Book 2 of seven in the Harvest Moon Week: small books written the week the Harvest Moon rose full over the ridge, each one set on one of the nights that follow it. This one is set on Sunday, September 27.
Before the Chapters: Dusk at the Coop
It is a little after seven on a Sunday evening, two nights past the Harvest Moon, and the ridge has gone the color of cold dishwater. The sun slid behind the western hills at about four minutes to seven. The wind has come around out of the northwest, and it smells the way it only smells in the last week of September: wet leaves, woodsmoke from somewhere down the valley, and something metallic underneath that means frost is thinking about us.
The steward walks out to the coop with a flashlight that is not turned on yet. There is still just enough light to see the path. There is not enough light to see into the coop, and that is the whole point of the walk.
Inside, the hens have gone up. You can hear them before you see them: the low, grumbling conversation of birds settling, one complaint, a shuffle, another complaint. The flashlight clicks on and swings up to the roost bar, and there they are, not in a tidy row like birds in a picture book but in one heap at the far end, packed wing to wing, with a couple of heads poking out of the top and the rest buried. A drift of feathers against the wall.
The steward counts. It takes a moment, because you have to count tails and feet and the odd shape of a comb where it shows between two backs. One, two, three on top. Four and five pressed underneath. Six by the wall. Seven, the little one, nearly invisible. Eight, facing the wrong way, as she always does.
Eight. The door is latched. The flashlight goes off. The walk back to the house is darker than the walk out.
This is a small thing that happens every night on every homestead that keeps birds. It is also, it turns out, one of the oldest problems in the world. This small book is about it.
THE SNOW-DRIFT PROBLEM: THE EVENING MEASURE
+----------------------------------------------------------+
| Photoperiod .......... about 11h 55m, shrinking daily |
| Sunrise / Sunset ..... about 07:01 / about 18:56 |
| Moon ................. Waning Gibbous, about 9/10 lit |
| Lunar age ............ about 16 days, rising after dark |
| Night low ............ near 38 F, first frost close |
| Season ............... Winter Finding past, dark half |
+----------------------------+-----------------------------+
|
+----------------+----------------+
v v
+---------------------------+ +---------------------------+
| ISA: THE ICE | | KENAZ: THE TORCH |
| The broad bridge hides | | Light shows as far as |
| what moves beneath it. | | it reaches, and no more. |
+---------------------------+ +---------------------------+
|
v
Say what you saw. Say where your seeing stopped.
Chapter 1: I Can Only See Three
Friday was the night of the Harvest Moon. It had been a long working day, the kind that leaves you sitting at the kitchen table at dusk with your boots still on, too tired to get up and too restless to sit. The moon had not yet cleared the trees. The light was going fast.
So the steward did what a modern household can do, which is ask instead of walk. There is a camera inside the coop, mounted high in a corner, and there is a machine helper that can look through it. The question was the simplest one a keeper of hens ever asks: are they in?
The helper looked. It came back and said, in plain words, that it could see three hens on the roost. It did not say the flock was in. It did not say five were missing. It said what it saw, and it said the number, and it made it clear that three was the number of hens it could see, not the number of hens that were there.
Now think about the two easy answers it did not give.
The first easy answer is the comfortable one: “Yes, they’re in.” There were hens on the roost. It was dusk. Hens go to roost at dusk. Any reasonable guess would say the flock was in. And on this night the guess would have been right. That is exactly what makes it dangerous. A comfortable guess that is right nine nights out of ten teaches you to stop walking out to the coop, and on the tenth night there is a fox, or a hen who got shut out behind the woodpile, or a door that did not latch, and the comfortable answer is the one that costs you a bird.
The second easy answer is the alarming one: “Only three are in, five are missing.” This is the answer a careless counter gives, one who treats what the camera shows as the whole world. It would have sent the steward out into the dusk with a flashlight and a pounding heart, hunting for five hens that were sitting quietly two feet from the other three. A false alarm is not free. It costs a walk in the dark, a little adrenaline, and a little trust. Enough of them and you stop believing the alarm at all.
