Studia Metrica et Poetica sisu 2_1.indd Reuven Tsur Playing by Ear and the Tip of the Tongue Amsterdam/Philadelphia, Johns Benjamins, 2012 Eva Lilja Reuven Tsur created cognitive poetics, and from 1977 on his perspectives greatly shaped the fi eld. His chosen problems explain many of the classical questions within aesthetics. English cognitive stylistics, however, which more strictly follow the cognitive discoveries of Lakoff & Johnson, look more limited by comparison. Specifi cally Tsur’s poetry analyses are both enlightening and fi nished with a certain feeling for the poem. His criticism always pays atten- tion to artistic qualities before dry theory, and this focus is something that strengthens his credibility. His new book Playing by Ear and the Tip of the Tongue (2012) continues in many ways his What Makes Sound Patterns expressive? (1992), with parts of the older book even being repeated in service of those who haven’t read it. Th e somewhat peculiar title refers to what Tsur labels the TOT (Tip of the Tongue) phenomenon – when you have a word on the tip of your tongue but cannot quite get hold of it. Almost all parts of the word are at hand but one detail is missing. His point is how words consist of many parts, phoneti- cally and semantically, that to some extent act separately. A word is a stable confi guration out of long time memory consisting of a great many distinctive features, anyone of them potentially changed or manipulated. Th is is used within poetic language. Speech sounds as well consist of clusters of qualities in combination – for example a sound might be opened or closed. Diff erent combinations give rise to various meanings. Th is is a main theme of the book – no linguistic phe- nomenon is unambiguous because they all consist of combinations of a great many small characteristics. As the preface says, this book explores how poetic language attempts to escape the tyranny of conceptual and phonetic categories. Th e specifi c poetic eff ect originates from disturbances in cognitive pro- cesses, Tsur says. Speech sounds are in actuality complex, but uncomplicated when listened to and categorised. Still, certain devices of poetic language delay categorisation in order to facilitate precategorial information. In that way, poetry creates meanings which are both unexpected and optimal. Normally categorisation takes place at lightning speed, but the poetic language possesses doi: dx.doi.org/10.12697/smp.2015.2.1.07 Studia Metrica et Poetica 2.1, 2015, 134–139 dx.doi.org/10.12697/smp.2015.2.1.07 135Review articles tricks to delay it and provoke emotion. Th e good reader perceives phonetics as well as semantics in a text, something that will take some time. And aff ects need time to incubate, growing emotions that will be indispensable elements in the experience of that poem. Precategorial information Reuven Tsur has developed gestalt psychology so it can be relevant for ver- sifi cation studies. He is one in the great tradition from Arnheim, Cooper & Meyer and Barbara Herrnstein Smith. His fi ndings are just as important for free verse as for the metred verse that he himself mostly prefers. You might say that poetic forms consist of gestalts – according to the standards of gestalt psychology. More than that, Tsur also postulates a background mumble, that is not categorised but full of undetermined meanings originating out of the context – this context might be semantic as well as acoustic. Th is background murmur is said to be ‘thing-free’, or ‘shapefree’, and lacking gestalt. A precategorial meaning production should precede the gestalt. Th e book aims at explaining the relations between precategorial information and its semantic and phonetic circumstances. Tsur’s theory emanates from an era before the break through of brain research, which is why you should be careful with the results. Later psycholo- gists have discussed so called cognitive schemas applying patterns or gestalts to all information that could be perceived. Th e stability of such schemas has also been subject of debate: are they stable structures of the long term memory or more of momentary constructions? My experiences from versifi cation studies and verse history speak for the stability of such cognitive schemas. When a form pattern is established it will prevail in its culture. An example is the tactus or metre, that arised within the Germanic languages at the threshold of the Modern Age. Tactus is still a strong force in the mind – but nowadays perhaps more a pattern for poets to oppose. Th ere are patterns of form and patterns of culture or meaning. Th ose who mean that cognitive schemas are more temporary constructions