Do Bushtits Sing?

Do Bushtits Sing?

(Guest post by Ted Floyd, editor of Birding magazine and bird sound enthusiast)

Fish swim, snakes slither, songbirds sing. Okay, some songbirds’ songs aren’t especially songful: The “song” of the Cactus Wren sounds like an old car starting up, for example, and the “song” of the Yellow-headed Blackbird sounds like someone being strangulated. Nevertheless, the chugging of the wren and the tormented gasping of the blackbird qualify as song, pretty much any ornithologist will tell you.

But some songbirds actually are believed not to sing. Probably the best known example in the U.S. is the Cedar Waxwing. Oh, they’re plenty noisy, flocks of them trilling and lisping and wheezing their way from tree to tree. But they don’t actually sing, it is said. I can see why. I mean, they’re so over-the-top gorgeous, with their waxy wingtips, wispy crests, and just general, well, gorgeousness. One look at a Cedar Waxwing, and it’s love at first sight—whether you’re a human or a waxwing. No need for song.

Less well known is that Bushtits are said not to sing. Like waxwings, Bushtits roam about in noisy flocks. But they’re drab, they’re tiny, they’re not as widespread in the U.S. as Cedar Waxwings, and much of their range comprises hot, arid, rocky, low-diversity shrublands eschewed by many birders. Out of sight (and out of earshot), out of mind.


Of late, I’ve had occasion to question the conventional wisdom that Bushtits do not sing. Here’s the deal: In just the past couple years, Bushtits have invaded my suburban neighborhood near Denver, Colorado. Suddenly, Bushtits are everywhere. There’s a certain novelty for me about seeing and hearing Bushtits. I find myself paying special attention to Bushtits.

In particular, I’ve noticed that one Bushtit vocalization—a short, high-pitched trill—tends to be given by solitary birds or by birds in pairs. Yes, this vocalization can be heard from birds in large flocks and at any time of the year; in the same way, the Red-winged Blackbird’s classic song may be heard in huge roosts on the wintering grounds. I can see from my field notes, as well as from my comments posted to Xeno-Canto, that I’ve been wondering for a while if this short, high-pitched trill functions as the song of the supposedly songless Bushtit. And I had an experience the other day that really got me thinking.


Friday, March 6, 2015, was the first sort-of warm day we’d had around here in a while. Even through my closed window, I could hear House Finches singing their twangy warbles and collared-doves chanting “eighteen” in Greek. (Every wondered why they’re called decaocto?) I could also hear the trilling of a Bushtit, so I went outside to investigate. A Bushtit was teed up on a branch in my yard, vocalizing away. I went back into the house, got my recorder, and came back outside again. The bird was still at the same perch and still vocalizing, and I recorded it. Then I did something absurd.

The male Bushtit apparently singing, Boulder County, CO, 3/6/2015. Photo by Ted Floyd
The male Bushtit apparently singing, Boulder County, CO, 3/6/2015. Photo by Ted Floyd

I went back inside, looked for my scope, and looked for my cell phone. I found them, gathered them up, and went back outside. The Bushtit was still there, still vocalizing. I set up the scope, fired up the phone, fiddled with various settings, and got a bunch of photos. That’s absurd! Bushtits aren’t supposed to stay put for ten minutes, as this bird did. They’re not supposed to stay put for ten milliseconds.

But this guy (confirmed as a male by the digiscoped images) was at his perch, teed up, constantly giving the same vocalization. And when I reviewed the soundfiles, I saw something I hadn’t consciously noted while with the bird in real life: Another Bushtit, quite some distance away, was counter-vocalizing with the bird in my yard.

So we’ve got a bird perched for ten-plus minutes, delivering the same vocalization over and over again; the bird’s a male; it’s a sunny and sort-of warm morning in early spring; and another bird is dueting with it. If these were cardinals or chickadees, we’d unquestioningly say they were singing.

But were the Bushtits “really” singing?

The literature says Bushtits don’t sing. I checked Sarah A. Sloane’s BNA account [subscription required], which states: “There is no song in the usual sense. The closest Bushtits get to a song is the somewhat musical, but apparently unstructured, ‘twitterings’ given by many individuals simultaneously when in a close group.” The sprawling account by Arthur Cleveland Bent doesn’t treat song, best I can tell. And a detailed 1903 treatise by Joseph Grinnell likewise contains no mention of song.

