Squirrels in Traffic: How I Cracked the Code
I.
One might readily postulate that if squirrels understood the mortal danger of running out into traffic, they wouldn’t do it. Yet squirrels do, repeatedly, run out into traffic, where they often meet their deaths. With such propositions one can only logically conclude that squirrels do not, in fact, understand the danger of running out into traffic.
If such is the case, however, then why do squirrels appear to understand the danger of running out into traffic — once, that is, they run out into it? That is the underlying issue of the questions I propose to address here.
You have likely witnessed the phenomenon yourself, from your position behind the wheel of a car approaching one of the subjects under discussion. You see the squirrel crouched on a lawn near the curb up ahead, casually motionless, blinking perhaps, seemingly content to soak in the atmosphere, a mere observer of the tranquil suburban vista. Nevertheless, you are aware of the possibility the squirrel may decide to jump out in front of you. You know — and this is key — that if it does decide to do so, it will do so at the last second and no other. You become aware, in a way you weren’t before, of the traffic situation behind you. Your foot on the gas pedal prepares to react, paradoxically tensing up yet reducing downward pressure. The squirrel, however, remains motionless on the lawn as you approach, until you have clearly passed what is in your mind the last possible moment any psychologically functioning member of the animal kingdom might attempt crossing the road in front of you. Only then, only after you’ve mentally closed that door with a relieving sigh, convinced the squirrel won’t try anything stupid — he goes.
This, then, is the first anomaly that must be accounted for, the squirrel’s counter-intuitive timing, his apparent need to eject himself bodily into the street only at the moment of greatest risk. It cannot be considered, however, without acknowledging a second, potentially related, anomaly, which first requires a point of clarification regarding appropriate terminology for the behavior under discussion, that of “running out into traffic.” Squirrels are not known to cross the road. Dogs cross the road. Possums cross the road. Chickens, if free range, cross the road. Squirrels run out into traffic.
In this we find our second anomaly. Instead of making a mad dash straight for the far curb — which madness is necessitated by the first anomaly — the squirrel reaches the point directly in front of you such that your license plate is bearing down on it, the point of very greatest danger, equidistant from safety forward or back — and there he halts. He looks up at you. Then inevitably he launches into what appears to be a crackbrained and futile effort to get himself out of this mess. He darts a few erratic bounds toward the far curb, offering you the anticipation of relief, but then makes a U-turn and a pell-mell dash back toward the original curb, much farther away, which he seems to realize when he stops again — he bobs his head, goes left goes right, perhaps even ricochets off on an arbitrary tangent toward your car, while your foot switches pedals to hover over the brake as you weigh in tenuous, white knuckle equilibrium the safety of the squirrel against the safety of the rear of your car, which could be mangled by the texting teen on a learner’s permit you’re now well aware of in your rear-view mirror. Sometimes the squirrel will bound all the way across your path, which triggers however only a premature sigh of relief as he for some reason finds his newly arrived position least safe of all and begins a last moment frantic dash all the way back across the wheel base of your approaching car to his original curb and in the process disappears below the horizon of your hood. You pray please no thump. Sometimes, in a kind of three-dimensional chess version of this encounter, with cars going both directions, he may clear the path of your car after his song and dance only to stop again right in the path of the oncoming car, which might prompt him either to perform the same song and dance for the other driver and eventually send him onto the far curb, or to implement backwards evasive tactics right back into your lane, where he gives you an encore performance.
From our perspective nothing here makes sense. Throughout the entire encounter the squirrel never seems to do what is in his best interest, which is get the hell out of the way, but instead spends precious milliseconds bobbing and weaving around the very epicenter of danger, right in front of your moving car, an act which one supposes might pass as some sort of taunting, like pro football smack talk, if he didn’t look so terrified doing it. The squirrel appears to be calculating relative risks, like any animal would, but then pursuing whichever risk is greatest, which no animal would. Further, it’s often all for naught, since he can easily end up right back in the yard he came from. Squirrels, who seem circumspect and judicious in all other matters, like small town bankers, patiently gathering their nuts and so on, will, when it comes to crossing the road, assume the risk tolerance behavior of experimental test pilots. It is baffling. By one set of observational measures squirrels seem bent on their own survival, yet by another set made simultaneously seem to do all in their power to put that survival at risk.
