The Condor 107:765±774 q The Cooper Ornithological Society 2005 FIRST DOCUMENTATION OF COMBINATORIAL SONG SYNTAX IN A SUBOSCINE PASSERINE SPECIES DANIEL W. L EGER1 Department of Psychology and Nebraska Behavioral Biology Group, University of Nebraska, Lincoln, NE 68588-0308 Abstract. Birds with songs having two or more acoustically distinct elements can arrange them either rigidly (i.e., in the same sequence) or ¯exibly. Flexible song syntax can be achieved either by varying the number of repetitions of elements or by combining elements in different ways. Combinatorial syntax has been documented only in the songs of oscine passerines and in one nonpasserine, but not in the suboscine passerines. Dawn and day songs of a tyrant ¯ycatcher, the Flammulated Attila (Attila ¯ammulatus), were recorded in Costa Rica. Flexible syntax was noted in both dawn and day song. Attilas not only varied the number of repetitions of their song elements but also combined elements in various ways. This appears to be the ®rst reported case of combinatorial song syntax in a suboscine species. Key words: Attila, song, suboscine, syntax, tyrant ¯ycatcher. Primer Registro de Sintaxis Canora Combinatoria en una Especie Paserina Suboscine Resumen. Las aves canoras que poseen dos o maÂs elementos acuÂsticos distintivos pueden acomodarlos ya sea de modo rõÂgido (i.e., en la misma secuencia) o de modo ¯exible. La sintaxis ¯exible de cantos puede alcanzarse ya sea variando el nuÂmero de repeticiones de los elementos o combinando los elementos de diferentes modos. La sintaxis combinatoria ha sido documentada soÂlo en los cantos de paserinos oscines y en un ave no paserina, pero no en paserinos suboscines. Los cantos del amanecer y diurnos de un tiraÂnido atrapamoscas, Attila ¯ammulatus, fueron registrados en Costa Rica. Encontramos evidencia de sintaxis ¯exible en cantos del amanecer y diurnos. A. ¯ammulatus no soÂlo varioÂelnuÂmero de repeticiones de su canto, sino que tambieÂn combino elementos de varios modos. Este parece ser el primer registro de sintaxis canora combinatoria en una especie suboscine. INTRODUCTION other extreme, the singer may change song types Bird song exhibits organization at all levels of each time it sings a song. its expression, ranging from the most basic units Notes, syllables, and phrases, which I gener- of sound production to repertoires of songs (No- ically will call ``song elements,'' may be ordered wicki and Podos 1993). By ``organization'' I am relative to each other. Among bird species hav- referring to the predictability with which song ing more than one type of song element, the se- elements or song types follow others. The most quencing of the elements can be either rigid or basic level of song organization consists of ¯exible. In rigid syntax, transitions from one el- notes, which are usually de®ned as continuous ement to the next are perfectly predictable. tracings on a sound spectrogram. If notes con- Some variability from one song to the next can sistently are associated with other notes, they occur even with rigid syntax, but it comes about form a syllable, and syllables or notes may be either by terminating the song at various points, combined to form a phrase. A song consists of or by shifting the frequencies of the notes, both one or more different notes, syllables, or phrases of which occur in the Nightingale Wren (Micro- that may be combined with one another in var- cerculus philomela, Leger et al. 2000). Rigid ious ways to make different song types. Finally, syntax can occur even in species that have song a song type may be repeated several times be- repertoires, provided that each song type has dif- fore switching to a different song type, or at the ferent elements and that element sequencing is invariant (Hailman 1989). Flexible syntax is present whenever uncer- Manuscript received 10 March 2005; accepted 1 Au- tainty exists in the sequencing of song elements. gust 2005. Stated differently, there may be points in the 1 E-mail: [email protected] song where the singer has a ``choice'' as to [765] 766 DANIEL W. LEGER which element will occur next. Choices can be are rare. The large number of song types that of two types. First, song elements may be re- result is still only a fraction of the total possible peated a variable number of times before mov- number (Howes-Jones 1985). Ficken and Popp ing on to the next element. For example, in one (1992) have described a similar arrangement in song the bird may sing AAABBCC but in the the gargle vocalization of Black-capped Chick- next it may sing ABCC, where letters refer to adees. acoustically distinct song elements. This form of In some cases combinatorial syntax is ¯exible syntax is similar to rigid syntax in that achieved by repeating the song without the nor- the same sequence of elements is maintained mal intersong interval. For example, Blue- from one song to the next. throated