The honest answer sat between those two, and it was the least satisfying of the three. I can only see three. It did not close the question. It handed it back.
And the steward, who has kept these birds long enough to know their habits, knew what it meant. On a cold night, hens do not spread out along the bar the way they do in July. They pile up. They crowd into one corner, against the wall, wedged together for warmth, and they stack like snow blown up against a fence. A camera mounted high looks down on that pile and sees only the top of it. Three heads above the drift. Five bodies beneath.
All eight were there. The steward went out later anyway, and counted by hand, because that is the rule of this house: the camera is a question, not an answer.
What stayed with us afterward was not the hens. It was the shape of that honest answer. It named the edge of its own sight. It did not fill the blind spot with a guess in either direction. It said: here is what I can see, and here, by implication, is where my seeing ends, and beyond that line I will not pretend.
We started calling it the snow-drift problem. Once we had a name for it, we saw it everywhere. Before the week was out we had found it in four other instruments on the ridge, and in a few old stories, and in ourselves.
Chapter 2: What the Wind Does
If you have lived through a hill-country winter you know what a drift is, but it is worth slowing down and looking at one properly, because a drift is a kind of lie the land tells, and it tells it beautifully.
Snow does not fall evenly and stay where it lands. It falls, and then the wind picks it up again and moves it. Wind carries snow along the ground until it hits something that slows it down: a fence, a hedge, a stone wall, the side of a barn, a dip in the field. Where the wind slows, it drops what it carries. The snow piles up in the still air behind the obstacle, and it keeps piling until the hollow is filled and the wind can pass over smoothly again. That is why a snow fence works at all. It does not block snow. It slows the wind so the snow drops where you want it, in a harmless heap out in the field, rather than across the lane.
The result is a surface. A drift is smooth and white and even. It rounds over the tops of fence posts. It fills ditches level with the road. It fills the low spot where the old stone pile used to be, and the gully where the spring runs, and the hollow by the gate where the ground is always soft. You look out across a drifted field the morning after a storm and it looks simpler than it did in summer. It looks like there is less there.
There is not less there. There is exactly as much there as before. The wind has simply hidden the shape of it under a smooth, confident surface. The drift does not tell you about the ditch. It tells you about the wind.
The old hill shepherds knew this in their bones, because the drift was where they lost sheep. Sheep caught out in a blizzard will shelter in the lee of a wall, which is exactly where the snow comes down deepest. By morning the flock can be buried, and the field looks empty and clean. A shepherd who looked out, saw a smooth white field, and reported “no sheep in the lower pasture” would have been telling the truth about what he saw and a terrible lie about what was there.
So the shepherds went out with long poles and dogs and probed the drifts along the walls. They looked for small holes in the surface where the breath of a buried animal had melted a little tunnel up through the snow. They knew that a sheep under snow can live a surprisingly long time, kept warm by the snow itself and by its own fleece, and that the smooth surface was not evidence of absence. It was evidence of wind.
That is the snow-drift problem in its plainest form. The surface you can see was shaped by something other than the thing you are looking for. The camera on the coop sees the top of the pile because it is mounted high and the pile is deep. What it shows is a fact about the camera’s angle and the night’s temperature. It is not a fact about the number of hens.
Every instrument has a wind. Every instrument shows you a surface shaped partly by the thing you care about and partly by its own position, its own limits, its own weather. The skill is not in having no wind. That is impossible. The skill is in knowing which way your wind blows, and where your drifts pile up, and in going out with a pole to the places where the smooth surface is most likely to be hiding something.
A field in January is honest about its surface. It is only we who make the mistake of reading the surface as the ground.
Chapter 3: The Eye in the Well
The old stories knew all of this long before there were cameras. The best of them is about Odin, and it is worth telling carefully, because it gets retold in careless ways.
Odin is the one-eyed god. In the old poems he is the wanderer in the broad hat and the grey cloak, the asker of questions, the one who is never satisfied with what he knows. Under one of the roots of the world tree Yggdrasil there is a well, and the well belongs to Mimir, whose name is bound up with memory and deep knowing. The water of that well holds wisdom. Mimir drinks from it every morning.
