The Temple Mount / al-Ḥaram al-Sharīf

BETA

north up · 1 px ≈ 1 m · {{ vexNote }} · drag to rotate, shift-drag to pan, scroll to zoom

STANDING LIMITATION The platform surface has never been excavated.

Everything drawn above the retaining walls is reconstruction — from ancient texts, from the walls themselves, from excavations outside the enclosure, and from redeposited debris sieved off-site. Nothing on the esplanade has been dug under controlled conditions. Every shape in the key carries a confidence tag; click any shape on the plan to see what it rests on.

OVERLAY PERIODS

Earlier enclosure Bedrock anchor

VIEW

rotate {{ yawLabel }} tilt {{ tiltLabel }} zoom {{ zoomLabel }} height {{ vexLabel }}

LABELS

PERIOD

Present-day outline

MILESTONES

KEY — WHAT IS DRAWN, AND ON WHAT EVIDENCE

EXCAVATED / STANDING INFERRED FROM REMAINS TEXTUAL ONLY CONJECTURAL CONTESTED

COMPETING RECONSTRUCTIONS — WHERE DID THE SANCTUARY STAND?

Before there was a temple here, there was a hill. The place we now call the Temple Mount began as the northern end of a narrow limestone ridge running down toward the junction of two valleys — the Kidron on the east and the Tyropoeon on the west. In the Bronze Age, roughly the third and second millennia BCE, this high point was bare rock, scrub and thin soil. It was not the center of anything. The town of the period, known from Egyptian texts as Urusalim, sat downhill to the south, on the lower spur archaeologists call the City of David.

The reason the town sat there and not up here is simple: water. The Gihon Spring rises in the Kidron Valley at the foot of the lower ridge, and it was the only reliable perennial water source for the settlement. Springs decide where Bronze Age towns go. The massive stone fortifications excavated around the Gihon — the so-called Spring Tower and Pool Tower, built of boulders weighing several tons and dated by most excavators to the Middle Bronze Age, around 1800 BCE — show how much effort people were willing to spend defending that water. Nothing remotely comparable has ever been excavated on the summit to the north, because the summit has never been excavated at all.

That is the first and most important thing to understand about this entire visualization. The 144,000-square-meter platform is a living religious site under the administration of the Islamic Waqf, and no controlled archaeological excavation has ever taken place on its surface. Everything we know comes from four indirect sources: the retaining walls that hold the platform up, which can be studied from outside and below; excavations in the areas around it, especially the Ophel to the south and the tunnels along the Western Wall; ancient written descriptions; and debris — soil that was removed from the mount and dumped elsewhere, then sieved by the Temple Mount Sifting Project. Redeposited soil can tell you that an object existed on the mount. It cannot tell you where on the mount it lay, or in what layer, or with what else. Archaeology is mostly about context, and on this site the context is gone.

What we can see is the rock itself. Near the middle of the modern platform, inside the Dome of the Rock, a great outcrop of Cenomanian limestone breaks through the paving. It measures roughly 18 by 13 meters and rises about 1.5 meters above the floor. In Arabic it is al-Sakhra, the Rock; in Hebrew tradition it is the Even ha-Shetiyah, the Foundation Stone. It is a real geological feature — the natural summit of the ridge, or very close to it — and it is the single fixed point that every reconstruction on this site has to work around.

Beneath it is a hollow, reached today by a stair on the southeast side, known as the Cave of the Souls. Whether it began as a natural karstic cavity, a rock-cut tomb, a cistern, or something enlarged in stages over centuries is genuinely unsettled. The cave has never been excavated to bedrock floor under archaeological conditions. Cut channels, basins and small shafts are visible on and around the rock surface, and they have been read in wildly different ways: as drainage for sacrificial blood, as the sockets of a much later structure, as quarrying scars, as post-Roman modifications. All of those readings are interpretations of shapes, not of stratified deposits.

The platform is also honeycombed with cisterns — around three dozen have been mapped, most famously by Charles Warren's survey team in the 1860s and 1870s. Some are enormous; the one called the Great Sea holds several million liters. Rock-cut cisterns are notoriously hard to date, because cutting into bedrock destroys the evidence of when the cutting happened, and cisterns are cleaned out, replastered and reused for millennia. A few of the simplest bottle-shaped ones may be very early. Most are almost certainly Herodian or later. On the map here they are marked as approximate and undated, which is the honest position.

Was the summit a sacred place before Israelite Jerusalem existed? Many scholars think it likely, on general grounds: prominent rocky high points near ancient towns often attracted cult activity, and later biblical tradition remembers the site as a threshing floor bought from Araunah the Jebusite, a foreigner in possession before the Israelite king. But "likely on general grounds" is not evidence. No Bronze Age installation, no altar, no temple wall, no votive deposit has been excavated on the summit — because nothing at all has been excavated on the summit.

So the drawing for this period is deliberately thin: contour lines suggesting a ridge falling steeply east into the Kidron and more gently west, the outcrop near the crest, the cave beneath it, a scatter of undated cisterns, and an arrow pointing south to the spring and the settlement that actually mattered. The dashed present-day outline is drawn over it so you can see how much of what feels like a natural hilltop is, in fact, later engineering: the modern esplanade is a made surface, filling and vaulting over slopes that once dropped away sharply on every side but the north.

