In De medicina 5.25, Celsus calls pills that relieve pain through sleep anodyna, but says they should be used only under strong necessity, because their ingredients are powerful and bad for the stomach.
Reading Library · Anesthesia Epochs · 中文版
Watch the education video (for clinicians)
1st Century CE: Celsus on Sleep and Pain — Limits Before the Dose · Anesthesia Epochs Education Series

Before listing preparations intended to bring sleep and relieve pain, Celsus places a restriction on their use.
Book V, chapter 25 of De medicina begins by noting that pills are numerous and serve different purposes. Those that relieve pain through sleep are called anodyna. The next sentence says that their use is improper unless a strong necessity presses, because their ingredients are powerful and unfriendly to the stomach.[1, 5.25.1, pp. 58–59]
Both statements belong in the history of anesthesia. The first links sleep with pain relief. The second shows the text considering when the desired effect might justify using the preparation.
One name, several preparations
The chapter supplies more than a single preparation of “opium and wine.”
The first formula includes papaveris lacrima, rendered by Spencer as “poppy-tears.” A second lists mandragora among its ingredients and is described as more soporific but worse for the stomach. A third again contains poppy-tears, connects the preparation with several painful conditions and includes raisin wine in its account of preparation.[1, 5.25.1–3, pp. 58–61]
Collapsing these passages into one formula loses distinctions that matter: different ingredients, different stated uses and an explicit comparison of undesirable effects. This article preserves those distinctions without converting quantities into doses or identifying every ancient plant name with a securely established modern species.

A warning has a particular object
The opening caution is not the chapter’s only qualification. At 5.25.5, after describing another preparation for a painful condition that prevents sleep, the text specifically warns of danger if it is given during fever.[1, pp. 62–63]
That warning belongs to that particular passage. It is neither a modern prescribing rule nor evidence that every preparation in the chapter carried an identical risk. It does show that the account includes selection, comparison and conditions of use. Extracting only the promises of sleep or pain relief would remove those conditions.
The effects reported in the chapter also remain textual claims. These passages do not supply a clinical case with a named patient, documented procedure, measured anesthetic depth and an outcome that could be independently checked. They provide no basis for calculating a success rate.[1, 5.25]

Modern-looking quantities, ancient materials
The inspected English pages contain grams and volume measures. Those numbers can make the prescriptions look reproducible. Yet Spencer’s introduction to the volume gives an explicit qualification: although quantities have been converted into modern measures, the standard strength of the original preparations cannot be established. He therefore says they cannot be dispensed or directly compared with prescriptions of his own time.[1, introduction, p. vii]
The translator’s warning helps explain why the present article does not turn the page into a usable recipe. Identifying the effect an ancient writer sought and establishing a dose that reliably achieves it require different evidence.
The printed witness has a date of its own. Its title page reads 1961, while the copyright page records first printing in 1938. The copy inspected is the 1961 reprint of Spencer’s volume II. Its publication date must remain separate from the date of Celsus’ writing.[1, title and copyright pages; 2]

A first-century textual setting
In his introduction to volume I, W. H. S. Jones places Celsus’ probable activity in the reign of Tiberius, while stressing how little is known of his life and the dispute over whether he practised medicine. This article therefore uses a first-century CE setting. The year 50 is only the midpoint used for browsing that century, not a verified date of drug administration or surgery.[3]
A comparison with Dioscorides also brings the different questions into focus. Celsus organizes this passage around sleep, pain and restrictions on use; Dioscorides’ mandragora entry explicitly mentions cutting or cautery at 4.75.3 and 4.75.5.[1, 5.25.1; 4, pp. 235–236] The articles are linked through that concrete textual comparison. No direct transmission is established, and no conclusion about all of Celsus’ surgical chapters is intended.
Image note: all four images are AI-generated editorial illustrations. The scrolls, vessels and books are generic props, not antiquities, authenticated covers or page facsimiles. Labels are modern reading aids. No image documents a patient, an administration procedure or an established ancient formula.
Sources and reading limits
[1] Celsus. De medicina. Spencer WG, trans. Vol. II. 1961 reprint; first printed 1938. London: William Heinemann; Cambridge, MA: Harvard University Press. University of Toronto scan. The complete English chapter 5.25 was read on printed pp. 59, 61, 63, 65 and the opening sentence of 67, with Latin 5.25.1–16 and the beginning of 17 compared. Introduction pp. vii–viii were read. The entire 376-page scan was not read; no ancient manuscript was independently collated.
[2] Berlin-Brandenburg Academy of Sciences and Humanities. Celsus edition catalogue. The edition table was opened and read; it lists Spencer’s Books 5–6 under 1938. Used for edition identity, not as additional clinical evidence.
[3] Jones WHS. Introduction to Celsus, De medicina, Vol. I. Public limited preview, Logos/Biblia. The visible opening, probable Tiberian setting and discussion of practitioner status were read. The rest is truncated; the complete introduction and its original printed pages are not covered.
[4] Dioscorides. De materia medica. Wellmann M, ed. Vol. II. 1906;4.75.3, 4.75.5. BIUSante scan. Read: the digital main text and printed pp. 235–236, solely for the textual comparison.
Continue exploring
Related reading connects shared questions and contexts; it does not by itself establish historical influence.
Anesthesia Epochs · The Source Room · About & Editorial Method
Updated October 30, 2026. Sources and reading limits are stated above. Illustrations are labelled as original source images or AI-made. Historical accounts are educational; they are not current treatment instructions.