The two anomalies cited above, the first being the squirrel’s delayed engagement with your car and the second his hesitant evasive tactics, bring up a further point for discussion, possibly a third anomaly: Why doesn’t the squirrel simply wait for you to pass, as any sane rabbit, donkey, or other animal would? One may counter with deer as a potential subject for comparison, but deer I contend should be approached and examined separately. It has been well-documented that deer most often meet their demise in the hours of dusk and dawn, when they are feeding, while it is light out yet drivers have their headlights on, because deer are blinded by headlights due to the internal structure of their eyeballs. Squirrels, on the other hand, whose eyeballs have no such shortcomings, will get themselves killed any time at all.
It was a singular event offering observation of this third anomaly, witnessed first-hand and as it happened, that initially led to the current study, an event that assuredly warrants closer examination before any potential hypotheses can be posed.
II.
I was riding my bike to work one morning and had just turned on to a quiet, winding street when I spotted him, up ahead on the right, crouched in the grass near the curb, empty-pawed and looking across the street. There was no traffic but me. As noted, I was on a bicycle, smaller and less lethal than his usual vehicular threat by several orders of magnitude, and furthermore riding at the speed of melting butter: Surely, I thought, he has the spatial reckoning necessary to account for my approach and calculate a time and trajectory that will see him safely across the street. If, that is, he intends to cross at all. I couldn’t tell.
At first he did nothing. He seemed to be aware of my existence without having acknowledged it in any manner, the way an attractive person you see every day does. As I neared, however, his outward demeanor underwent a slight but distinct change. Without moving a single tiny muscle, the squirrel, at first idle and seemingly content to sit on the grass enjoying the fine autumn morning, perhaps digesting breakfast, mutated before my eyes into a being of edgy, hair-trigger despair, like an escaped con wading down a stream to lose the bloodhounds. I saw it and know what I saw.
All of a sudden the squirrel darted out onto the street and into my path — where it abruptly stopped as if frozen by some magnetic force. He hesitated — a paw, a headbob — then swung his haunches around to face my charge. I was still a reasonable fifteen yards away; I coasted and put my fingers over my brake levers. The squirrel doubled back toward his starting point and I breathed a mental sigh, believing I could now coast by without incident, believing like a dope that in contrast to every other squirrel who has ever existed this one possessed a sense of the probable — but he stopped after only a few feet, perhaps belatedly realizing after all that I was no match for his speed and agility.
One thing must be made clear before we move on. Seeing as how I was riding a bicycle rather than barreling down on him in a car, for the squirrel all this happened in slow motion. The basketball great Michael Jordan once said that at times he was able to achieve a mental state on the court in which it appeared to him that his opponents were all moving in slow motion. It was like that, only more so, since the squirrel was not facing an opponent of similar but marginally lesser athletic abilities, but an opponent of next to no abilities at all. It was not Michael Jordan facing Charles Barkley or Larry Bird, but rather Michael Jordan facing, say, Snuffleupagus.
I continued approaching, the squirrel in the street but on the right. At one or another millisecond after the very last reasonable millisecond — I was nearly alongside now — the squirrel made a sudden go for the promised land after all. Too late: I heard a cartoonish poing! as the squirrel’s skull bounced off the spokes of my front wheel. For just an instant he lay there apparently lifeless then — while I was still passing — he snapped out of it, sprang in one motion from his supine position onto all fours — turning a half-roll in mid-air, mind you — and then, crouched low on the pavement, seemed to mull his options with a kind of electrified urgency. I saw a little of Steve McQueen from The Great Escape in his low crouch, head swinging left and right scouting danger. Unbelievably, despite his dangerously close call, the squirrel spent a fraction of a moment contemplating another shot at the far curb, then still more unbelievably attempted it, darting directly toward my rear wheel. I braced for another poing! but this time after that single sixteenth note of a mad dash the squirrel made an instantaneous decision and an abrupt U-turn, retreating — to fight another day, no doubt — back to the curb whence he came.
