Sounds Wild and Broken, page 17
In the white-crowned sparrow, the extent of geographic variation in song depends on the migratory behavior of the birds. On the California coast, where the birds live year-round on stable territories, the songs are structured into small neighborhoods, sometimes of no more than a few territories. All the birds within a neighborhood share a similar pattern of whistles, buzzes, and sweeps, although each male adds a signature of his own. This fine-grained sectarian geography of sound is, like the tree-centered cultural world of yellow-rumped caciques, a product of the behavior of incomers, itself a result of the sexual preferences of females and the rules of territorial engagement with males. When a young male establishes his first territory, he must match the style of his song to fit with social norms, a strong pressure to conform. Each neighborhood was likely founded when vegetation grew back after fire cleared the land and drove out resident sparrows. Sparrow colonists to the rejuvenated habitat brought their own quirks of song, which were then passed down as cultural variants particular to each patch. The small sizes of cultural units on the coast, then, are a product of small-scale disturbances that create a patchwork of song types. In the years without fires or other calamities, song differences within each patch are maintained by vocal learning within a conformist sparrow society.
White-crowned sparrows from the mountains or the edge of the boreal forest migrate south every winter and don’t live in highly stable communities. Their songs vary from place to place but on the scale of hundreds of kilometers, not tens of meters. This is a pattern typical among mobile bird species with wide distributions. One of the joys of traveling is to hear the regional variants of each bird species. Songs that are familiar at home take on new inflections or add peculiar elements when we step outside our familiar paths. This geography of sound varies in its scale and texture by species, depending on the particular balance of creativity and conformity of each. It is often the homebodies, those that seldom move and whose young settle nearby, that have the most compact and parochial geographic distributions. A morning walk in the San Francisco Bay Area takes us through several neighborhoods of white-crowned sparrows. To hear the same for the song sparrow, we’d need to drive several hundred kilometers. The white-crowned sparrow has no distinct regional dialects, although a novel song variant that turns ohhh-sweet-canada-canada into ohhh-sweet-cana-cana has been sweeping across the continent in the last two decades, the rapidity of its spread helped by the wide-ranging migratory behavior of the birds.
These geographic variations of birdsong, regardless of scale, are usually referred to as dialects. But this word is perhaps loaded with too much human meaning to help us hear the many layers of cultural variation among vocal learning birds. The caciques innovate and change on a week-by-week basis, and every colony tree has its own changing flow of popular sounds, more like Top 40 music playlists than a dialect. California white-crowned sparrows cram more patois into a small area than even the most highly structured human languages. White-throated sparrows sing with such consistency across their range that the new variant is perhaps analogous to the spread of a single idea or catchphrase.
Culture, then, can diversify sound, in forms unique to each species. In doing so, culture combines its powers with those of genetic evolution. In white-crowned sparrows, for example, the rate of trilling is partly a product of culture and partly the result of the genetic evolution of beak size. Birds trill where this sonic ornament is popular and stick to whistles and sweeps where it is not, a learned behavior. Birds with large beaks—a result of genetic adaptation to local foods—cannot trill very fast and so their song is partly a reflection of their beak size, a character largely shaped by genes.
Vocal culture can also fold back into genetic evolution. White-crowned sparrows in coastal California settle into stable neighborhoods as early as their first autumn. They therefore need to rapidly slot into the sounds of the home where they will likely live out their lives. Sparrows in the mountains, though, migrate away from their parents’ nesting site, then return in spring not to their place of hatching but to a breeding site whose location the young birds cannot predict. They’ll settle, depending on the vagaries of opportunity and chance, in one of a wide range of locales spread across the vast breeding range of the species. The brains of each population of white-crowned sparrow have evolved learning mechanisms to match the demands of their life histories. Coastal birds start learning songs relatively late, extending their learning into the autumnal period, when they match their songs to the new territory. Their learning is focused and accurate, picking out the single best option for their social context. Birds from the mountains learn song earlier, picking up a variety of potential songs during the few weeks available to them between hatching and when they migrate. They remember this wide variety and, when the time comes, practice multiple variants, settling on their adult song when they arrive in their breeding territory. These differences in the timing and breadth of learning persist among captive birds in the lab, indicating that evolution has shaped the nervous system of each to match the social context in which its song is delivered. Sparrows from different populations also have a genetic predisposition to pay attention to and learn songs from their own region, a preference that presumably evolved to help them focus on the most relevant and useful sounds. Genes make culture possible, by providing a blueprint for animal bodies that are able and eager to learn. Culture, once developed, then shapes genes, favoring the blueprints most suited to their cultural milieu.