Odin wanted a drink. The price was an eye. He paid it. The Voluspa, the old poem in which a seeress tells Odin what she knows of the world’s beginning and end, says it outright: she knows where Odin’s eye is hidden, in Mimir’s famous well.
What gets lost in the careless retellings is how strange that bargain is. Odin traded sight for sight. He gave up half of what his body could see in exchange for a kind of seeing that the body cannot do. He looked into the well with one eye, and his other eye stayed there, looking up from under the water, forever.
We have come to read this as a story about the edge of vision. Odin did not pretend his two eyes saw everything. He knew that the surface of the world, the part the eyes take in, was only a surface. There was more under it, down in the dark water at the root, and the only way to reach it was to admit that his ordinary seeing was not enough and pay the price of knowing more. The price was not small. You do not come back from that well with the eyes you went in with.
Odin has other instruments, too. Two ravens, Huginn and Muninn, whose names are usually rendered as Thought and Memory, fly out over the whole world every day and come back to sit on his shoulders and tell him what they saw. They are, if you like, the first remote cameras. And in the poem Grimnismal, Odin says something that every person who has ever run a network of sensors ought to have carved above the door. He says he fears each day that Huginn will not come back. But he fears more for Muninn.
Think about that. The god who gave an eye for wisdom, who has two ravens reporting from the ends of the earth, still worries every single day that his instruments will not return. He does not assume that silence means all is well. He knows that a raven who does not come home is not a report of an empty world. It is a report of a lost raven.
Then there is Heimdall, the watchman at the rainbow bridge. The prose Edda gives him senses beyond any creature’s: he needs less sleep than a bird, he can see a hundred leagues by night as well as by day, and he can hear the grass growing on the earth and the wool growing on the sheep. If any being in the old lore could be trusted to see everything, it is Heimdall.
And yet he keeps a horn. Gjallarhorn, the great horn whose blast will warn the gods when the end comes. The finest sentinel in all the worlds does not rely on his own perfect senses alone. He has a signal meant for others, one that can be heard across the worlds. Some readers of the Voluspa even think Heimdall, like Odin, left something of his own at that same well beneath the tree, his hearing or his horn; the old words are unclear, and we will not pretend they are clearer than they are.
Three lessons, then, from three old figures. Odin: your ordinary sight is a surface, and seeing further costs something. The ravens: silence from an instrument is not news of an empty world. Heimdall: even perfect senses need a way to raise the alarm that does not depend on any one pair of eyes.
The coop camera is a very small raven. It flies no further than one corner of one henhouse. But it carries the same obligation. When it comes home, it should tell the truth about what it saw, and it should never let us mistake its silence for the world’s.
Chapter 4: The Camera That Looked Like a Camera
The second instance of the snow-drift problem that week was quieter, and in some ways worse.
There is a camera in the backyard. It watches the open ground between the house and the tree line, where deer come through at dawn and where, once in a while, something with a longer tail does too. It has been there a long time. It has a little light on it. It shows up in the list of cameras with the same icon as all the others. By every outward sign it is a working camera.
It has not seen anything in a while.
Somewhere along the way the lens fogged over, or got a film of grime on it, or a leaf or a spider’s work settled across it. We are not sure yet which, and we will not know until someone climbs up there with a rag. What we do know is that every time it is asked for a fresh picture, it tries, and fails, and the failure is quiet. Ask it what the backyard looks like and it either gives you nothing at all or hands back the last image it managed to take, some time ago, as if that were now.
This is the fogged lens problem, and it is the snow-drift problem’s more dangerous cousin. The coop camera, at least, was seeing something real. Three hens were really there. The error was only in forgetting that the view was partial. The backyard camera is seeing nothing, and it looks exactly like a camera seeing an empty yard.