Despite the lifelong familiarity with squirrels that I, like most others, possess, I found myself unable to put this encounter or its potential import out of my mind. Yet the more I thought about his actions, the choices he made at each grave millisecond, the more confounding and idiotic they all seemed. Not even a gazelle running helter skelter from a documentary filmmaker’s helicopter is that stupid.
Q: Why did the squirrel cross the road?
A: To get to the other — no STOP! RETREAT, RETREAT! No, wait a — CHARGE!! CHARGE. No, just keep — if I just turn here and — MY GOD THAT THING IS GONNA KILL ME!!!
Reviewing this particular encounter, there appeared to be two salient factors involved: first, the timing of the attack, and second, the squirrel’s hesitation once engaged. Both were quite perplexing. On more extensive reflection, however, placing this instance in context of the whole set of squirrel encounters, it appeared that these two factors were not unique but in fact symptomatic of every squirrel encounter and illustrative of all squirrelkind. They had never seemed perplexing before, a fact that in and of itself was perplexing.
In subsequent review, I discovered that many of the hypotheses one may posit to account for squirrel behavior in traffic, even traffic of a bicycle nature, tend to overlook one or another of the anomalies cited previously without explicitly doing so. Take, for example, the first and most obvious hypothesis that might be suggested regarding squirrel motive: There are nuts in the yard across the street. Now return to my encounter as described above, and you will see an equally obvious question: Why didn’t he just go get them? He was sitting there like grandpa on the front porch, so motionless he might’ve been asleep, and then I closed in and bam! — Steve McQueen. Indeed, in examining this topic closely, I found myself exploring possibilities in what can only be described as a squirrel-like fashion, leaping for an explanation right into the path of an obvious flaw and scurrying backward then taking a sideways reconsideration of things and leaping off toward a new hypothesis and into the path of another flaw, etc.
For this reason, I feel it necessary to provide here the working hypotheses that I formulated, as well as counter-factuals and points that challenged or undermined them. Those hypotheses that I considered scientifically implausible at the outset (for example, squirrels are engaged in a never-ending contest with one another, and getting across the road is considered a “score,” like a touchdown) will be omitted. While some readers may rather hope that I would first pose a carefully framed research question before entertaining hypotheses, I have resisted doing so, determining that a research question framing the issue presupposes confidence in the assessment of what it is squirrels are actually doing. However, what appears to us to be running out into traffic may not be that at all, in the same way that laughing at the boss’s joke is not really laughing, in comparison with laughing in other more genuinely human circumstances, as your boss would understand if he bothered to make the comparison.
III.
The first hypothesis to be discussed is one that rests on the hesitation that I witnessed in the squirrel which I encountered on my bike. He wanted to go to the other side, decided against it, decided for it, decided against, then went, then decided against it, etc. The hypothesis holds that squirrels naturally select for hesitation, and are thus compelled to express this hesitation in some way, just as peacocks naturally select for pretty tail feathers and have to show them off.
Bear with me. Cars have been running over squirrels for quite some time now, through hundreds of squirrel generations, and are still doing it today. The theory of natural selection, however, suggests that this propensity — to meet one’s demise at the wheels of a moving vehicle — should have been bred out of the squirrel population long ago, since those who meet their demise in this way will not survive to pass on their genes. The fact that it has not thus suggests that hesitation in the face of oncoming traffic is a survival trait, like opposable thumbs.
Two questions immediately arise in pursuing this hypothesis: One, what advantage could there possibly be to naturally select for hesitation? Two, how would a species go about naturally selecting for hesitation? Can one suppose that a good hesitator will live to see another day, and hence reproduce, or is it the bad hesitator who does? How does one even differ between a good hesitator and a bad one?