The most dramatic way that culture might affect genetic evolution is by causing species to split. Songs during the breeding season serve both to connect animals with similar preferences and songs and to exclude those with different tastes and vocal displays. As with genetic evolution, if animals with similar preferences and songs stick together, these sexual dynamics can cleave a population, producing two or more gene pools. Over time, these differences can create new species. It makes no difference whether the inheritance of the preferences and songs is genetic or cultural; what matters is whether a link develops between the forms of the songs and the sexual preferences for them. If they do, populations can be broken into cliques that breed among themselves but not with others.
For more than half a century, scientists have studied the question of whether song learning can cause speciation. Their work reveals that cultural differences in birdsong types are widespread, but these are only occasionally associated with genetic differences among populations. White-crowned sparrows provide one of the clearest examples. In northern California and southern Oregon, the resident population of the California coast meets the migratory population of the Pacific Northwest. Each has its own song “dialect,” with the northerly birds singing with longer whistles and shorter sweeps and trills. Playback experiments show that birds respond more vigorously to songs of their own dialect, suggesting that shared songs unite each population and keep them separate from others. But in border areas where the two populations mingle, these behavioral differences were slighter. This suggests that although cultural differences in song do seem to keep populations apart, this force may weaken in areas of extensive contact.
Vocal learning also affords a degree of flexibility that connects divergent populations, delaying speciation. Female birds sometimes prefer the song types of their home region, but this preference is not universal and can be erased by exposure to other song varieties. In a neighborhood near San Francisco where song types are uniform, females will therefore likely enforce conformity, preferring familiar songs. Farther north, on the Oregon border, females hear many song types and will have more catholic and flexible tastes, potentially selecting males from other areas. For males, too, culture can smooth over geographic differences. By molding his song into the style of the neighborhood, a young male setting up his first territory can partly break free from his parental inheritance. He’s stuck with his genes but can find a new vocal identity through learning.
In addition to its role in promoting or slowing the evolutionary splitting of populations, vocal culture can make endangered species more vulnerable to extinction. If population densities drop too low, it is harder for animals to find one another and young birds fail to learn the species’ full songs. In the Blue Mountains of Australia, for example, the population of the regent honeyeater, a black-and-gold nectar-drinking bird, is down to just a few hundred birds. In recent years, many of these birds have started singing atypical songs, including the songs of other species. Compared with recordings from previous decades, contemporary birds also sing simpler songs. Lacking suitable tutors, young birds pick up snatches of sound from other bird species or invent their own songs. Males with these malformed and often stunted songs are less attractive to females. At the edge of extinction, then, social learning of song can become a liability. As Hawaiian honeycreepers on the island of Kaua’i declined in numbers, the diversity of their songs plummeted, likely due to the loss of social connections that formerly sustained the cultural richness of song learning. In endangered whales too, it seems that cultural diversity is lost when populations shrink. Such losses have been heard in endangered and declining sperm and orca whales. But we have no record of the vocal diversity of pre-twentieth-century whales, and so the full extent of the loss is unknown. The decline of vocal diversity may have been severe among those species that were scythed down to 10 percent or less of their former abundance.
Among all nonhuman animals with vocal learning and cultural evolution, the white-crowned sparrow is one of the best understood. The species’ geographic variation in sound is obvious even to human ears unaccustomed to parsing the details of birdsong. The species offers us an imaginative window into the possibilities of culture in all vocal learning species, most of them unstudied by science. Wherever vocal learning happens, cultural evolution can unfold, driven either by the creative impulses in animal minds or by the simple accumulation of copying errors as each generation learns from its elders. These cultural changes cause sound to change through time and to fan out in a richly textured geography.
Birds offer the most well-studied examples, but geographic variation is common among other vocal learners such as marine mammals. New song variants of humpback whales, for example, spread in just months across entire ocean basins, often originating with whales that live in an innovation zone—an incubator for whale sonic creativity—off the coast of Australia, then spreading worldwide. Why one patch of ocean should be the origin of so much new whale song is unknown, as are the causes for the sudden spread among whale singers of one song variant and not others. Cultural variation in the sounds of toothed whales, like sperm whales, orcas, and dolphins, reveals subtle hierarchies of affiliation within each species, from parent to offspring, to clans, to large regions. Sperm whales, for example, live in matrilineal groups that range over thousands of kilometers. These matrilines remain stable over decades, likely held together by shared patterns of vocalization, learned by youngsters from elders within each group. Sperm whales communicate with short bursts of loud clicks. When the whales are close together, these pulses of clicks are like the excited chatter of human friends gathering at the weekend, overlapping one another in a frenzy. Individual whales seem to have distinctive voices or accents—unique ways of using click groupings. This individuality is embedded within a larger spatial and social structure. Matrilines have their own distinctive clicking styles and are themselves part of regional “dialects.” In the Pacific, these dialect groups overlap in range, but whales within each do not associate with one another, seeming to disdain the company of whales with the “wrong” way of clicking. In the Atlantic, whales in each dialect group stick to their own nonoverlapping subregions of the ocean. When a sperm whale clicks, other whales presumably can immediately identify its region, family, and individual identity, just as we humans can infer the identity and biography of those we hear speak.