Those two things feel identical when you look at them from the kitchen. An empty yard and a blind camera both produce the same result: no alerts, no motion, no deer, no fox. The quiet in both cases is total. But one quiet means the yard is empty and the other means we do not know anything about the yard at all.
Think of a night watchman in an old town. In many places, for centuries, the watch walked the streets after dark, calling out the hour and the weather and that all was well. The people in their beds heard the voice and turned over and slept. Now suppose the watchman fell asleep in a doorway at midnight. The town would hear nothing. And nothing, to a sleeping town, sounds exactly like “all is well.” The only way anyone would know the difference is if they were listening for the call itself, not just for the absence of alarm. A town that noticed at one in the morning that no one had called the hour would know that something was wrong with the watch, even if nothing was wrong with the town.
That is the heart of it. An instrument that can go blind without saying so needs a second thing to watch it: a heartbeat, a call of the hour, a sign that it is still actually looking. For a camera, the simplest heartbeat is the timestamp on its last good picture. If the newest image is from Tuesday and it is Friday, the camera has not been watching the yard since Tuesday, no matter how green its little light is.
The fix we have written down is modest. Every camera on the ridge should be asked, at least once a day, not “did you see anything?” but “when did you last see anything at all?” A camera that cannot answer the second question with a recent time is not a quiet camera. It is a blind one, and it should be listed as blind, in plain words, at the top of the morning report, until someone climbs the ladder with a rag.
There is a small humility in that. It means admitting that part of the yard is dark to us right now. It would be more comfortable to leave the camera in the list with its green light and let it go on looking like a watchman. But a watchman asleep in a doorway is worse than no watchman, because the town sleeps easier for him.
Chapter 5: The Empty Sky
The third instance came from above.
There is a small radio on the ridge whose only job is to listen for airplanes. Aircraft passing overhead broadcast little messages about who they are, where they are, how high, how fast. A modest receiver with a decent antenna, sitting on a hill, can hear a surprising number of them: the long-haul flights crossing high over the hill country, the regional hops, the occasional small plane putting along below the clouds. The radio keeps a count. On an ordinary day the count is never zero. There is always something up there.
For several days, the count was zero.
Zero aircraft. Morning, noon, evening. The report came in every day like a clean ledger: nothing heard, nothing seen. If you had read only the number and knew nothing else, you might have wondered for a moment whether something had happened to the sky.
Of course nothing had happened to the sky. The planes were up there as always, with their lights winking and their passengers asleep against the windows. What had happened was on the ground: the antenna was disconnected or broken. The radio was listening with its ears stopped up, and it was dutifully reporting the silence it heard.
This one is instructive because the number itself was so clean. Zero is a respectable number. It does not look like an error. An error, in most systems, looks like a crash or a garbled line or a warning in red. Zero looks like a fact. It fits in the same column as five hundred. It adds up. It averages. It goes into a chart without complaint.
But this zero was not the same kind of thing as a zero on a quiet night. It was not a measurement of the sky. It was a measurement of the antenna, dressed up as a measurement of the sky.
There is a phrase we keep coming back to: nothing there is not the same as nothing seen. The difference between them is the difference between the empty field and the drifted one, between the sleeping town and the sleeping watchman. When an instrument says nothing, the first question is not “what does nothing mean?” It is “could this instrument have seen something, if something were there?”
For the airplane radio, the answer is easy to check once you think to ask it. On a hill with a working antenna, the count should never be zero for long. The sky over any inhabited country is never empty for days at a time. A reasonable expectation, a baseline, is built into the physics of the thing. When the count falls far below that baseline and stays there, the most likely explanation is not a sudden change in the world. It is a change in the listener.
Old farmers had their own version of this. The rain gauge on a post in the garden is a simple instrument: a tube, some markings, a funnel. A farmer who read no rain in his gauge for three weeks in a wet spring, while his boots came in muddy every evening, did not conclude that the rain had skipped his garden. He went and looked at the gauge, and found a wasp nest in the funnel, or a crack in the tube, or a leaf over the mouth of it. The mud on his boots was a second witness, and the second witness disagreed.