seem to refer to cultural patterns and their semantic meanings. Th is begs the question: are form schemas more stable than the categories of culture? Th is is to be further investigated. An important idea in gestalt psychology is the fi gure-ground scheme. Tsur’s conception of a murmur below the gestalts is analogous to it, and he gives 136 Review articles extensive room for discussing examples of this model. However, the fi gure- ground scheme shows up to be rather unstable when applied to poetry. Tsur argues that it operates more like in music – sometimes ‘ground’ is missing or ‘ground’ turns out to be ‘fi gure’. But perhaps this scheme is not relevant for music and poetry? Tsur imagines gestalts fl oating upon uncategorised information that nev- ertheless slips into consciousness. I agree that perceptions are premodal, but does that really mean they lack gestalt? My experiences from versifi cation studies say that the settling of form patterns in a fi rst step is premodal, and that details are added later on when the pattern has found its modality. With such a model, perceptions that are not patterned just disappear. Roman Jakobson, however, claimed the existence of subliminal signifi cation. Th is is where he comes close to Tsur’s idea of uncategorised ‘thing-free’ sounds. But maybe the background mumble also possesses some kind of form? Also, as Tsur correctly points out, phonemes are coded according to the acoustic context. A speech sound is pronounced (and coded) depending on the surrounding sounds and their meanings. Th e background murmur seems to function in a similar way. Th en is it really lacking gestalt? Th e poetic text is spatial as well as temporal. In his Poetic Rhythm, Tsur shows how the gestalt’s second limit contributes to reshape the gestalt, by so called back-structuring. Only when the gestalt is closed you know its form for sure – it is, so to say, structured or understood backwards. Th is is one of several devices that spatialise the text giving it a quality of balance by suspending the fl ow of time. In other words, this quality adds to the concentration and charged signifi cation of poetry. Modalities act in diff erent ways. Tsur brings qualities from music and image into the art of poetry testing the results; furthermore, he is tracing the play between modalities. Sight appears to be the strongest and most dif- ferentiated of the senses, and visual gestalts are stable and diff erentiated in comparison. Th ey are rapidly categorised, while acoustic gestalts typically are unique, undiff erentiated and miss adequate descriptions – in other words they are signifi ed by delayed categorization. Some modalities don’t have a working terminology but use descriptive terms from, primarily, visuality. Th at means that temporal lapses oft en are described in spatial terms. Sight possesses most descriptive expressions, aft er that comes hearing, but tactility seems to be woolly. Meanwhile, less diff erentiated senses borrow expressions from the more diff erentiated ones. Here we have a good explanation for the synaesthesia of common everyday language. Sound is mostly described in spatial expressions such as ‘high-low’ and so on. But Tsur shows us how acoustic details are really understood. 137Review articles A sound has three physical dimensions: rapid – slow, broad – narrow (both aiming at the vibrations of the sound wave) and thick – thin (aiming at the source of the sound). He also shows how the formants of a sound infl uence semantics. For example, the formants of a ‘g’ are situated where you also fi nd metallic sounds – in that way the ‘g’ gets a metallic quality. More than that, diff erent aff ects have their typical pitches. Th e intonation of happiness is marked by big jumps with rounded tops. Anger jumps too, but the curves are somewhat lower and sharper in form. Tsur also repeats what we already know – small things cooperate with high frequancy and big things with low frequency, and so on. While interpreting poetry, facts like these explain a great many ‘subliminal’ meanings. Words have very many more signifi cations than those listed in the dictionary. Brain halves Tsur’s measurements at Haskins Laboratories around 1980 are of epoch-mak- ing importance. Today, however, they are thirty fi ve years old and should be revised in the light of modern neurology. Brain research has developed rapidly, and new results are continuously knocking at the door. A poem can be listened to and looked at, and now neurology discusses the relationship between sight and hearing, and how those senses are supported in the brain. Among other things, it has been proposed that the sense of hearing has less