Aaron stands at the entrance of the tent of meeting with two goats and casts lots over them. One lot is for the LORD. The other, in Leviticus 16:8, is for Azazel. The first goat becomes a sin offering. The second stays alive, receives the confessed sins of Israel, and is driven out into the desert.
Leviticus never says who Azazel is.
The name appears three times in the chapter, at verses 8, 10, and 26. In the first of those it sits in direct parallel with the personal name of the God of Israel, which is about as conspicuous a position as a word can occupy in the Torah. The text supplies no gloss, no backstory, and no warning label.
Here is what I think the chapter will and will not carry. Leviticus 16 does not settle Azazel’s identity, and no amount of pressure on the Hebrew is going to make it. What the chapter does establish, without ambiguity, is a movement. Both goats are stood before the LORD before either one is used. The living goat receives Israel’s iniquities by confession and by the weight of the high priest’s hands, and then it carries them out past the boundary of inhabited space. This is a chapter about purging contamination from holy ground. The figure in the wilderness is where the contamination goes, not what the ritual is about.
The word scapegoat is already an answer
Most English readers never run into this problem, because most English Bibles solve it before the reader arrives.
That solution has a history. As Liane Feldman lays out, early Greek translators rendered the phrase as “the one sent off,” and the Latin Vulgate called the animal the emissary goat. The English word arrives in 1530, when Tyndale coined “escape-goat,” which the King James Version shortened in the following century into the form everyone now recognizes. Feldman calls the familiar term a modern interpretation of a perplexing phrase, and that is exactly the right description. Each of those renderings is a decision about what the Hebrew means.
Why would translators make that decision? Because the alternative is to print an unexplained proper name in the middle of Israel’s most solemn ritual and leave the reader holding it. The NET Bible does exactly that, keeping Azazel in verses 8, 10, and 26 and letting the difficulty stand. That is the better choice, though the discomfort behind the other one is easy to understand. “Scapegoat” tells you what the goat does. The Hebrew tells you where the goat goes, and the destination has a name.
Four ways to read a name the text never explains
The first reading takes Azazel as a personal being belonging to the wilderness. Anna Angelini’s survey notes that the spelling in the Masoretic Text has no clear meaning in Hebrew and is probably a correction of an earlier form, one attested in some manuscripts of the Samaritan Pentateuch, in the Temple Scroll’s version of this ritual, and in the Qumran text known as the Pesher on the Periods. That earlier form joins a Semitic root meaning fierce or strong to el, either the divine name El or the ordinary word for a god. On that reading the name means something along the lines of “the god is fierce.”
The second reading takes Azazel as a place. Later revisers of the Septuagint, including Aquila in the second century of the common era, specified a rocky and fortified location. Rabbinic tradition preserves the same instinct. The Babylonian Talmud at Yoma 67b glosses the name as land that is hard and rough, and in one line as the roughest of the mountains.
The third reading treats the word as an abstraction for removal. In verse 26 the Septuagint abandons the vocabulary it had used earlier and refers instead to the goat’s release, a rendering that may reflect the Aramaic root azal, meaning to go away.
The fourth reading identifies Azazel with the goat itself. Philo of Alexandria, working from the Greek, understands the term as a description of the banished creature and turns the rite into an allegory about a soul running away from wisdom.
I find the first reading the strongest of the four, and my reason is structural rather than etymological. The third and fourth options have to account for a lot formula that puts this word exactly where the divine name goes. Angelini presses the same observation, saying the ritual appears to present Azazel as a deity precisely because of that parallel, while also arguing that the chapter has demoted whatever the figure once was.
But I cannot get there from Leviticus 16 by itself, and that belongs in the open rather than in a footnote. The proposed etymology is reconstructed from a spelling the Masoretic Text does not actually have. The parallel in verse 8 is suggestive rather than decisive, since a lot can designate a destination as easily as a recipient. What would change my mind is a clear reference elsewhere in the Hebrew Bible treating Azazel as a person, and I do not know of one.
Right about here a careful reader starts to get nervous, and the nervousness is legitimate. If Azazel is a being, and a goat is assigned to him by lot, is Israel making an offering to a demon?
No, and the chapter works hard to prevent that reading. Both goats are first stood before the LORD at the entrance of the tent (Leviticus 16:7). The goat designated for Azazel is then presented alive before the LORD (16:10). Nothing is slaughtered for Azazel, no blood is directed toward him, and no altar is built. Angelini’s own description is that Azazel receives a warding-off rite rather than a sacrifice. One chapter later the Torah bans the very practice the nervous reader fears, insisting that the people “must no longer offer their sacrifices to the goat demons” (Leviticus 17:7, NET). Whatever Azazel is, the legislation surrounding this ritual treats sacrificing to wilderness powers as apostasy.
What the chapter is certain about
What is actually being cleansed on the Day of Atonement? Most people answer that it is the worshipers. Leviticus answers differently.