Pondering this hypothesis takes one in confusing overlapping circles that I would discourage one from attempting to wind one’s way through, but it did lead to a secondary thesis that I thought may prove useful in a different context: Perhaps squirrels are so successful at managing their hesitation that, like functional alcoholics, they pass on to their offspring not only the genetic flaw but also the character trait that enables them to survive with the genetic flaw. Perhaps squirrels are functional alcoholics. The evidence doesn’t contradict it.
Unlike the first, the second hypothesis can at least withstand a modicum of scrutiny without causing migraine symptoms. This hypothesis holds that squirrels, like fish, have inadequate memory capabilities, either short- or long-term. This would have a dual effect of leaving them with no experience of what happened the last time they launched themselves out into oncoming traffic, and likewise preventing them from forming any lasting bond with the lawn they’re already on, such as cats possess. It is precisely due to this lack of memory that fish can bear living in an aquarium day after day, indeed hour after hour. It also explains why fishing lures don’t have to be very lifelike but can be fluorescent orange or green with yellow stars or any other preposterous color and shape no other animal would even glance at twice, let alone try to eat. Thus it is, according to this hypothesis, with squirrels: As far as they know, they’ve never seen a moving car before, and really don’t see a problem with it when viewed in light of the potential riches of nuts that lay hidden in the grass of the yard across the street, a yard which unbeknownst to them they’ve scoured a thousand times.
This hypothesis adequately accounts for their hesitation in traffic, since only after getting themselves into it do they fully realize what they’ve gotten themselves into, but as you may have anticipated it does not begin to address one of the central anomalies described: Why do they always make a run for it at a moment after it seems too late? All data they need is present and obvious from the outset; i.e., they can see how fast the car is going. Yet they wait.
Additionally, this hypothesis is fatally undermined when considering a separate and well-known trait of the species: gathering nuts. It makes no sense that a squirrel can remember with a high degree of accuracy the many different locations across several yards where he buried nuts last fall, a wisely diversified portfolio in anticipation of leans times, yet plumb forget the asphalt terror he encountered this morning. (As flimsy as this hypothesis is, it does neatly explain the squirrel’s occasional penchant, after having gone to the trouble to penetrate car-held territory, for making a mad dash straight back to the yard he came from.)
Before examining the next hypothesis, it is necessary to introduce a supposition on which the hypothesis is based. In my own deliberations of this phenomenon I had reached a dead-end, as the above I believe makes clear, and felt it necessary to return to the evidence of the phenomenon that I had, that we all have. It occurred to me that all evidence of squirrels trying to cross the road came to me through first-hand experience, while I was on a bicycle or in a car — in short, while I was present. I couldn’t be sure that I had witnessed a squirrel tangling with other bicyclists or drivers, but one thing seemed irrefutable: there was always a vehicle. No one has ever sat in their study gazing idly out a window of their home, pondering a lost love or the bitter inevitably of hair loss perhaps, and while sitting there witnessed a squirrel sauntering across an empty street. It does not happen. Squirrels do not cross empty streets, nor saunter.
This observation led to a key and fundamental supposition: the car somehow triggers the act. In a kind of modified Heisenberg principle, wherein the presence of an observer alters the behavior of the thing observed, squirrels alone in a yard will behave in a manner that accords with their animal instincts for self-preservation unless and until a vehicle appears, at which point, for some as yet unidentified reason, they instantaneously mutate into mindless beings hell-bent on immediate fulfillment of some desperate primal mission, like zombies on fast-forward.
Despite the significant advance that this observation offered, the hypotheses that arose out of it still required examination. In the first hypothesis that came to hand, the squirrel sees a vehicle as a natural threat to its environment, like a bad storm — since so many of his kin have lost their lives due to vehicles — and rushes to protect himself and that which is most important to him, namely the nuts that he buried in the yard across the street way last summer.
Though this theory seemed reasonable and workable as far as it went, it did nothing to account for the singular anomalies mentioned earlier: Why does the squirrel wait until the last moment to plunge out in front of you? And what’s with all the teen pop idol gyrations once they’re smack dab in the middle of your lane? Wouldn’t a squirrel simply race for the far curb as soon as possible, and take the quickest, straightest route to get there? And furthermore, if the goal is to protect the nuts across the street why would squirrels be satisfied, as they often seem to be, ending up right back in the yard they came from?
