The Lighthouse Keeper’s Log

🕐11 min read

Routine Maintenance, Station Greyrock

I’ve been a civilian contractor for the Coast Guard for eight years, specializing in AIS transponder systems and automated lighthouse maintenance. Most of the lighthouses along the New England coast haven’t had a permanent keeper in thirty or forty years. They run on timers and sensors, and my job is to go out every six to eight weeks, check the optics, test the backup power systems, replace anything that’s worn, log my findings, and come back. It’s methodical work and I like it. I like being on the water and I like working alone and I like the fact that the job is purely technical — the light either works or it doesn’t, and I either fix it or I don’t.

Greyrock Light is on a small island about nine miles offshore in Penobscot Bay, Maine. The lighthouse itself was built in 1892 and automated in 1979. The keeper’s cottage is still standing and still used — the Coast Guard maintains it as a bunk facility for contractors who need to stay overnight during extended maintenance, which I do twice a year, in spring and fall. The spring visit is usually four days: I open the facility after winter, test everything, do any weather damage repairs, restock supplies. The fall visit is the same in reverse. For more listening ideas, check out our best horror audiobooks.

The cottage has two small bedrooms, a kitchen, a bathroom with a composting toilet, and a front room with a woodstove. There’s no Wi-Fi, no television, no cell service to speak of. The Coast Guard keeps a marine radio on the shelf above the woodstove. There’s a bookshelf with paperback novels that various contractors have left over the years, and a flat-topped storage trunk in the corner that I’d always assumed was empty or full of equipment.

On my fall visit two years ago, I opened the trunk.

The Original Log Books

The trunk contained nine bound log books, the kind that look like oversized composition notebooks, in black oilskin covers. They were in reasonable condition — a little musty, pages slightly foxed at the edges, but legible. The handwriting inside was small and precise, in ink that had faded from black to brown.

The first volume was dated 1921. The keeper’s name, written on the inside front cover, was W.A. Osgood.

I spent the first evening reading through parts of it. Osgood’s logs were what you’d expect — daily weather observations, shipping traffic he’d observed, maintenance notes on the light mechanism, notes about supply deliveries. The entries were brief and professional. He recorded the hours the light burned, any irregularities in the mechanism, the state of the fog signal. He noted when ships passed close to the island and what their probable course was. He noted the behavior of the seabirds and the weather patterns in that specific way that people who live alone on an island for years develop — not a scientific catalog, more like long familiarity.

I read three volumes the first night. I found them absorbing in the way that diaries and logs always are — you get the texture of a person’s daily life, the things they noticed, the things that worried them. Osgood mentioned a problem with the clockwork mechanism in October of 1923 that took him two weeks to diagnose and repair. He mentioned a vessel he watched run aground on the outer bar in a November storm in 1924 and his unsuccessful attempt to reach the crew. He mentioned being resupplied late in March of 1925 because of ice in the bay, and the specific meals he ate during the nine extra days he was stranded waiting.

I went to sleep and didn’t think about the logs again until the second morning.

The Ninth Volume

The ninth volume was the one that made me stop.

I’d assumed the logs ran chronologically from 1921 through the late 1920s and that the ninth would be the latest, ending around 1929 or so when a new keeper presumably took over. But when I opened the ninth volume, the dates weren’t what I expected.

The dates were from this year.

I’m going to be specific about this because I want you to understand what I mean. The handwriting was the same as in the first eight volumes. The same small, precise hand, the same brown ink faded the same way. The same oilskin cover, the same foxed pages. The paper had the same texture and color as the other volumes. But the entries were dated with the current year, and the entries described a Coast Guard maintenance contractor arriving at Greyrock Light in late September to conduct fall maintenance.

The entries described me.

What the Entries Said

I’m going to quote from the entries because I made photographs of every page on my phone before I left the island. I want to be precise:

The first entry in the ninth volume read: “September 24. The contractor arrives by skiff from Stonington. He brings a blue canvas duffel and the gray Pelican case with the diagnostic equipment. He is alone. He ties up at the south dock. Takes a moment to look at the lantern room from the dock before going in.”

I had arrived on September 24. I had come alone, from Stonington, in my personal skiff. My diagnostic equipment is in a gray Pelican case. I had a blue canvas duffel. I had stood on the south dock and looked at the lantern room. It is something I do every time I arrive — I look at the light from the water before going to work, the way a painter might stand back from a canvas. It’s a habit. I don’t know that I’ve told anyone about it.