It is worth being clear about how the bedrock shape is known at all, since no one has surveyed it from above. Most of it comes from below and from the sides. Charles Warren's teams sank shafts to bedrock outside the walls and crawled along the foundations, recording the level at which the masonry sat. The floors and ceilings of the platform's cisterns give scattered spot heights inside. Excavations in the Ophel, to the south, and in the Western Wall tunnels add more points. Stitch those together and you get a rough surface: a crest near the Dome of the Rock at roughly 743 meters above sea level, falling to about 700 at the southeast corner. That 40-meter difference is why the southeastern walls are so tall, and it is the single most important fact about building anything on this hill.

The Sifting Project has produced a small quantity of Bronze Age material, including sherds and a few Egyptian-style objects. That is consistent with people using the summit in the second millennium BCE, but the material was carried there in truckloads of disturbed soil, so it cannot show whether there was a shrine, a farmstead, a threshing floor or simply a path across the ridge. The pre-Israelite summit is therefore the period we know least about, and it is drawn here with the fewest lines.

The First Temple is the most famous building in Jerusalem's history and the least archaeologically known. Not one stone of it has ever been excavated. Not one wall line, not one floor, not one foundation trench. Everything on this drawing comes from written descriptions written down long after the building went up, checked against temples of similar type that have been excavated elsewhere in the region. That is why every element in the key for this period is tagged "textual only."

The main description is in 1 Kings 6–7, with a parallel in 2 Chronicles 3–4 and a related visionary plan in Ezekiel 40–42. Those texts describe a tripartite longroom building: you enter through a porch called the ulam, pass into a long main hall called the hekhal, and at the far end, raised and windowless, is the debir — the innermost room, the Holy of Holies, a cube housing the ark. The overall interior is given as sixty cubits long, twenty wide and thirty high, which works out to something like 30 by 10 meters inside, with side chambers built against the outer walls in three storeys.

Two free-standing bronze pillars stood at the entrance, named Jachin and Boaz. They supported nothing; they were markers. In the courtyard stood a huge bronze basin called the Sea, resting on twelve bronze oxen, plus ten wheeled lavers for rinsing sacrificial parts, and an altar of burnt offering. The texts describe cedar panelling, carved palmettes and cherubim, and a great deal of gold overlay.

The reason scholars take this description seriously as architecture, rather than as pure literary invention, is that excavated temples in Syria and southeastern Anatolia match the plan closely. At ʿAin Dara in northern Syria, a temple of the tenth to eighth centuries BCE has the same porch-hall-inner-room sequence, a raised innermost room, side chambers wrapping the building, carved lions and sphinxes along the base, and column bases flanking the entrance. At Tell Taʿyinat, a smaller temple beside a palace shows the same tripartite plan and the same relationship: the temple is a royal chapel attached to the king's residence, not a separate civic institution. Tell Tayinat's Building II even sits on the same side of the palace complex. Other parallels come from Hazor, Ebla and Emar.

What that comparison establishes is that the biblical description is a plausible Iron Age Levantine temple of a known type, not an anachronistic fantasy. What it does not establish is that this particular building stood on this particular hill, at this particular size, in this particular decade. Those are separate claims, and the archaeology of the platform cannot test them.

Dating is contested too. The biblical account credits Solomon with the building in the tenth century BCE. Some archaeologists accept a tenth-century royal complex on the summit; others argue that Jerusalem in the tenth century was a modest highland town and that monumental construction on this scale belongs later, perhaps to the ninth or eighth century, with the tradition attaching itself to Solomon afterwards. Excavations in the City of David and the Ophel — the strip between the platform and the older settlement — have produced substantial Iron Age buildings, a large stepped stone structure, a monumental gate complex and fortification walls, and the interpretation of those remains is one of the most argued-over topics in the field. By the eighth and seventh centuries, the evidence for a large, administratively complex Jerusalem is overwhelming: seal impressions, a hugely expanded settlement onto the Western Hill, and the Siloam Tunnel.

Objects that plausibly relate to First Temple period activity on the mount do exist, but always at one remove. The Sifting Project has recovered Iron Age pottery, a bulla with a Hebrew name, arrowheads and a small piece of a proto-Aeolic-style capital fragment — from redeposited soil, so they prove presence, not placement. An ivory pomegranate once celebrated as a temple object was shown to carry a modern forged inscription on an ancient object. The Tel Dan and Mesha inscriptions attest a "house of David" as a political entity in the ninth century, which matters for the dynasty but says nothing about the building.

So the drawing here is deliberately schematic and small. The temple is placed over the rock only because the central reconstruction is the most widely held; move to the Herodian period and you can switch between three competing positions, and the same uncertainty applies backwards to this period. The dashed precinct around it has no measured basis at all — the texts describe an inner and an outer court without giving the enclosure's outer dimensions, and the platform of this period was certainly far smaller than the one you see today. What is drawn confidently is only this: a natural summit, a royal-temple complex of a recognizable regional type somewhere on it, and a city growing beneath it.