Were the Bushtits “really” singing? I think we need first to address another question: What is birdsong? Immediately, I’m reminded of the famous quip by U.S. Supreme Court justice Potter Stewart: “I know it when I see it.” (Or, in the case of birdsong, hear it.) That’s good enough for jurisprudence perhaps, but inadequate for biology. David A. Spector took up the matter in a 1994 paper in the Journal of Theoretical Biology: “Definition in Biology: The Case of ‘Bird Song.’ ” The problem is that different authorities have different definitions, many of them intrinsically circular.


I’m a scientist by training, but more a philosopher by temperament. I get it, at a scientific level, that the matter of song in the Bushtit is unresolved. It remains to be proved that the male Bushtit’s short, high trill functions as song. I’ve stated a hypothesis; now someone needs to go out and test it.

But I’m temperamentally a philosopher, I said, and I can’t resist ending on that note. The morning after my encounter with the provisionally singing Bushtits, I came upon a Black-capped Chickadee up on a branch. The bird was vocalizing. So was another chickadee, farther away. I made a quick-and-dirty recording; you can clearly hear one bird and sort of hear the other bird, and they’re inarguably singing—singing one of the best known and most beloved songs of any bird in the U.S.

Weren’t the Bushtits doing the same thing?

17 thoughts on “Do Bushtits Sing?

  1. Is it my ears? I cannot hear a thing other than a hiss in the Bushtit recording. I can hear the Chaickadees just fine.

  2. Hi Bill,

    Sorry to say, it probably IS your ears (though there’s a possibility it might be your speakers). Head over to Xeno-Canto and see if you can hear any of the other Bushtit recordings. Have you noticed any trouble hearing that species in real life? It’s very high-pitched so it will be one of the first victims of age-related hearing loss.

  3. It might be your ears, Bill. The Bushtits’ vocalizations are slightly above 7,000 Hz; the chickadees’ are below 5,000 Hz. And the Bushtit cut here is “cleaner” than the chickadee cut, i.e., with less hiss. Plus, the Bushtits are, at least in the sense of decibels, louder.

    (The question of the perception of loudness is fascinating to me, and I hope someone takes it up at some point. I ventured a few thoughts about this, a while ago in another forum, and got a fair bit of pushback. We all “know” that Hamerkops are loud and Brown Creepers are soft–yet I’ve made recordings that objectively show the opposite result! Our perception of loudness depends importantly on culture, psychology, and the physics of human hearing. Here’s an early, classic paper on the matter:

  4. Hi Ted, the call you recorded is primarily an alarm call. A good percentage of my Accipiter detections come when I hear a flock of Bushtits trilling and I immediately look up. I have occasionally seen a bird teed up giving this call repeatedly; I always got an “upset” vibe from it and wondered if the bird had become separated from its flock or mate. I would only describe a functional bird song as one related to territory defense or mate attraction, not distress. However, your observation of apparent countersinging suggests that perhaps the alarm call is occasionally co-opted and used as a song, perhaps a developing behavior in Bushtits expanding into new habitats where territoriality is more important (apparently territoriality is quite plastic in some songbirds, like wintering American Redstarts, depending on geography or habitat). Alternatively, perhaps this once WAS a true song and only later became co-opted as an alarm call as social behavior developed. Amazing little creatures–thanks for the post.

  5. Hi, Tristan. I’m really glad you brought this up. In fact, I was going to do. But I’d run out of my 1,000 words for this post.

    Okay, so please check out this recording:

    An accipiter (it was a Sharp-shinned Hawk) blasted through at the 11-sec. mark, and you can hear the Bushtits’ response. And that’s the call I associate with response to a predator. Spectrographically, it is distinct from the vocalization that I conjecture to be the song. Compared to what I treat as the song, it is lower-pitched, the individual notes are strongly down-slurred, there are more notes in the series, and it is subjectively “loud” (but see above).

    I’m reasonably certain (to the extent that one can be certain with Bushtits) that we’re dealing with two different stereotyped vocalizations here. To be clear, the vocalization I conjecture to be song is the one that can be heard (and “seen”) at 1.1 sec., 3.1 sec., 7.8 sec., and 15.5 sec, whereas the vocalization I take to be an alarm call is the one that can be heard (“seen”) at 11.0 sec. and 12.5 sec.