That is the other half of the lesson. A single instrument can go deaf or blind and never know it. Two instruments that ought to agree can catch each other. The airplane radio could have been checked against the plain fact that a plane goes over the ridge most evenings, winking its way east, and anyone standing on the porch can see it. The zero count, set against one clear evening with a plane in the sky, would have fallen apart at once.
We have made a note to give the airplane radio its own call of the hour. If it hears nothing for too long, it should say so, not as “zero aircraft” but as “I have heard nothing, which is unusual, and my ears may be the problem.” That sentence is longer than a number. It is also true.
Chapter 6: The Grid Corners
The fourth instance of the snow-drift problem turned out not to be a problem at all, and that is exactly why it belongs in this book.
On the ridge and around it there are a few little mesh radios: small, cheap, long-range devices that pass short messages from one to the next across the hills without any tower or wire in between. Among other things, each one can report where it is. For a while we watched those positions come in, and they looked wrong. A radio sitting still on a windowsill would report itself in one place, then in another place a good distance away, then back again. It looked like a bad fix, the kind a cheap receiver gives on a cloudy day under heavy tree cover. It looked like something jumping between two points.
We started to troubleshoot. We thought about antennas and sky view and interference. We thought about whether one of the radios was faulty.
Then we read the settings, which is what we should have done first. The radios had been told, deliberately, to round off their positions for privacy. Instead of reporting exactly where they sat, they snap their reported location to a coarse grid, with cells somewhere around seven hundred meters one way and five hundred the other. Any radio inside a cell reports itself at the same point. A radio sitting near the line between two cells can flicker from one side to the other as its raw reading wanders by a few meters, and each time it flickers, its reported position leaps to a different grid point. The jump looked like hundreds of meters of error. It was really a few meters of ordinary drift, magnified by the grid.
Nothing was broken. The instruments were doing exactly what they had been told. They were being honest about a resolution we had forgotten we had chosen.
This is the gentlest form of the snow-drift problem, and the most common. The instrument’s view is limited, and the limit is known and written down, but the person reading the result has lost track of it. The coop camera’s limit is its angle. The backyard camera’s limit is its fogged glass. The mesh radios’ limit is a grid we set on purpose. In each case the reading is only as good as your memory of the instrument’s edges.
There is a long human history of exactly this kind of honesty, and it lives in weights and measures. For centuries, the question of how much grain is in a bushel or how long a yard of cloth runs was not left to each merchant’s judgment. Towns and kingdoms kept official standard measures, and inspectors checked the measures used in markets against them, and a measure that had been checked might be stamped or sealed to show it. The point of all that effort was not that every measure was perfect. The point was that everyone knew what a measure meant and how far to trust it. A bushel was a declared thing.
Surveyors have always worked the same way. A good survey does not only record where a boundary lies. It records how the measurement was made and, in a careful hand, how close the surveyor believes it to be. A line drawn on a map with no word of its precision invites a lawsuit.
The mesh radios, in their small way, are good surveyors. They carry the resolution in their settings. The fault, if there was one, was in the reader who took a coarse grid point for a precise fix and then went hunting for an error that was never there.
We have a new habit now. When a reading looks strange, before we ask “what is wrong with this instrument?”, we ask “what did we tell this instrument to do?” Half the time the answer is already written down. The instrument is not lying. It is reporting its world at the grain we asked for, and the grain was coarser than we remembered.
A snow drift is honest about the wind. A grid corner is honest about the grid. Both will fool you only if you forget what shaped them.
Chapter 7: Fog Ghosts and the Muted Roost
There are two more cameras on the ridge, and between them they finish the picture, because they show the snow-drift problem running the other way.
The first is a night camera on the porch. After dark it lights the steps and the dooryard with its own infrared light, invisible to us, bright as day to its sensor. On a clear night it is a fine watchman. It sees the raccoon at the compost, the cat on the rail, the deer at the edge of the lawn, all in that flat grey night-vision world where eyes glow white.
On a foggy night it goes a little mad.