The texts also say something about how the building was made, and that part can be checked against the wider region. 1 Kings 5 describes an agreement with Hiram of Tyre for cedar and skilled labour, and Phoenician craftsmen working in bronze. Iron Age temples and palaces along the Levantine coast do show exactly that shared repertoire: cedar beams laid in courses with stone, ashlar masonry with header-and-stretcher patterns, proto-Aeolic capitals, ivory inlay, carved palmettes and cherub figures. Fragments of that repertoire have been excavated in Jerusalem and at Ramat Rahel just to the south. So while the temple itself is missing, the decorative and structural vocabulary the texts describe is genuinely attested in Judah.

One more caution about the sources. The descriptions in Kings, Chronicles and Ezekiel do not agree with one another, and none of them was written to help an architect. Chronicles gives the porch an impossible height. Ezekiel's temple is a vision of an ideal building, not a survey of a real one. Mishnah Middot, written after both temples were gone, is a legal text organizing memory. Every reconstruction drawing of the First Temple — including this one — quietly chooses between those sources, and the choices are rarely visible in the finished picture. That is why the elements here are drawn small, plain and tagged as textual.

In 587 or 586 BCE — the exact year depends on how you reconcile the regnal dating systems — a Babylonian army under Nebuchadnezzar II took Jerusalem after a siege of about eighteen months, burned the temple and the palace, broke down the city walls, and deported a substantial part of the population. This is one of the few events in the sequence where archaeology and text line up tightly, although once again the confirmation comes from everywhere except the platform.

In the City of David, excavators have found collapsed and burnt buildings of the early sixth century with charred beams, smashed storage jars, and a thick layer of ash and rubble tumbling down the slope. Bronze and iron arrowheads of both Judahite and Scythian-type forms — the latter used by the Babylonian army — were found together in the destruction debris of the so-called House of Ahiel and the Burnt Room. In the Jewish Quarter across the valley, a tower and a fallen mass of masonry with arrowheads at its base tell the same story. The Lachish Letters, ostraca found at another Judahite city, record a lookout watching for the fire signals of neighbouring towns and reporting that one of them can no longer be seen.

On the platform itself, none of that is visible, because nothing on the platform is visible. Whatever destruction layer belonged to the temple was removed, buried, or built over by later terracing — Hasmonean, Herodian, Umayyad, medieval. If any of it survives, it is under fills that no one has excavated and no one is likely to excavate.

The decades that follow are dim. The biblical books of Lamentations and parts of Jeremiah describe a ruined city. Archaeologically, Jerusalem in the sixth century looks small and poor, though not entirely empty; there is evidence of continued activity in the region around it, and the book of Jeremiah mentions people still bringing offerings to the site of the house after its destruction, which would imply some kind of continuing altar or ritual presence on the ruin. That is a text-based hint, not an excavated fact.

The return begins under Persian rule. Cyrus the Great took Babylon in 539 BCE, and the biblical account in Ezra 1 credits him with authorizing the return of exiles and the rebuilding of the temple. The Cyrus Cylinder, an actual Babylonian document, shows that this kind of policy — restoring sanctuaries and returning displaced cult images and peoples — was a real Persian practice, though the cylinder itself concerns Babylonian temples and does not mention Jerusalem. Persian imperial administration generally favoured local temples as instruments of provincial order and taxation, so the story is entirely consistent with the period, whatever the details.

Ezra 3 describes the altar being rebuilt first, before any building went up, so that sacrifices could resume. Work then stalled for years amid local opposition and, according to Haggai and Zechariah, a general lack of enthusiasm. The temple was finally dedicated in 516 BCE — traditionally the sixth year of Darius I — under the leadership of Zerubbabel, a descendant of the Judahite royal line, and the high priest Jeshua. Ezra preserves what purports to be a Persian building decree specifying dimensions and imperial funding.

The sources agree that this building was modest. Haggai 2 records old men who had seen the first house weeping at the sight of the new one; Ezra 3 describes shouting and weeping mixed together so that no one could tell them apart. Later Jewish tradition lists things the second temple lacked, above all the ark, which does not reappear in any historical source after 586 BCE. No description of Zerubbabel's building's plan survives in any detail, which is why it is drawn here as a dashed copy of the earlier plan: the assumption that it followed the old footprint is reasonable — rebuilders usually reuse foundations and remembered layouts — but it is an assumption.

Persian-period Jerusalem was a small provincial center, the capital of the province of Yehud. Excavated evidence is thin: some pottery, Yehud stamp impressions on jar handles, a few coins, scattered building remains. The city was probably a few thousand people at most, concentrated on the eastern ridge. Nehemiah's wall-building is famous, but identifying his wall on the ground has proved contentious, with different excavators proposing different lines. What matters for this visualization is the scale: the sacred precinct at this stage was a small enclosure on a hilltop, nothing like the vast esplanade that came later. The dashed present-day outline over this drawing shows the difference dramatically — five centuries of expansion still lay ahead.

Babylonian records add an outside voice. A cuneiform chronicle records Nebuchadnezzar's capture of Jerusalem in 597 BCE, when the young king Jehoiachin was deported — an event the Bible also reports. Ration tablets from Babylon list provisions for Jehoiachin and his sons, showing him alive in exile at the Babylonian court. Documents from a Judahite community settled in Babylonia record ordinary business dealings over several generations. None of these mention the temple's fate, but together they anchor the exile as an administrative reality, not just a scriptural theme.