Other questions remained unanswered and it was clear that a tenable hypothesis was still wanting that accounted for or built upon the insight that the car triggers the act.
IV.
In analyzing this issue from several different angles, looking for connections and clues, I found myself contemplating dogs, some of whom are also triggered to act by the sight of a moving car. In such an event, however, their actions are nearly the opposite that of squirrels: they chase the cars, barking the cars away. And while dogs, after such an encounter, trot back home demonstrating what appears to be a kind of gloating, like a chaperone at a Christian junior high dance, squirrels afterward appear skittery and jangled, one step away from a nervous breakdown.
This analogy seemed fruitless, then, and strongly suggested I was grasping at straws, until it suddenly hit me that a dog thinks the car is alive. It wouldn’t gloat otherwise. Dogs aren’t chasing and barking at the driver; indeed, with eyes only one or two feet off the ground they can’t even see the driver. They’re barking at the car.
And it was at this point that all the elements fell into place, and a complete and fulfilling hypothesis — which I suddenly felt could now be called a theory — lay before me in all its turning, crystalline logic, elegant as a DNA chain.
To a squirrel, the car is not only alive, but a natural enemy, and what does an enemy want to do? Take your nuts and women. Due to its diminutive size, however, the squirrel can’t hope to confront the threat of your approach with barking and chasing as a dog does, which it knows would be ridiculous and embarrassing, the sort of thing somebody would be sure to bring up at the reunion, and so it employs a unique and singular advantage that natural selection has bestowed on it: lightning-quick reflexes. When it sees you approach, the squirrel does not run to protect the nuts across the street; no, the squirrel tries to distract you from the nuts across the street. The squirrel running into traffic is trying to fake you out.
This more than adequately explains, first of all, the squirrel’s lunatic dance while your car bears down on him. What better way to protect his nuts from you, the marauding horde, than to go where the nuts aren’t — in the middle of the street — and there act like a maniac to make you forget the riches in the vicinity. Hey buddy, over here! Hey, whatcha watching? Bet you can’t do this! Or THIS! See my tail? Ha — made you look! Hey, pull my finger.
This theory also explains the squirrel’s delay, its diabolical penchant for lunging into the street only at the last moment. This delay, as has often been demonstrated by great running backs over the years who don’t cut left until after their defender has committed right, is actually a highly sophisticated survival instinct. If he ran out into the street too early, you would simply come to a stop and assess the situation, leaving the squirrel with no advantage. The squirrel knows this.
This also explains why the squirrel’s excursions are sometimes fatal — why, that is, the gene accounting for excessive risk has not been bred out of the species over generations. The squirrel’s reflexes are so much greater than yours that your reflexes should not even be compared to a squirrel’s but to those of some other species, mold perhaps. While the squirrel has engaged most assiduously in an attempt to distract you for what is to him long exhaustive moments, you have remained — from the squirrel’s perspective — practically catatonic. And then suddenly you’re not. In other words, the squirrel zigs, and then zags, and then zigs again — and only then do you get around to reacting to the first zig. Most of the time, because you do indeed react, the squirrel concludes that its efforts do pay off. Occasionally, however, this unfortunate mismatch of reaction time leaves its zig2 and your zig1 on intersecting trajectories, which is the cause of the squirrel’s untimely demise.
Areas for future research include exploring whether the natural advantages of a squirrel’s lightning reflexes can be harnessed somehow for mankind’s benefit, perhaps in the field of quantum computing, and of course testing hypotheses for preventing accidental squirrel death. Ad hoc experiments could be carried out by any interested reader with an aim toward shaping future research questions. Such experiments will likely prove most fruitful by accepting the premises arising from this study. In your next encounter with a squirrel, implement the single strategy these premises all seem to point to: Try to kill the squirrel. Speed up, swerve, honk, open your door and stick out your foot, etc. — try anything at all. It is conjectured that you will fail because, like Michael Jordan’s opponent, you are moving in slow motion. Squirrels die only when you try not to kill them.