Fog is a cloud that has come down to sit on the ridge, and on this hill it does that often in autumn, rolling up out of the valley in the evening and settling in so thick you cannot see the barn from the porch. To the porch camera, fog is not darkness. It is a million tiny droplets hanging in the air right in front of the lens, and every one of them catches the camera’s own infrared light and throws it back. The camera sees drifting, glowing motion everywhere. It decides something is moving. It raises the alarm.
Ghosts. The camera is lighting the fog and then startling at its own reflection. Nothing is on the porch. The camera sees something anyway, because the thing it sees is its own light, bounced back at it.
The second is the roost camera, the one that started this whole book. At night we keep it muted. Not blind: muted. It still watches, and it still records, but it does not send alerts after dark. The reason is simple. Hens on a roost are not still. They shift, they shuffle, they re-settle when one of them decides she wants the warm spot in the middle. Every shuffle looks like motion. Left alone, the roost camera would ping all night long, every alert a hen adjusting her feathers.
Here the snow-drift problem comes around the other side. The drift hides things and makes a field look emptier than it is. The fog does the opposite: it fills an empty porch with phantoms. One instrument says less than is there. The other says more. Both are shaped by their own wind.
And the muted roost teaches a third thing, which is that sometimes you choose your blind spot on purpose. We decided not to listen to the roost camera at night. That was a real choice with a real cost: if something did get into the coop in the small hours, the roost camera would see it, and we would not hear about it from that camera until morning. We accepted that cost because the alternative was a camera that cried wolf so often we would stop listening to it at all, and then it would be worse than muted. It would be ignored.
The difference between a chosen blind spot and a forgotten one is everything. A chosen blind spot is written down. You know it is there. You cover it some other way: a well-latched door, a solid run, a hand count at dusk. A forgotten blind spot is a fogged lens with a green light. It is the airplane radio with its ears stopped up. It is the shepherd reading a smooth field as an empty one.
The old lighthouse keepers, by long tradition, kept a log of their watch. They wrote down the weather and the state of the light and what passed on the water. What made a keeper’s log trustworthy was not that it recorded everything. No keeper saw everything on the sea at night. It was that the keeper wrote down the conditions: fog, heavy rain, visibility poor. A ship not recorded on a clear night meant something. A ship not recorded in thick fog meant much less. The honest log carried its own weather with it.
So we have started doing the same. The porch camera’s alerts on a foggy night are marked as fog-night alerts, and taken with a grain of salt. The roost camera’s night silence is marked as chosen silence, not as all-clear. The record should say not only what was seen, but how well we were able to see.
Chapter 8: The Shepherd at the Gate
This chapter is a story. It is not history, though it is built from old things that were true. Tell it by the stove, if you have a stove.
There was once a shepherd on a high hill farm who kept a flock of some two hundred ewes. He was not a clever man in the ways that are admired at market, but he was careful, and he knew his sheep, and every evening in the dark half of the year he counted them in through the gate of the fold.
He counted the old way. He did not stand in the middle of the field and try to tally the flock where it grazed, because a flock in a field is a snow drift of its own: sheep behind sheep, sheep lying down in the hollows, sheep standing so that two look like one. You cannot count sheep in a field and trust the number. So he opened the gate of the fold only partway, just wide enough for one or two at a time, and let them pass through the narrow place, and counted them as they went. Every score, every twenty, he cut a notch in his tally stick, and when the last ewe went through he counted the notches and the leftovers and knew his number.
The gate was the whole trick. The field was wide and confusing. The gate was narrow and honest. A sheep in the field might be hidden. A sheep through the gate had been seen.
One evening in the first cold week of the year, with fog coming up out of the valley, the count came out two short.
His neighbor, who farmed the next hill over, was leaning on the wall and watching. “Your stick’s off,” the neighbor said. “Two ewes slipped past you in the fog. They’re in there. I saw the whole flock go down the lane this morning and nothing’s taken any of them.”