The temple vessels are a thread worth following, because they show how memory works around an absence. 2 Kings says the bronze fittings were broken up and carried off. Ezra says gold and silver vessels were returned by Persian order and lists them. Later traditions kept multiplying claims about where the ark and the great candelabrum ended up: hidden on the mount, taken to Babylon, carried to Rome. None of it can be tested. The one image with any independent weight is the relief on the Arch of Titus, and that shows objects from the second temple, five centuries after Zerubbabel.

One last point about this period is easy to miss. The rebuilt temple began as a small provincial sanctuary, but the institution around it grew steadily under Persian and then Hellenistic rule: a priesthood with legal authority, an archive, a treasury, and a network of contributions from Jewish communities abroad. Documents from the Judahite garrison at Elephantine in Egypt show such communities corresponding with Jerusalem's authorities about religious practice in the fifth century BCE. The building on the hill stayed modest for a long time; the institution attached to it did not, and that is what made the later expansions possible.

Between the modest Persian-period sanctuary and Herod's colossal esplanade, there was an intermediate stage: a smaller stone-walled enclosure, expanded at least once, that Herod's engineers later wrapped around and buried. We can say that this earlier platform existed because we can see where Herod's masonry stops and something older begins.

The clearest evidence is a feature known as the "straight joint," visible in the eastern retaining wall about 32 meters north of the southeast corner. Above ground you can see a clean vertical seam. On one side the stones are unmistakably Herodian: large ashlars with finely drafted margins and flat, smooth bosses, laid in courses of consistent height. On the other side, to the north, the masonry is different — different margin widths, rougher or more prominent bosses, different course heights, different tooling. Charles Warren, tunneling down to bedrock in the 1860s, traced the seam far below ground and confirmed that the two builds go all the way down as separate constructions.

Everyone agrees the joint marks the southeast corner of a pre-Herodian enclosure. What is argued about is the date of the older masonry and the shape of the platform it belonged to. Proposals range from Persian-period work, to Hasmonean construction of the second century BCE, to the possibility that some of the lowest courses are older still. The style is often compared to Hellenistic bossed masonry at other sites, which pulls most scholars toward a Hasmonean date, but style-dating standing walls without excavated foundation deposits is inherently soft.

Onto this thin evidence, a much bigger reconstruction has been built. Mishnah Middot, compiled around 200 CE, states that the Temple Mount measured five hundred cubits by five hundred cubits — a square of roughly 250 meters a side. Leen Ritmeyer proposed that this square can be located on the ground: he takes the straight joint as its southeast corner, identifies a step in the bedrock and a section of wall in the northwest as another corner, and reconstructs a square enclosure with the rock inside it. Under his reading, part of that square could go back to the First Temple period, with Hasmonean extension northward.

This reconstruction is elegant and widely reproduced, and it is also contested. Critics point out that Middot is a rabbinic text written more than a century after the temple was destroyed, describing an idealized building whose measurements elsewhere disagree with Josephus, who actually saw the place. Reconstructing a physical platform from an idealized number, then anchoring it to one visible seam, is a chain of inferences where each link is plausible and none is proven. On the map here, the pre-Herodian square is drawn as a semi-transparent block with a dashed label, and it is tagged "contested" — because that is what it is.

The historical context is better known than the architecture. Under Persian and then Ptolemaic and Seleucid rule, Jerusalem remained a temple-city governed largely by its priesthood. The high priesthood became a political prize. In the 170s and 160s BCE, conflict between Hellenizing and traditionalist factions, and the intervention of the Seleucid king Antiochus IV, produced the desecration of the temple, the Maccabean revolt, and the rededication in 164 BCE commemorated as Hanukkah. The Seleucids built a fortress called the Akra to control the city and its temple; its location was debated for a century, and excavations in the Givati Parking Lot, west of the City of David, uncovered a fortification with Seleucid coins, lead sling bullets and bronze arrowheads that the excavators identify as the Akra. That identification is itself debated, but the finds are real and stratified.

The Hasmonean dynasty that emerged from the revolt ruled for about a century, and both the books of Maccabees and Josephus describe fortification and building around the temple, including a citadel at the northwest corner called the Baris, which Herod would later replace with the Antonia. Josephus also says the Hasmoneans enlarged the temple platform. So there is textual support for exactly what the wall masonry hints at: a phase of expansion before Herod, probably substantial, in the second century BCE.

Compare this period's outline with the dashed present-day one and you see the scale of what came next. Herod did not remodel this enclosure; he engulfed it. Everything drawn for this period lies inside the northern half of the later esplanade, and almost all of it is now buried under fills and paving that no one has dug.

Warren's underground work remains the backbone of what is known here. Because the ground outside the walls has risen many meters since antiquity, the seam at the straight joint continues far below the modern surface, and only a tunnel can follow it. Warren recorded that the older wall to the north sits on bedrock and that its courses do not bond with the Herodian work to the south — the two builds simply stand against one another. Later study of the masonry styles, the bosses, and the tool marks refined the picture but did not settle the date, because no excavation has ever reached a sealed deposit against the older foundation.

