Wednesday olives and wishes.
Karen captured this fantastic image this morning as the sun reflected the boat onto the glassy surface of the water. Great job, Karen! - HP
This swimming yacht woke at Dișlice Island this morning (green marker on map below). This place has these vertical towers of what looks to me like sedimentary rock, surrounded by lots of blue water and little tunnels to swim in and through. We ended the swim (more of a snorkel) on a shallow rock in the bay for some photos.
Like many athletes, swimmers have to be careful at breakfast: you can’t eat too little or too much before a morning swim, and sometimes you return famished. Our Gulet crew puts out a nice spread and leaves much of it out for post-swim second breakfasts.
After our crossing to our new location, Captain Seyfi kindly cut up a cloth napkin to give us each a strip of fabric for the wishing tree. So thoughtful. - HP
Back to that big amphora fragment we saw yesterday: rather than paw through hundreds of pages on dozens of web sites, I uploaded a photograph of it to ChatGPT and asked for identification. I was fairly shocked at the speed and detail of the reply: it compared several elements of the fragment (the rim, the handles) to known types of amphorae and made an identification: Dressel 2–4 (ca. 1st c. BCE – 2nd c. CE), a roman amphora used for transporting large quantities of wine. I’ve pasted the ChatGPT response at the end of this post.
We motored to Kameriye Island, close to mainland Turkiye (red marker on map below). About ½ mile from where we were anchored, there’s a Christian monastery dated from 225CE. Of course we had to swim to it. Along the way, pottery sherds littered the shoreline, cemented into rubble on the sea floor. Our deckhands brought the dinghy around so we could change into shoes, and we hiked up the hill to check things out. The church is still intact, and apparently a pretty big tourist attraction. Some of the ancient (2000 year old?!) olive trees are now “wishing trees”, their branches covered in thousands of ribbons from hopeful visitors.
On making landfall, we were greeted by a herd of expectant goats! - HP
Most of the buildings are long in ruins, but the tiny church remains, and it’s a delight to spend a few minutes in this ancient place. Andrew spied a large stone used for pressing olives just behind the church. There’s a small courtyard with a lovely mosaic, the center of which had been removed, I assume sometime after the 1300s, when the Ottomans arrived. Like many of the far-flung ruins we’ve seen in Turkey, there is very little here in the way of signage, restoration, or stabilization.
On our swim back, we saw more amphorae sherds, and in a particularly large pile of them, an octopus. We were met back at the Gulet with massive brownies for tea time. This rolled into cocktail hour (aperol spritzes), and a mixed Turkish Grill for dinner.
Stargazing turned to snoozing as this boat full of 14 souls rocked to sleep.
Hopper
P.S. Here’s our photo of the amphora and accompanying conversation with ChatGPT
Chat GPT: let’s do a typology comparison. Since we only have the rim, neck, and handles, we’ll focus on those features.
Key Features in Your Photo:
Wide, everted rim (flaring outward)
Thick, solid neck
Large, rounded handles starting high at the rim and sloping down
Proportions suggest it was a large transport amphora, not a fineware piece
Comparison to Known Amphora Typologies:
Dressel 1 (ca. 140–10 BCE)
Thick rims, tall necks, heavy construction
Often used for Italian wine exports
Rims are more upright than flared — so less likely a Dressel 1
Dressel 2–4 (ca. 1st c. BCE – 2nd c. CE)
Developed from Dressel 1, but with broader, flaring rims
Handles connect high on the rim, like yours
Widely used in the Roman Empire for wine and oil
Strong candidate — your amphora’s rim/handle combo fits this group well
Dressel 20 (ca. 1st – 3rd c. CE, Baetica/Spain)
Big, bulbous olive oil amphorae
Rim is thickened but usually more rounded and less flared
Yours looks too open and flaring to be Dressel 20
Late Roman Amphorae (3rd – 7th c. CE, e.g., LRA 1, LRA 2)
Some have flared rims, but usually thinner and less massive than yours
Handles also tend to be narrower
So, less likely a Late Roman type
Most Likely Match:
Your amphora fragment most closely resembles Dressel 2–4 types, produced in Italy, Gaul, and elsewhere in the early Roman Empire. These were wine amphorae, used during the height of Roman seaborne trade (roughly 1st century BCE through 2nd century CE).
⚓ That makes sense given it’s underwater — possibly from a Roman shipwreck.
ChatGPT: Would you like me to put together a side-by-side visual guide of amphora rim types so you can compare your photo more directly with Dressel forms?
Me: Yes, please.
Here are some photographic (and diagrammatic) examples of amphorae, especially of Dressel 2–4 / Roman wine transport types, that you can compare with your find:
What the Images Show & How They Help
Top-left (image 5): A fairly intact Roman wine amphora (likely Dressel 2–4).
Note the pronounced flaring rim, straight neck, and sturdy handles joining near the rim.
The body tapers downward to a pointed base, typical of transport amphorae.
The marine encrustation confirms long submersion.
Top-right (image 4): Close-up of a rim section of a Dressel 2–4 amphora.
You can clearly see the rim profile: outward flare, thickened lip, more rounded in cross-section.
Bottom-left (image 0): A full view of a Dressel 2–4 in a gallery/auction listing.
Helps see the full proportions: rim, neck, shoulder, body taper.
The rim shape is a match to what your find suggests.
Bottom-right (image 3): A diagrammatic comparative illustration of several amphora types including Dressel 2–4.
Useful for seeing how rim–shoulder–body transitions differ in other types, reinforcing that your specimen is closest to Dressel 2–4.
Comparison & Assessment
The photographic examples show very similar rim geometry to your fragment: a flaring lip, thickened edge, and handles that join close to the rim.
The structural proportions (rim + neck) line up well with standard Dressel 2–4 examples.
The encrustation patterns and underwater corrosion on the examples are analogous to what your fragment exhibits.
Based on this visual comparison, your amphora fragment indeed bears strong affinities with Dressel 2–4 (1st BCE – 2nd CE Roman wine amphorae).
If you like, I can also pull up examples of non-Dressel amphora rims (e.g. later types, regional variants) so you can see what it isn’t. Do you want me to do that?
(Etc!)