    I assume you’re listening mainly to Bushtits in California west of the Sierra, yes? I was also going to write in my original note–but I ran out of words–that my observations and conjectures in Colorado do not necessarily apply in California… 🙂

    I read the Grinnell paper cited in my main text, and I couldn’t really match some of Grinnell’s descriptions to what I hear and “see” (spectrographically) in Colorado. In his paper, Grinnell states perceptively, “Of course I realize how hard it is to describe bird voices. And also, as I have often had opportunity to note, hardly any two persons receive the same impression of a single birds’ song. No two people seem to hear exactly alike.” 100+ years later we have access to great freeware like Audacity, and we can study the matter much more objectively.

    I don’t mean to be definitive, and I hope I’m not coming across as dogmatic. Alls I’m doing is putting a hypothesis out there, and I’m grateful for the gentle pushback from Tristan. I think the next two questions are:

    1. Alright, so we’re dealing with two different vocalizations (among many made by Bushtits). If one functions as song, that’s interesting. If *both* function as alarms that’s interesting, too; why would the species have two different alarms?

    2. And do observations and conjectures from California even apply to Colorado, and vice versa?

    Hey, I think we can agree on at least thing: The vocalizations of the Bushtit are fantastically varied. I confess, it wasn’t all that long ago that I would have said the calls of Bushtits are more-or-less undifferentiated twittering and sputtering. But thanks to Audacity and Xeno-Canto (and experts like Tristan and Nathan), I’ve learned otherwise. And it’s wonderfully humbling and inspiring to see that Joseph Grinnell figured it out only a few years after the discovery of the electron.

    I have a final question for you, Tristan: When are you going to teach us how to ID LWHGs (“large white headed gulls”) by call? No, seriously, Tristan and coauthors have a fantastic article on gull ID in a recent issue of Birding:

  6. Hi, Tristan & all. Just two more comments from me:

    1. This “counter-singing” (according to my hypothesis) was prolonged. My post above contains a link to, but note that “counter-singing” (if that’s what it is; I’m trying to be careful!) is also occurring in the two other cuts I uploaded to Xeno-Canto: and So this back-and-forth went on for a little while.

    2. I appreciate your insight about behavioral plasticity in expanding bird populations. Earlier, I said it’s important to think about geography (e.g., Colorado vs. California), but I like your way of thinking about ecology and evolution: the adaptive “needs” of expanding Bushtit populations (as where I live) vs. the adaptive “needs” of more sedentary ones (the birds that stay behind?). And maybe proneness to exploration in Bushtits is correlated with vocalization function.

    [A note to local birders: Most of the Bushtit “singing” I’ve been hearing of late in my neighborhood is in and around the backyards (accessible via Open Space) near 39.9962 N, 105.1155 W and 39.9953 N, 105.1193 W. The “singing” birds spill over into my yard only infrequently.]

  7. What I’m hearing in the recording is clearly a long-distance contact call. Bushtits make this call, which carries a long distance (compared to the quiet “spits” they make when foraging together), to stay in touch and to locate each other when lost. When I hear this call, it’s almost always a lone bushtit (male or female) and, when another bushtit is not too far away, you can hear the same response call. That’s very common during incubation as the male and female take turns. The one that has just left the nest spends a good amount of time, using this call, to find the flock.

    The other time I hear this call is when the flock is starting to move. It’s a signal to me that I better pay attention if I want to stay with them!!

    Rarely, I hear this call during courtship and in mild confrontations, coupled with loud “spits.”

    But I have never in the 30 years I’ve been studying bushtits, thought that they actually “sing” in the sense of advertising for a mate or a territory. Their close social relationships throughout the year (including during nesting season) makes that entirely unnecessary. But I could be wrong, of course.

    (And this is definitely not a predator warning call….that is more of a weak descending warble that causes all the surrounding bushtits to freeze!)

  8. BTW, just because I said there is no song (in the BNA account), doesn’t mean bushtits don’t have an extremely varied set of vocalizations! That’s really clear when they are in captivity — quite amazing!

  9. Hi, Sarah. Thanks for contributing your expertise in this forum. Check this out: Right now, this very instant, a presumably mated pair of Great Horned Owls are hooting, audible through my open window. (It’s a mild morning, relatively speaking, here in Boulder County, Colorado.) And I ask the question: Are the owls “singing”?