Why does the size of the earlier platform matter so much? Because it constrains where the temple could have stood. If the pre-Herodian enclosure really was a 250-meter square with the rock inside it, then the sanctuary must have been near the rock, and the central reconstruction gains strength. If the earlier enclosure was a different shape or sat differently on the ridge, the northern and southern proposals become easier to argue. So a single vertical seam in a wall, and one number in a rabbinic text, end up carrying an enormous amount of weight. Recognizing that is not scepticism for its own sake; it is the difference between a reconstruction and a measurement.

Two smaller pieces of evidence round out the period. First, tombs: rock-cut burial chambers of Hellenistic and early Roman date ring the city, and their architecture shows the same bossed masonry and Greek-influenced ornament as the enclosure's earlier walls, which is part of why a Hasmonean date is favoured. Second, coins: Hasmonean bronzes are extremely common in Jerusalem's excavated fills, and where they appear beneath a floor or behind a wall they give a date after which that construction happened. Neither can be applied to the platform, because there are no excavated fills on it — but they set the regional benchmark against which its masonry is read.

Around 20 or 19 BCE, Herod the Great began the project that produced the platform you can still walk on. He did not so much rebuild the temple as rebuild the mountain. The enclosure was roughly doubled, to about 144,000 square meters — some fourteen hectares, larger than any other sacred precinct in the Roman world. This is the one period where the archaeology is abundant, because the retaining walls that made the expansion possible are still standing, and the ground outside them has been excavated intensively for more than a century and a half.

The engineering problem was that the summit is a ridge, not a plateau. To make a level esplanade, Herod's builders raised enormous retaining walls around the slopes and filled the space behind them with earth, rubble and, where the drop was greatest, systems of vaults. The trapezoidal outline on this map is the result: about 315 meters along the north, 280 along the south, roughly 485 along the west and 470 along the east. The walls survive to great height, most spectacularly at the southeast corner, where they stand about 40 meters above the ancient ground level and continue perhaps 20 meters below it.

The masonry is unmistakable: ashlars with flat bosses and precisely cut drafted margins, laid in courses that step back slightly as they rise. Most blocks weigh two to ten tons. In the Western Wall Tunnel, the so-called Master Course contains stones far larger — one measured at roughly 13.6 meters long and estimated in the hundreds of tons, among the largest building stones ever moved in antiquity. No mortar was used. This is directly measurable evidence, and it anchors everything else.

Access to the platform was monumental. At the southwest, a broad staircase rose from the main street on an arch whose springing stones still project from the wall — Robinson's Arch, named for the American scholar who recognized it in 1838. Excavation below it revealed the pier that carried the far end, a row of vaulted shops in its base, and the paved street running north–south past them. Farther north, Wilson's Arch carried a bridge and aqueduct across the Tyropoeon Valley from the Upper City; it still stands, and recent excavation beneath it produced dated construction material and evidence of later rebuilding, including a small Roman theater-like structure tucked beneath. Barclay's Gate, its massive lintel still visible, opened at street level into a passage that climbed inside the platform. Warren's Gate, further north, is now deep inside the tunnel system.

On the south side, the two Hulda Gates — the Double Gate and the Triple Gate — were the main public entrances. Pilgrims climbed a broad monumental staircase of alternating wide and narrow treads, which is excavated and partly reconstructed, passed through the gates, and walked up sloping vaulted tunnels that emerged onto the esplanade. The domed vestibule behind the Double Gate, with its carved decoration, survives. Ritual bathing installations, or miqvaʾot, cluster below the stairs, which fits what the texts say about purification before entering.

Under the southeastern corner, where the natural ground falls away most sharply, the platform is carried on rows of piers and vaults, an area known since Crusader times as Solomon's Stables. The name is a medieval misnomer. The vaults as you see them today are largely medieval reconstruction over a Herodian-period substructure; distinguishing the phases is genuinely difficult, and the space was converted into the el-Marwani prayer hall in 1999.

Along the southern edge of the esplanade stood the Royal Stoa, described by Josephus as a basilica with four rows of columns forming a central nave and side aisles, running the whole width of the platform. Nothing of it stands, but architectural fragments — Corinthian capitals, cornices, column drums, carved friezes — were found in the debris fallen from above the southern wall, and they are consistent with the scale he describes. This is the pattern for everything on top of the platform: fallen fragments and texts, no in-place remains.

At the northwest, Herod rebuilt the earlier Hasmonean Baris as the Antonia Fortress, named for Mark Antony. Josephus describes a fortress with four towers overlooking the temple courts, with a garrison that could intervene during festivals. Its exact size and footprint are disputed; some reconstructions make it enormous, others compact, and the rock scarp at the northwest corner is variously read as its base or as an unrelated quarry face. It is tagged contested here.

Inside, the space was organized as a series of nested courts of increasing restriction. The vast outer area, the Court of the Gentiles, was open to anyone. Within it, a low stone barrier called the soreg marked the boundary beyond which non-Jews could not pass; two Greek inscriptions from it have actually been found, one complete in 1871 and a fragment in 1936, warning that any foreigner who enters is responsible for his own death. They are among the very few objects that can be tied directly to the temple's own architecture. Inside that came the Court of the Women, then the Nicanor Gate, the Court of Israel, the Court of the Priests with the great altar of unhewn stone, and finally the sanctuary itself, its facade sheathed in gold.