Marc Zvi Brettler points out that the chapter uses terms related to kippur sixteen times, most of them applied to the tabernacle and its furniture rather than to persons, and he argues that day of purgation is the better translation of the festival’s Hebrew name. Blood functions here as something close to a detergent, applied to the atonement lid, the Holy Place, the tent, and the horns of the altar. What is at stake is God’s willingness to go on residing in the middle of a contaminated camp.
The live goat handles a different problem, and the NET Bible states the procedure plainly at Leviticus 16:21:
Aaron is to lay his two hands on the head of the live goat and confess over it all the iniquities of the Israelites and all their transgressions in regard to all their sins, and thus he is to put them on the head of the goat and send it away into the desert.
Notice what that sentence does and does not do. It specifies the transfer, the mechanism, the confession, and the direction of travel, and the next verse adds that the goat carries the iniquities “into an inaccessible land.” It says nothing whatsoever about the nature of whoever or whatever is out there. The recipient is named once and then dropped, while the cargo and the trajectory get all the attention.
Feldman, summarizing Jacob Milgrom’s argument (reported here from her summary rather than from the commentary itself), explains why a second goat is needed at all. The earlier purification offerings clear away contamination from impurity and unintentional sin. Contamination from deliberate sin cannot be neutralized that way. It can only be relocated. That also explains an odd practical detail in verse 26, where the man who leads the goat out has to wash his clothes and his body before he comes back into camp. He has been handling something.
This is where the chapter should change the way you read a great deal of the rest of the Old Testament. In the priestly system, sin is not merely a legal status attached to a person. It has weight and location. It accumulates somewhere, it has to be moved, and it has a direction of travel, which is out.
The ritual form was not unique to Israel. Noga Ayali-Darshan gathers a series of send-away rites from the wider region: a goat fitted with a silver bracelet and driven to the steppe at Ebla, Hittite rituals in which animals designated as sent-away carry off an evil word or a false oath or a curse, a bound mouse released toward high mountains, and a Neo-Assyrian billy goat led alive into the steppe with a disease loaded onto it. Her conclusion is carefully framed. Rather than Israel borrowing a rite from a neighbor, she describes a shared practice of the Syro-Anatolian region that Israel inherited along with everyone else living in it.
That comparative material establishes something real and something limited. It establishes that a priest and a congregation in the ancient Levant would have recognized the grammar of what was happening, the way you recognize a handshake. It does not establish what an Israelite believed about Azazel, it does not demonstrate borrowing in either direction, and it cannot show whether the author of Leviticus 16 thought of the wilderness as occupied.
What 1 Enoch adds, and what it cannot settle
By the Second Temple period, Jewish writers had stopped leaving the name blank.
In 1 Enoch 8, in R. H. Charles’s translation, Azazel teaches humanity to make swords and knives and shields and breastplates, to work metals, and to produce bracelets, ornaments, eye paint, and dyes, after which, in Charles’s rendering, there arose much godlessness. Two chapters later God orders Raphael to bind Azazel hand and foot, throw him into darkness in the desert of Dudael, and cover him with jagged stones until the day of judgment. Angelini observes that this punishment recalls the Leviticus ritual in several particulars, including the desert setting, the rough terrain, and the transfer of sin onto the figure.
Does that tell us what the author of Leviticus meant? No, and this is the discipline the whole topic requires. 1 Enoch is not Scripture, and it is not a commentary produced by anyone with access to the priestly author’s intentions. It is evidence, and genuinely useful evidence, for how Jews centuries later understood a name their Bible had left unexplained. Only one direction of influence is actually visible in the sources: Leviticus handed Enoch a name and a desert. Whether Enoch handed back anything the priestly author would have recognized is not something these texts can show.
The same restraint belongs on the Christian side. Feldman notes that scapegoat thinking surfaces in the New Testament, including in Hebrews 9. Brettler shows the shape of the argument there, with Hebrews 9:11-14 insisting that Christ entered a greater tent once for all with his own blood rather than the blood of goats and calves, and the argument running on through 9:23-28 toward a single offering that never needs repeating. Every move in that passage is a Day of Atonement move: high priest, sanctuary, blood, cleansing. But Hebrews never names Azazel and never says that Jesus is the live goat. A reader who makes that identification is doing theology on top of the text, which is a different activity from reading it. I would rather keep the typology where Hebrews actually puts it, on the priest and the blood and the sanctuary, than build a tidier correspondence the letter itself declined to draw.
So go back and stand where Aaron stood.
Two goats. Both of them presented before the LORD. One dies, and its blood goes in through the curtain to scour the place where God has agreed to live. The other lives, takes the full spoken weight of Israel’s iniquities on its head, and walks away until nobody can see it anymore, out into land the text calls inaccessible.
You are not told what is waiting out there. That silence has bothered readers for two thousand years, and I would rather Leviticus had explained itself. It did not. What it gave instead is a picture of what forgiveness costs and where sin goes, and the picture holds whether or not the name on the second lot is ever identified. Read the chapter again with the English word scapegoat set aside, and watch how much of it turns on a boundary, and on the fact that something has to cross it.