    I would say so.

    But others would say that owls don’t “sing.” Why, they hoot, of course! But if we allow that owls are in theory capable of singing, do these Great Horned Owls’ vocalizations qualify as song? The male isn’t advertising for a female (because I think the pair bond is established), and I don’t think he’s defending a territory, either. He’s engaging in a behavior not all that different from what you’re ascribing to the Bushtits: He’s staying in touch, if you’ll allow the anthropomorphism.

    Which brings us back to that David Spector paper in J. Theor. Biol. about the very definition of song. Spector starts out with this overview: “Our names for categories are socially constructed handles that allow logical manipulation of those categories.” Does a bird have to sing for the purpose of defending a territory or advertising for a mate? (The answer, at least according to Spector: No, not at all.)

    Moreover, couldn’t a particular class of vocalization serve dual or even multiple functions? At the nearby feedlot, on cold and dreary afternoons in early December, I can go out and hear the unmistakeable songs of Red-winged Blackbirds and Western Meadowlarks. But do those “songs” in fact function as song, at that time of day and year, and in that ecological context?

    (I hasten to point out that Tristan has already said much the same thing. Call function can change in a changing environment. Evolution in action!)

    Well, we all seem to agree that Bushtits have an “extremely varied set of vocalizations,” as Sarah says.

    P.s. And then there’s the matter of “song” in waterfowl on the wintering grounds! Now that’s one that blows apart the category of “song.”

  10. In these comments, I see several mentions about birds in California not necessarily following the same “rules” as the Colorado birds. I have heard ruminations about there being potential for a Bushtit split of interior vs coastal populations, as we see in other bird species (eg, Western Scrub-Jays, Winter/Pacific Wren, Sagebrush/Bell’s sparrows, etc). Are these comments that people are making related to this point, or more to the differences in population expansion as Ted has mentioned for his Front Range birds?

  11. Hi, John. I too was picking up on such “chatter,” or “ruminations,” but perhaps a decade ago. Haven’t heard much, though, in the past couple years.

    As John notes, the (relatively) recent ruminating has centered on a split between Pacific-slope and Interior West populations of the Bushtit. But if you go back well into the 20th century, the split was between the northern Bushtit and the southern (“Mexican,” or “Black-eared”) Bushtit. The historico-taxonomical situation with Bushtits finds an analogy with Snow Geese. Back in the day, the Blue Goose and the Snow Goose were treated as separate species. (And the name of the old Blue Goose, caerulescens, is retained in the specific epithet of the Snow Goose.) But today Lesser Snow Goose vs. Greater Snow Goose is thought to be where it’s at, taxon-wise.

  12. Here’s a Bushtit that appears to be using the short, high trill for precisely the purpose that Sarah Sloane describes:

    At least, that’s a plausible interpretation… 🙂

    However, the vocalization seems less vigorous and less songful, if you will, than the presumptive songster in my yard a week earlier.

    (I videoed this particular bird yesterday in Louisville, eastern Boulder Colorado, a couple miles from the Bushtit in my yard.)

    I still wonder about the possibility that this vocalization functions as song, in certain individuals, populations, and circumstances. (See earlier comments about Great Horned Owls, Red-winged Blackbirds, and Black-capped Chickadees.)

  13. Ted, in one of your responses above, you state the following: “Do these Great Horned Owls’ vocalizations qualify as song? The male isn’t advertising for a female (because I think the pair bond is established), and I don’t think he’s defending a territory, either.” I think you’re overthinking this. I would say few folks would deny that owls sing, and at the same time, I think the functions of the songs you are talking about in the case of your Great Horns are 1) territorial defense (because most birds that have a pair-held territories usually start their “day” by declaring it) and 2) maintenance of pair bond (until the female is on the nest and incubating, Great Horns, as most owls, usually duet). Why would you Say territoriality is not their reason for singing?