And here is the honest problem: we do not know where on the platform that complex stood. Mishnah Middot and Josephus both describe it, and they disagree with one another on measurements. Neither can be checked against excavation. The three reconstructions offered above — central, northern and southern — are the main contenders. The central one, in which the Foundation Stone lies inside the Holy of Holies and the Dome of the Rock sits over the sanctuary, is the most widely accepted; the other two are minority positions built on level calculations and alignments. Switch between them on the map and watch the entire inner complex slide across the esplanade. That movement is the most accurate thing this visualization can show you about the state of the evidence.

The logistics deserve attention, because they are part of the evidence. The stone came from quarries in and around the city; one enormous cavity under the Old City, known as Zedekiah's Cave, shows how quarrying was done, with channels cut around a block and wedges driven in to split it free. Blocks were dressed at or near the quarry, moved on rollers and sledges along prepared ramps, and set without mortar, their weight doing the work. Josephus says the courts took about eight years and the sanctuary itself about a year and a half, with work on the wider complex continuing for decades — indeed until a few years before the revolt. The scale of employment implied by that timetable was itself a political fact in Jerusalem.

Around the walls, excavation has recovered the ordinary machinery of pilgrimage. There are dozens of ritual bathing installations south and west of the enclosure, cut into bedrock with the divided steps typical of the period. There are shop fronts along the western street, stone weights and measures, coins struck by Roman governors and by the rebels, and quantities of chalk-stone vessels, used because stone was held not to transmit ritual impurity. Bones and cooking debris in the fills tell us what visitors ate. This is the best-documented aspect of the whole site, and it exists because it lies outside the walls where digging has been possible.

As for the three sanctuary positions, it is worth knowing what each side actually argues. The central case rests on the rock's prominence, on cuttings that Ritmeyer reads as wall sockets, and on the assumption that the Dome of the Rock deliberately marks the older sanctuary. The northern case rests on bedrock exposed near the Dome of the Spirits and on aligning the temple axis with the eastern gate. The southern case rests on level calculations — the claim that the rock is too high for the sanctuary floor if the courts are to fit — supported by thermographic and radar survey of questionable resolution. What would settle it is a controlled excavation of the esplanade, which will not happen. Absent that, honesty means drawing all three.

The Jewish revolt against Rome began in 66 CE. By the summer of 70, after months of siege under Titus, the Roman army broke into the temple precinct. The buildings on the platform burned and were then systematically dismantled. The city was taken, its walls torn down, its population killed, enslaved or scattered. The temple was never rebuilt.

Our narrative comes almost entirely from Josephus, who was present on the Roman side and whose account is detailed, vivid and self-interested. He wanted to explain the catastrophe to a Roman audience and to protect his own reputation, and he insists that Titus tried to spare the sanctuary and that the fire started against orders. Later Roman-era writers give a harsher version in which the destruction was deliberate policy. Historians read Josephus carefully rather than literally; his topography is often good, his numbers usually inflated, his motives always in play.

What excavation adds is startlingly direct, and it is all at the foot of the walls. Along the base of the Western Wall at the southwest corner, Benjamin Mazar's excavations in the late 1960s and 1970s uncovered the Herodian street still paved, and lying on it — cracking the paving stones where they landed — a heap of ashlars pushed from the top of the wall. They were not robbed away or reused; they lie where they fell. Standing there, you are looking at the physical moment of dismantling, preserved because the site was abandoned rather than rebuilt.

Among those fallen stones, in 1968, excavators found a block carved with a Hebrew inscription reading "to the place of trumpeting to de[clare]" — the text breaks off. Josephus mentions a point at the top of the wall where a priest stood to sound the trumpet announcing the beginning and end of the Sabbath. The stone appears to be from a parapet at the southwest corner, the very corner overlooking the Upper City and the main street. It is one of the few inscriptions that can be linked directly to the functioning of the temple, and it was found in the rubble of that function ending.

Elsewhere in the city, the destruction layer is unambiguous. In the Jewish Quarter, Nahman Avigad excavated what became known as the Burnt House, a basement of the priestly period with soot-blackened walls, smashed vessels, a stone weight inscribed with the name Bar Kathros — a priestly family named in rabbinic sources — and the skeletal remains of a forearm near a spear. The Palatial Mansion nearby, with its frescoes, mosaics and ritual baths, burned in the same event. Coins of the revolt years give the date.

On the platform itself, again, nothing has been excavated. Sifting Project material includes burnt debris, arrowheads and coins from the revolt, and a set of decorated opus sectile floor tile fragments whose geometry matches Herodian-period patterns; researchers have proposed they paved the porticoes. As always, they come from displaced soil, so they establish presence, not position.

The Roman side of the event is commemorated in Rome itself. The Arch of Titus, erected after his death, shows soldiers carrying spoils from the temple — a menorah, trumpets, what is usually read as the table of showbread. Coins struck across the empire read IVDAEA CAPTA. Vespasian redirected the annual half-shekel temple tax paid by Jews into the fiscus Judaicus, a levy for the temple of Jupiter Capitolinus in Rome. The destruction was converted into imperial revenue and imperial imagery.