Hummingbird (Lampornis clemenciae) Second, ¯exibility may be achieved by com- type 1 song consists of ®ve units, designated A± bining elements in various ways. Such combi- E. After singing the E unit the bird usually ter- natorial syntax (sensu Hailman 1989) can be minates the song but sometimes immediately re- achieved through different processes. For ex- peats units B, C, D, and E. In this case, the re- ample, the Black-capped Chickadee (Poecile peated units are accompanied by omission of the atricapillus) employs combinatorial syntax in its A unit (Ficken et al. 2000), a phenomenon also call system and does so by omitting one or more of the elements. The sequencing is always in the known to occur in Wood Thrush (Hylocichla order ABCD, but one or more of the elements mustelina) song (Dobson and Lemon 1979). may be omitted. Black-capped Chickadees also Combinatorial syntax has been referred to as vary the number of repetitions of the elements a procedural hierarchy (Todt and Hultsch 1998), (Hailman and Ficken 1986). Similar observa- branching (Dawkins 1976, Gil and Slater 2000), tions have been made on the Mountain Chicka- and as path organization (Ficken et al. 2000). All dee (P. gambeli, Bloom®eld et al. 2004) and the the avian examples of combinatorial syntax Mexican Chickadee (P. sclateri, Ficken et al. would be classi®ed as ``®nite state grammars,'' 1994), among others. and although some of them are similar to the Another method of combinatorial syntax oc- ``phrase structure grammar'' found in human curs when the bird has a pool of song elements language, the similarity is thought to be only in its repertoire that can be strung together in super®cial (Fitch and Hauser 2004). Examples different combinations or permutations. For ex- of combinatorial syntax are fairly common ample, Winter Wren (Troglodytes troglodytes) among oscine passerines, but seem to be ex- song phrases, which can be several seconds in tremely rare in other taxa (but see Ficken et al. duration and consist of many notes, are inserted 2000). in various positions in the song, resulting in very In this paper, I describe the ®rst case of com- large song repertoires (Kroodsma 1980). binatorial syntax in a suboscine passerine, the Lengthy and combinatorially complex songs Flammulated Attila (Attila ¯ammulatus), a Neo- also occur in several other species, such as the tropical tyrant ¯ycatcher. This species varies its European Starling (Sturnus vulgaris, Adret- repetition of song elements, but also combines Hausberger and Jenkins 1988). its song elements in varied ways. Attila songs Some elements are more likely than others to are longer than those of most other tyrant ¯y- occupy the ®rst position in a song, while other catchers. Attilas sing two acoustically distinct elements tend to occur in middle or terminal po- song types, one of which is sung primarily at sitions (Catchpole 1976, Zann 1993). Further, some sequenced pairs of elements are more dawn and dusk and the other primarily during common than others. These tendencies mean the day (Skutch 1971). Leger and Mountjoy that the observed number of unique song types (2003) described geographic variation in the vo- is much lower than the number that could be calizations and morphology of Bright-rumped produced if the available elements were com- Attilas (A. spadiceus, sensu AOU 1998) and bined in an unbiased manner. For example, War- concluded that the differences between birds in bling Vireo (Vireo gilvus) song begins with one the northern (Panama to Mexico) and the south- of a small number of notes in the bird's reper- ern (South America) parts of the range were toire, followed by various combinations of other great enough to justify recognition of the north- notes, some of which are common while others ern birds as a distinct species, A. ¯ammulatus. COMBINATORIAL SONG SYNTAX 767 METHODS species); these were not strong enough to permit I recorded songs at the La Selva Biological Sta- reliable spectrographic analyses, but were ade- tion (108269N, 838599W) in northeastern Costa quate to describe diel patterns of singing. To avoid the statistical problems that might Rica during March 1998 and March 2000, and arise from treating songs from the same bout as at the Las Cruces Biological Station (88479N, independent data points, I calculated the mean 828579W) in southwestern Costa Rica in March for each variable for each bout before using 2002. I also analyzed recordings made by D. J. these values to compute statistics for the entire Mountjoy at La Selva in 2000. dataset. Values reported are means (6 SD). The 1998 recordings were made with a Sony TCM5000EV cassette recorder and a Sennheiser RESULTS ME67/K6 microphone. The recordings in 2000 were made with Tascam DA-P1 and Sony TCD- SINGING BEHAVIOR D8 DAT recorders, both of which were equipped The earliest singing began at 05:15 and the latest with Sennheiser ME67/K6 microphones.