The shepherd looked at his stick and at the fog and at the fold full of dim, grey, shifting backs. It would have been easy to agree. It was cold. He was hungry. The count had been right every night for a month. The fog was thick, and it was likely enough that two had pushed through together and he had missed one.
“Could be,” he said. “And could be not. I counted what came through the gate. The gate says two short. I did not see two slip by. So I will not say they did.”
The neighbor laughed at him and went home to his supper.
The shepherd lit his lantern and walked back up the hill field in the fog. He did not find anything in the open, where the lantern showed only grey. He went to the walls, because walls are where things end up: the lee side, the corners, the places where the wind drops what it carries. And in the far corner of the top field, where the wall met a thicket of thorn, he found one ewe cast on her back in a hollow, legs in the air, unable to right herself, as happens sometimes to a heavy sheep in a dip of ground. A cast ewe left all night on a cold hill may not see morning. He rolled her up and held her until she found her feet, and walked her down.
The other he did not find that night. He went back in the grey dawn, with the dog, and found her tangled in the thorn not twenty paces from where he had found the first, scratched and cross and very much alive.
The neighbor’s count that season, as it happened, was a comfortable one. He counted his flock in the field, and it always came out about right, and he never walked the walls on a foggy night, because his number always said there was no need. It was spring before he learned what his comfortable number had been hiding, and by then it was a sad thing to learn.
The old tellers would stop here and let the fire speak for a while. Then they might add the plain point, for the children. The shepherd was not wiser than his neighbor because he saw better. In the fog he saw worse than anyone. He was wiser because he knew where his seeing stopped. He trusted the gate, which he had built to be honest. He refused to trust the fog, which had never promised him anything. And when the honest gate and the comfortable guess disagreed, he believed the gate and went out with his lantern.
Every system worth building has a gate somewhere in it: a narrow place where things pass one at a time and can truly be counted. The work of a steward is to build good gates, to count through them faithfully, and, when the count comes out short, not to explain the missing ones away.
Chapter 9: Isa and Kenaz, the Ice and the Torch
We have tagged this little book with two runes, and it is time to say why.
Isa is the ice rune, a single straight upright stroke. Of all the runes it is the simplest to carve and one of the hardest to read. The old Norwegian rune poem gives it a short verse that has always struck us as strange and exactly right: ice we call the broad bridge; the blind man must be led. The old English rune poem sees it differently, as something fair to look at, cold and slippery, glittering like glass and like jewels, a floor made by frost.
Put those two together and you have the snow-drift problem in the form of a pond.
Every hill-country household knows the pond in December. On a clear cold morning it is a sheet of glass, flat and bright and inviting, a floor made by frost. It looks like a road. It looks like a bridge. It looks, above all, like one thing, one surface, the same all the way across.
It is not the same all the way across. Ice over moving water is thinner than ice over still water. Ice near the inflow, near the reeds, over a spring, near a dock post that holds the day’s warmth, is thinner than ice out in the open. The surface tells you almost nothing about any of this. The broad bridge looks equally broad everywhere. And so, in the verse, the blind man must be led: someone who cannot see beneath the surface must not trust the surface alone. In one sense all of us are blind men on the ice. None of us can see through it by looking.
The old answer to that blindness was not better eyes. It was the auger and the axe. You do not judge ice by looking at it. You cut a hole near the shore and look at the edge of the cut, and then you walk a little way and cut another, and you trust only what the holes have told you. Each test hole is a small gate. It turns the smooth, confident, unreadable surface into one honest number at one honest place. Everywhere between the holes, you remember that you do not know.
Kenaz is the torch. In the old English poem it is a flame known to every living thing, bright and shining, burning where noble folk rest indoors. It is the light of the hall and the hearth, the light you carry into a dark place so you can see what is there.
But a torch has an edge. Anyone who has carried a lantern into a barn at night knows the shape of it: a warm circle, bright near the hand, fading out into shadows that seem darker for the light beside them. The torch does not light the whole barn. It lights a circle, and it makes the dark beyond the circle look even deeper than before. A person with a torch who says “there is nothing in this barn” has only earned the right to say “there is nothing in my circle.”