    Then you end that thought with:
    “And then there’s the matter of “song” in waterfowl on the wintering grounds! Now that’s one that blows apart the category of “song.”” Again, I’m not sure I follow. In the case of migratory temperate zone ducks, these “songs” are given as part of displays that are meant to exhibit a given male’s traits (usually in the presence of his “competition”) to potential mates. That they are given in winter may be the most unusual aspect to them, but their function is not particularly different from other bird songs. Duck song congregations are a similar idea to a lek of grouse or manakins or what have you. These species all perform songs to attract mates (not necessarily near where nesting takes place), but don’t defend traditional territories, so what is so “apart blowing” about duck songs? They are not traditional “songs” by temperate passerine standards (perhaps the ruler many folks in North America use to determine the meaning of “song”… and one that is rather narrow), but they certainly are not unusual within the broader context of bird vocalizations worldwide.

    I will admit that the concepts of “song” and “call” are quite open to interpretation, and there are many birds where it truly can be difficult to place a label as simple as these on various vocalizations (goodness knows I had such troubles as I was writing the voice descriptions for woodcreepers, furnariids, and many tyrant flycatchers for the Birds of Peru). Woodpeckers drum, which I think could be construed as a “song”, but they also vocalize, giving something I’d also consider a “song”. In my efforts to try to clarify what I was declaring “songs” versus “calls” in Birds of Peru, I said the following in the introduction: “We [in this book] use the terms “song” and “call” frequently, although there are many species for which it is difficult to label a particular vocalization as one or the other. Generally, we classify vocalizations that are produced in territorial defense, mate attraction, and pair bond maintenance as “songs.” In some cases, the “song” of a bird is not vocal at all, but rather mechanical: guans rattle wing quills during short predawn flights, for example, and woodpeckers drum on resonant substrates. Some species also combine mechanical sounds with vocal sounds in elaborate displays (especially among the cotingas and manakins, as well as other groups). “Calls”, on the other hand, are a class of vocalizations containing sounds with many different functions, such as to maintain contact among members of a pair, family, or flock; to warn others of danger; to mob predators; and sometimes in territorial defense. Most species have a wide repertoire of calls, many of which are seldom used.”

    This was probably only just scratching the surface, but I think the above statements encapsulate what most of us think of when we use “song” and “call”.

  14. Hey, Dan. Here’s the thing: I’m fine with the idea that hooting owls and buzzing goldeneyes are engaged in a form of song; and I’m good with your various other examples (guans, woodpeckers, furnariids, etc.). But function (e.g., territorial defense, mate attraction) is not the only criterion for defining song. If it were, then the concept of “song” could be extended to entirely non-acoustic behaviors for defending territories or attracting mates (e.g., bright colors, even the “perfume” exuded by Crested Auklets). Conversely, what about instances of “song” that are not necessarily attributable to territory defense or mate attraction (e.g., blackbirds “singing” in large roosts in early winter)?

    The David Spector paper in J. Theor. Biol. is worth reading. I have to say, I had a hard time finding it. Here’s a bit from the abstract:

    “Unrecognized different definitions for the same term can produce confusion among workers in a field. A review of over 80 definitions of bird song shows little agreement as to what defines bird song or differentiates it from calls. Defining criteria have included structural (e.g., duration), delivery (e.g., rhythmic production), physiological (e.g., hormonal control), developmental (e.g., learning), functional (e.g., territoriality), affective (e.g., musicality), and taxonomic (e.g., restricted to passerines) attributes of song.

    You’re keying in on function (e.g., territorial defense, mate attraction, pair bond), but there are many, many other definitions of song. Eighty! Well, seventy-nine, I suppose, now that we’ve knocked off function… 🙂

    On the one hand, Dan, I suspect I’m aligned with your thinking about what constitutes “song” and what differentiates it from “call.” On the other hand, I think it’s good to keep David Spector’s caveat in mind, namely, that there exists “little agreement as to what defines bird song or differentiates it from calls.” If we asked a biochemist or cognitive scientist or musician, we might get–we probably *would* get–quite different takes on the matter.

  15. Ted, Bill, Nathan – I got hearing aids 2.5 years ago. They have 4 modes, 2 specifically tailored to birding. Thanks to a birding friend, I discovered those birds which I couldn’t hear (e.g. Bushtits, Ruby-crowned Kinglet, Costa’s Hummingbird) and the audiologist fine-tuned the birding programs. Now when I hear Bushtits, I stop and enjoy the song, getting my $$$$ worth!

  16. Just to clarify that last sentence: I took recordings and spectrogram printouts of Bushtits etc to the audiologist, who then programmed the hearing aids to enhance those frequencies.

Comments are closed.

Comments are closed.