room than that of sight, or that the temporality of hearing depends on sight qualities. Visuality is spatial, and the superiority of sight might add spatial properties to the poem in spite of its basic temporal lapse. Around 1980 the general knowledge of the diff erent functions of the brain halves was agreed upon – the left half contains for example language and logic, while the right half is responsible for arts and emotions. Th e measurements at Haskins show that poetry uses both halves – this is something that gives it a unique position in human cognition. Th is double position explains some aesthetic questions and at the same time creates new ones. Sound might be coded in a speech mode (on the left side) as well as in a nonspeech mode (on the right). Th e speech mode is rapidly categorised – one hears something other than one really hears. In the nonspeech mode, however, one perceives the very sound, the qualities of the sound wave. Poetry allows us to take part with both modes, the coded meaning as well as the very sound – certainly a wonderful art form. In my studies of aesthetic rhythm, I have distinguished two dominating rhythmic movements, balance and direction. 138 Review articles Even these two seem to refer to diff erent brain halves. Direction goes to the left together with sequence and time, but balance belongs to the right half with space and relations. Th ese circumstances once more underline how poetic language works with time as well as space, something which partly explains the special character of poetic language. Not least the interesting facts Tsur presents about abilities of the brain raises the question of the age of his reference literature – it is oft en fi ft y to sixty years old. It is like repeating my own intellectual history – I remember when I as a young student lost myself in Wellek & Warren’s Th eory of Literature (1949) or Ullmann’s Principles of Semantics (1957), both of which are important for Tsur. But much has happened during these sixty years. Cognitive poetics has another theoretical base other than structuralism, and I am looking forward to the discussion of the connection between them. Poststructuralim is missing in Tsur’s impressive reference list, and I would like to have it explained what motivated its absence. Tsur is a great theory builder, he creates theory and has no duty to prove every step he takes, but I need more background. What is the relation between structuralism and Lakoff & Johnson’s Philosophy in the Flesh? And, how do I come from neurology and get to phenomenology? Signifi cation Patiently, Tsur uncovers layer aft er layer in the poem’s production of mean- ing. Every extra signifi cation has its own technical explanation. For example, repetitions add extra meaning because of their similarities – repetitions create similarity – they mix and disturb the rational lapses of language. Th e reader is forced to abandon the rational principle of succession for the emotional principle of similarity, and the properties of the text collide in a confusion of sound and content. Aft er this book, interpreting a poem will be arduous, sweaty work. Tsur elucidates ways between sound and meaning, and these connections are com- plicated. Meaning production takes place according to several models. One of them is iconicity in a broad sense of sound symbolism and structure resem- blance. Aff ects have their typical energy curves – the same curves that are seen in the poetic text. Brain mechanisms for religious mystery might enter the poem supplying its special signifi cation; synaesthesia of all kinds colour the text. Speech sounds have many potentials for meaning – the one realised depends on the context. 139Review articles In the old days one tried to explain the meaning production of the sound structure with the help of associations. Th ese associations should be intersub- jective. However, Tsur has proven his case that these so-called associations are based on (many subtle) facts, and this is something that means a change in paradigm for poetics. Th is book is one from a group where Tsur confi rms such a new paradigm. References Gibbs, Raymond W. Jr 2006. Embodiment and Cognitive Science. New York: Cambridge University Press. Lakoff , George; Johnson, Mark 1999. Philosophy in the Flesh: the Embodied Mind and its Challenge to Western Th ought. New York: Basic Books. Lakoff , George; Turner, Mark 1989. More than Cool Reason: a Field Guide to Poetic Metaphor. Chicago & London: Th e University of Chicago Press. Ullmann, Stephen 1957. Principles of Semantics. Oxford: Blackwell. Wellek, René; Warren, Austin 1949. Th eory of Literature. New York: Harcourt, Brace & World.