The second entry: “September 25. He begins with the optics. The outer lens has developed a stress fracture at the base — he will find this and spend most of the day documenting it for replacement. He eats the soup he brought from the mainland for dinner. He reads before sleeping. He does not use the second bedroom.”

The outer lens had a stress fracture that I had not yet found when I read this. I found it two hours later, at the base, exactly as described. I do bring soup from the mainland. I had used only the main bedroom. I had read before sleeping.

I went through the whole volume that afternoon, sitting at the kitchen table with my phone taking pictures. There were fourteen entries total — one for each day of my visit, though I was only scheduled for four days and was presently on day two.

The remaining twelve entries described things that had not happened yet.

What I Did With the Information

I want to be honest about this: I read those future entries. Of course I read them. Anyone would have. I read them carefully and I tried to decide if they were accurate predictions or if they were fabrications and the similarities in the first two days were coincidence.

The entry for day three said I would have trouble with the backup generator’s fuel transfer pump — a problem that would take me two hours to diagnose and involve a part I didn’t have with me. It said I would radio the Coast Guard station at 14:15 to request a parts delivery, and that the delivery would come the following morning by the Coast Guard Aux boat and the boat would be late because of fog.

The fuel transfer pump failed on day three, at around 11 AM. It took me close to two hours to diagnose. I called the station on the marine radio at 14:12. The parts came the next morning. There was fog. The boat was forty minutes late.

The entry for day six said I would find, under the south dock, a lobster pot that had drifted there and become tangled in the dock supports. It said the buoy was painted orange and black. It said I would spend an hour cutting it free and stacking it on the dock, intending to ask about it when I was picked up.

On day six I found the lobster pot. Orange and black buoy. I cut it free. I stacked it on the dock.

The Entry I Didn’t Follow

The entry for day eight said I would go up to the lantern room at night, after checking the fog signal, and I would spend time up there watching the beam sweep across the water, which I sometimes do when the night is clear. It said: “He stays up there until after midnight. He is thinking about his father.”

On day eight, I deliberately did not go up to the lantern room at night. I went to bed early. I wanted to see if I could break the pattern.

The entry for day nine said: “He went up at 11:45 PM after all.”

I had gone to bed at ten. I woke up at 11:30, inexplicably, for no reason I could identify — I simply woke up, fully alert, in a way I hadn’t during the whole visit. I lay there for a while. And then I got up and pulled on my jacket and climbed the stairs to the lantern room and stood there watching the beam go around until well past midnight, thinking, among other things, about my father, who had died the previous spring.

I don’t know if I went up there because I always would have, or because the log said I would. I don’t know if there’s a difference.

What I Reported

I photographed all fourteen entries and I wrote a summary of what I’d found in my official maintenance log, noting the location and condition of the historical logs in the trunk. I listed them as “archival materials, probable historical value, recommend transfer to Maine Maritime Museum collection for assessment.”

I did not describe the ninth volume in detail in my official report. I described it as “an undated volume of uncertain provenance, possibly a later addition to the collection, contents unclear.” This was not fully accurate. I made a decision that the accurate description would raise questions I didn’t have answers for and that those questions would not help anyone.

The Coast Guard sent an archivist from the district office three months later. She took all nine volumes. I asked her, when she called to confirm the pickup, what she’d found in the ninth volume. She said it appeared to be a continuation log in a different hand, possibly from a later caretaker, and that the dating was “irregular and possibly symbolic — a stylistic choice some keepers used.” She said it was “interesting” but not “historically significant.”

I asked her if the entries described anything specific.

She paused. “They’re fairly general,” she said. “Weather observations, maintenance notes. Nothing remarkable.”

I don’t know if she’d read them carefully. I don’t know if they described the same things when she read them that they described when I did. I know what I saw on those pages. I have the photographs on my phone and I’ve looked at them more than once to confirm I remembered them correctly. I have.

Greyrock Light is still operational. I do my visits in spring and fall. The trunk in the corner is empty now. I still stand on the south dock when I arrive and look at the lantern room before I go in. I’ve tried to stop doing it but it doesn’t seem to matter whether I try or not.

I’ve heard from other people who work alone in remote places at night that they develop a specific relationship with their location — a feeling that the place knows them, that it registers their presence in some way that’s hard to explain. A long-haul driver I know described a stretch of highway that seemed to know too much about the roads he’d traveled. I understand what he meant better now than I used to.

I’m going back in the spring. The light needs calibration on the new optics. I’ll be there four days. I’ve stopped trying to predict what will happen on each one.

More from the Night Shift

• Radio Silence Highway 50
• Guard Tower Camp Redmond
• The West Elevator