The consequences reshaped both Judaism and, eventually, Christianity. With the sanctuary gone, sacrifice ended, and the priesthood lost its central function. The forms of Jewish practice that survived and developed — prayer, study, the synagogue, the rabbinic tradition eventually recorded in the Mishnah and Talmud — grew in the space the temple's absence left. The detailed temple descriptions in Mishnah Middot were written by people who had never seen the building, preserving a memory and a legal ideal. That is precisely why using them as architectural blueprints requires caution.

On this map, the period is drawn as an intact enclosure with an emptied interior. That is not a graphic shortcut — it is the situation. The retaining walls, being retaining walls holding back millions of tons of fill, largely survived; the buildings they carried did not. Everything after this point in the sequence is built on top of a Herodian foundation that nobody has been able to replace.

It is worth asking why the retaining walls survived when everything above them did not. A retaining wall is not a building; it is the face of a solid mass of fill. Pulling one down gains an attacker nothing and requires moving the earth behind it. The buildings on top, by contrast, were free-standing, roofed in timber, and full of the gold and marble that made demolition profitable. So the enclosure remained while its contents disappeared — and later builders, from the Umayyads to the Ottomans, kept reusing the same platform because rebuilding it would have been unthinkable.

Stone from the ruins did not stay put. Herodian ashlars were reused across the city for centuries; you can spot their drafted margins in later walls, and blocks with clearly Herodian tooling turn up in Byzantine, Umayyad and medieval construction. That recycling is one more reason the platform's own history is hard to read: material has moved sideways and upward for two thousand years. Meanwhile the destruction became a fixed point in Jewish memory, marked annually on the ninth of Av, and in Christian argument as a proof text. The physical event and its interpretations have travelled together ever since.

The Roman army's own traces are archaeologically visible around the city. Siege ramps, camp remains, ballista stones and legionary equipment have been identified at several points, and after the war the Tenth Legion was garrisoned in Jerusalem for some two centuries, leaving stamped tiles, pottery kilns and bathhouse fragments. The city that survived 70 CE was a military installation with a civilian fringe, not a functioning temple town. On the platform, the absence of any securely dated building for the next five centuries is the clearest measure of what the destruction actually accomplished.

For about six decades after 70 CE, Jerusalem was a ruin with a garrison. The Tenth Legion Fretensis was stationed there; its stamped roof tiles and bricks turn up across the city, and pottery kilns serving the legion have been excavated. The civilian settlement was small. Then, sometime around 130 CE, the emperor Hadrian decided to refound the city as a Roman colony named Aelia Capitolina — Aelia after his own family name, Capitolina after the Capitoline triad of Jupiter, Juno and Minerva.

The refoundation is usually connected to the outbreak of the Bar Kokhba revolt in 132 CE, though whether it was a cause or a consequence of the revolt is argued both ways in the ancient sources themselves. The revolt lasted until 135 and ended in catastrophic defeat. Afterwards, Jews were barred from the city, and the Roman colony was laid out on a standard plan with a main colonnaded street, a forum, bathhouses and temples. Substantial parts of that plan are excavated or visible: sections of the Cardo, the Ecce Homo arch that was once a triple gateway in a plaza, and the eastern Cardo found in more recent digs.

What happened on the old temple platform is much less clear, and this is one of the most genuinely open questions in the sequence. Cassius Dio, writing about eighty years later, says Hadrian founded a temple to Jupiter on the site of the temple of the god. If that is right, a Roman temple stood somewhere on the esplanade. But the Chronicon Paschale and other sources locate Hadrian's building elsewhere, and the Bordeaux Pilgrim, who visited in 333 CE and describes the site in detail, mentions two statues of Hadrian on the platform but does not describe a functioning temple there. Later Christian writers, who would have had every reason to mention a pagan temple on the temple site as a mark of divine judgement, are inconsistent.

Archaeology cannot resolve it because the surface has not been dug. The one tantalizing object is a large inscribed block built face-down into the southern wall of the Haram, above the Double Gate, honouring the emperor Antoninus Pius. It shows that Roman imperial monuments existed in the area — but the stone is in secondary use, which means it was moved, and it tells us nothing about where it originally stood. Statue bases, fragments of imperial sculpture and Roman-period architectural pieces have surfaced in the Sifting Project debris, again without context.

So this drawing shows a mostly empty platform, a dashed rectangle in an unspecified position labelled with a question mark, and two markers for the statues the Bordeaux Pilgrim reports. The empty space is not laziness; it is a claim about the evidence. The most economical reading of what survives is that after 70 CE the esplanade lost its buildings and never regained a major functioning one until the seventh century.

One extraordinary episode belongs at the end of this period. In 363 CE, the emperor Julian — raised Christian, who returned to traditional Roman religion and is known to history as Julian the Apostate — authorized the rebuilding of the Jewish temple in Jerusalem. His motives were partly religious policy and partly polemical: a rebuilt temple would embarrass Christian claims that the destruction was permanent divine judgement. Funds were allocated, an official was appointed, and work began on clearing the site.

It stopped almost immediately. Ammianus Marcellinus, a pagan historian and a reliable one, writes that balls of fire burst out near the foundations, burning the workers and making the site inaccessible, and that the project was abandoned. Christian writers seized on the story as a miracle. Modern explanations usually invoke an earthquake — one is attested in the region in 363 — perhaps igniting gases in the old vaults and cisterns, or simply collapsing the works. Julian died on campaign in Persia that same year, and no one revived the project.