Sometimes cultural evolution crosses species boundaries. Parrots, lyrebirds, mockingbirds, and many other birds take snatches of the sounds of other species and weave them into their own sonic creations. In the case of Australian lyrebirds, these sounds are then passed down through the generations culturally. When lyrebirds were introduced by humans to Tasmania in 1934, they remembered and repeated the song of the whipbird as part of their mimicking display, even though whipbirds did not live in their new home. Thirty generations later, descendants of the colonist lyrebirds still sang the whipbird song, passed down to them by previous generations of lyrebird.
Nonhuman animal sounds also jump the fence into the culture of our own species. When recordings of humpback whales inspired a generation of ecological activists, when musicians from Sibelius to Pink Floyd weave bird sound into their creations, when our onomatopoeic verbs croak, twitter, and bellow, or when police sirens evoke the howl of wolves, fragments of animal sound from other species lodge into human imagination and spread through our own webs of listening, remembering, and responding.
Cultural evolution connects animals—both within and across species—in networks of learning more expansive than the parent-to-child inheritance. This weblike flow of information reawakens flexibility of evolution that the DNA of vertebrate animals has lost. Billions of years ago, our bacterial ancestors exchanged genes promiscuously through their watery surrounds, a back-and-forth that carried the DNA of one cell into another and back again. These movements were unconstrained by the rituals of sexual cell division and parental inheritance that would later control the genetics of complex animals. Cultural evolution, in breaking free from the rules of genetic inheritance, regains this lost rapidity and fluidity of evolution, allowing behaviors to jump from one animal to another through the process of learning. There are limits, of course. Genes and anatomical constraints set bounds on what animals attend to and copy. Sparrows will not learn the calls of ravens, and whales do not mimic toadfish. Within these boundaries, cultural evolution samples, remixes, and connects, reclaiming a little of the evolutionary nimbleness of our bacterial ancestors.
* * *
—
Songbirds and humans last shared a common ancestor more than 250 million years ago. The brains of birds and mammals have each taken their own paths since this split, resulting in parallel worlds of sensation and experience. Birds cram a higher density of nerves into their skulls than mammals, giving their small brains as many cells as much larger primates. The folds and layers of the forebrain have different geometries, hierarchically layered in mammals and clustered into nodes in birds. But despite the long separation of our lineages, our vocal learning converged on some similar processes. Social learning has some universal qualities.
The first of these parallels is evident when we hear human infants and young birds babble. My parents tell me that half a century ago I could not manage the sophisticated tongue and lip movements needed for cat and chocolate, and so in my infant voice felines were vuff and treats were clockluck. The trills of white-crowned sparrows are likewise beyond the capabilities of youngsters, and so the young birds squeak and waver, gradually building proficiency. But motor control is not the only aspect of maturation. The order, pacing, and form of sound in young birds and humans are more diverse than those of adults, coming in streams unconstrained by the rules that allow meaning to be conveyed. Maturation prunes this wide-ranging juvenile sound into precise adult forms. Vocal learning also gets harder as birds and humans age. An older white-crowned sparrow will not pick up new songs. Human adults struggle to grasp the rudiments of new languages, even though as infants we readily master any language we are immersed in.
The winnowing and clipping that birds and mammals experience when learning to vocalize also shape forms of growth and maturation in other beings and at other time scales. Twigs on a tree ramify in tens of thousands of directions. Only a few mature into stout branches, and the rest are dropped, food for worms. Animal bodies develop partly through expansive early growth later trimmed by the programmed death of cells. Evolution by natural selection first increases genetic variability through sex and mutation, then narrows this range of possibility as the physical and social environment picks out winners. The words on this page, too, are the few left behind after countless others were culled, along with hundreds of permutations of narrative and analogy. Arthur Quiller-Couch’s oft-quoted advice to writers, “Murder your darlings,” was inadvertently an insight into many of life’s creative processes.
In both birds and humans, vocal perception and memory are controlled by different parts of the brain than vocal production. Listening, memory, and action are each sequestered into their own spaces, and their activities are similar in humans and birds. The perception centers of the brain are tuned by unknown means to the sounds most relevant to each species. These centers feed sonic information to the parts of the brain that control muscles and nerves. Underlying these feedback loops in the brain are the genes that build the brain. The FOXP2 gene that is so important for human speech is also important in the early development of vocal learning pathways in the songbird brain.
When sparrows and human infants babble, we hear a deep-buried unity. The same genes build parts of the nerve network needed for vocal learning in humans and songbirds, despite the very different forms of the mature brains of humans and songbirds. The patterns and processes of learning, too, are similar. To smile when we hear the tumbling, inchoate songs of birds is not mere sentimentality, then. The pleasure rising within us is a reminder of kinship across difference.