Isa and Kenaz, side by side, are the two halves of honest seeing. Isa warns that a surface can look whole and hide what is beneath. Kenaz reminds us that even our light has an edge, and the edge must be named. The wise steward carries the torch onto the ice and is honest about both: here is what my light shows, here is where it stops, and here is the ice I have not tested.
That is the design lesson this whole week kept handing us, and it is worth saying as plainly as an engineer would say it.
An instrument on a homestead or in a data center ought to report three things, not one. It should report what it saw. It should report whether it was able to see, and when it last saw anything at all. And it should report, when it can, how far its seeing reaches: its angle, its resolution, its grid, its weather. A reading without the second is a watchman who might be asleep. A reading without the third is a grid corner mistaken for a precise fix. The coop camera, with the helper’s honest words behind it, did all three that night, in its way. I can only see three. That sentence contains a count, a confession that the camera was working, and an admission that its view had an edge.
We would rather have a thousand instruments that say “I can only see three” than one that says “all is well” and means “I did not look.”
Troth: The Edge of Seeing
Here is the week, distilled into rules for this house and anyone who keeps watch over anything.
- Say what you saw, and say where your seeing stopped. A count without its edge is only half a count.
- Nothing seen is not nothing there. Before trusting a silence, ask whether the instrument could have heard if there were something to hear.
- Give every watchman a call of the hour. An instrument that can go blind without saying so needs a heartbeat: a last good time, a baseline, a sign that it is still looking.
- Keep two witnesses. When one instrument says zero and the mud on your boots says otherwise, believe the disagreement and go look.
- Know your grain. Before blaming the instrument, read what you told it to do. A coarse grid honestly reported is not an error.
- Mark your weather. Record the fog along with the finding. A clear-night silence and a fog-night silence are not the same silence.
- Choose your blind spots and write them down. A muted camera is fine if you know it is muted and cover the gap another way. A forgotten blind spot is how flocks are lost.
- Build honest gates. Find the narrow places where things pass one at a time, and count there, not in the open field.
- Do not fill the dark with comfort. When the gate and the guess disagree, trust the gate and take the lantern out.
What Still Sits on the Scale
The backyard camera is still blind. Someone needs to climb up with a rag, and the ladder is heavy, and it is getting cold. Until then it sits at the top of the morning report as blind, which is where it belongs.
The airplane radio’s antenna is still on the list. It will need a dry afternoon and a little patience with a connector that does not want to cooperate.
There are a few larger questions, about which of the house’s instruments are quietly telling stories and which are telling the truth, that we have looked at and set aside for the long evenings. Some of those will take real work. None of them will be finished by the time the snow comes, and that is all right. The dark half is long, and careful work done slowly beats careless work done before the frost.
And there is one small question with no answer yet: how many other drifts are out there that we have not noticed, because they look so smooth. We suspect the honest answer is some. We intend to go out with a pole.
The Blessing
Hail Odin, who gave an eye at the well to see what the eyes cannot, and still feared for his ravens every day.
Hail Mimir, keeper of the deep water, who knows what lies under the surface of things.
Hail Heimdall, who hears the grass grow and still keeps the horn close at hand.
Hail to the land-wights of this ridge, who send the fog up from the valley and teach us humility by it.
Hail to the hens, all eight of them, drifted in the corner, warm and accounted for.
May every count come out whole, and when it does not, may we have the courage to walk the walls.
May our instruments tell us plainly when they cannot see, and may we have the patience to listen.
May the smooth field never fool us, and the fogged lens never lull us, and the grid corner never send us hunting for ghosts.
May the torch be bright enough, and may we always know where its circle ends.
May the ice hold where we have tested it, and may we never cross where we have not.
May the flock roost safe through the first frost, and the door latch true, and the hand count at dusk never feel like a chore.
The edge of what we can see is not the edge of what is there. Say where your sight ends, and carry the lantern past it.
Wes hal, and good night.
ᛁ ᚲ ᛗ