. Each chapter of the book focuses on the life and work of one of Kaila's aforementioned students, with a fourth chapter discussing works on logic by authors who would later become known within other disciplines. Through an extensive use of correspondence and other archived material, some insight has been gained into the persons behind the academic personae. Unique and unpublished biographical material has been available for this task. The chapter on Oiva Ketonen focuses primarily on his work on what is today known as proof theory, especially on his proof theoretical system with invertible rules that permits a terminating root-first proof search. The independency of the parallel postulate is proved as an example of the strength of root-first proof search. Ketonen was to our knowledge Gerhard Gentzen's (the 'father' of proof theory) only student. Correspondence and a hitherto unavailable autobiographic manuscript, in addition to an unpublished article on the relationship between logic and epistemology, is presented. The chapter on Erik Stenius discusses his work on paradoxes and set theory, more specifically on how a rigid theory of definitions is employed to avoid these paradoxes. A presentation by Paul Bernays on Stenius' attempt at a proof of the consistency of arithmetic is reconstructed based on Bernays' lecture notes. Stenius correspondence with Paul Bernays, Evert Beth, and Georg Kreisel is discussed. The chapter on Georg Henrik von Wright presents his early work on probability and epistemology, along with his later work on modal logic that made him internationally famous. Correspondence from various archives (especially with Kaila and Charlie Dunbar Broad) further discusses his academic achievements and his experiences during the challenging circumstances of the 1940s.

FIN-CLARIN: A Framework for Depositing and Disseminating Language Resources for R&D

Researchers and developers in academia and industry would benefit from a facility that enables them to easily locate, licence and use the kind of empirical data they need for testing and refining their hypotheses and to deposit and disseminate their data e.g. to support replication and validation of reported scientific experiments. To answer these needs initially in Finland, there is an ongoing project at University of Helsinki and its collaborators to create a user-friendly web service for researchers and developers in Finland and other countries. In our talk, we describe ongoing work to create a palette of extensive but easily available Finnish language resources and technologies for the research community, including lexical resources, wordnets, morphologically tagged corpora, dependency syntactic treebanks and parsebanks, open-source finite state toolkits and libraries and language models to support text analysis and processing at customer site. Also first publicly available results are presented.