No physical remains of Julian's attempt have ever been identified. A Hebrew inscription from Isaiah carved into a stone of the Western Wall has occasionally been connected to the hopes raised by his decree, but the dating is uncertain. On the map, the attempt is drawn as a hatched dashed strip with no fixed position, tagged textual only — a project that left a mark in the literature and, as far as we can tell, none in the ground.

The Bar Kokhba revolt left its own evidence, and some of it speaks directly to the temple. The rebels struck coins, usually by overstriking Roman ones, with legends such as "For the freedom of Jerusalem" and images of a four-columned facade with a star above it, a lulav and etrog, and trumpets. Whether the facade depicts the destroyed temple or a hoped-for rebuilt one, it shows that restoring the sanctuary was central to the movement's aims within sixty years of the destruction. Refuge caves in the Judaean Desert have produced the rebels' letters, contracts and household goods. What no source shows is any rebel construction on the platform itself.

After 135, Aelia Capitolina functioned as an ordinary Roman colony, and archaeologically it looks like one: a street grid, public baths, a theatre-like structure, drainage, inscriptions, and the standard imperial building types. The old temple enclosure sat inside that city as an enormous walled void. Whether it held a temple, a statue group, a plaza or nothing much, it was not the colony's religious center — that lay to the west. The most striking thing about the platform in this period is how little anyone seems to have said about it, and how much later writers, arguing about its meaning, wanted to fill that silence.

Two details of this period are often reported with more confidence than the evidence allows. The first is the claim that Hadrian renamed the city and banned Jews from it as a direct punishment after the revolt; the ban is well attested, but the sequence of refoundation and revolt is genuinely disputed among historians. The second is the idea that a statue of Hadrian stood exactly where the temple had been. The Bordeaux Pilgrim's two statues are real evidence that something imperial stood on the platform two centuries later, but he does not tell us where, or what had happened in between.

It is worth tracing how this drawing is built, because it is mostly built out of absences. The retaining walls are measured: they stand, and the Herodian courses can be counted, so the outline of the esplanade in this period is fixed by physical evidence. The surface is not. No Roman floor level has been excavated on the platform, so there is no stratigraphic horizon to hang a building on. The equestrian statue and the temple of Jupiter come from texts and coins — the coins of Aelia Capitolina show a temple, but coins are civic advertising and do not fix a location. So the Jupiter temple here is drawn as a faint dashed footprint at platform level, with no height at all: a claim about presence, not about architecture. Julian's attempt in 363 is textual, and drawn as a shallow scar rather than a building, because that is what the sources describe — foundations begun and abandoned.

When the Roman Empire became Christian in the fourth century, Jerusalem was transformed into a pilgrimage city — and the temple platform was pointedly left out of the transformation. This is one of the rare cases where an absence is itself a deliberate historical act, well attested in writing.

Constantine's building program concentrated on sites associated with Jesus. The Church of the Holy Sepulchre, begun in the 320s over what was identified as the tomb and the hill of the crucifixion, became the city's monumental center; the Eleona on the Mount of Olives and the Church of the Nativity in Bethlehem followed. Over the next two centuries, Jerusalem filled with churches, monasteries, hospices and pilgrim infrastructure. The Nea Church, dedicated by Justinian in 543, was enormous; part of its substructure was excavated in the Jewish Quarter, and it appears on the sixth-century Madaba mosaic map along with the colonnaded Cardo built to connect it to the Holy Sepulchre.

On the Madaba map, the old temple enclosure is not depicted as a building at all. That silence matches the written sources. The Bordeaux Pilgrim in 333 walks the platform and describes ruins, statues, and a perforated or pierced stone that Jews came to anoint and mourn over once a year, tearing their garments before departing. Later pilgrims and church writers describe an area of rubbish and neglect. Jerome, living in Bethlehem around 400, writes with satisfaction of the site's desolation as fulfilment of prophecy.

The neglect was theological. For Christian authorities of the period, the ruined temple mount was evidence: a visible demonstration that the old covenant had been superseded and that Jesus' prophecy of not one stone left upon another had come true. Building on it would have undermined the argument. So while the rest of the city was covered in marble and mosaic, the largest open space in Jerusalem was left as a dump. Some sources describe the city's refuse being carried there.

There are hints that this was not total. A handful of scholars have suggested a small church or chapel somewhere on the esplanade, based on ambiguous references and on Byzantine architectural fragments and mosaic tesserae found in later fills. The Sifting Project has produced quantities of Byzantine-period material: coins, pottery, tesserae, a few pieces of carved stone. That tells us people were on the mount and that Byzantine material ended up in its soil — for a site next to a busy city, hardly surprising, and not enough to place a building.

For Jewish communities, access varied with the politics of the moment. The ban imposed after the Bar Kokhba revolt was enforced with different degrees of strictness at different times; the Bordeaux Pilgrim's account of an annual visit implies some tolerated access, at least on the ninth of Av, the fast commemorating the destructions. Under Julian, briefly, everything seemed about to change; after him, it did not. Sources from the fifth and sixth centuries suggest periods of prohibition and periods of relative permission.

Two upheavals close the period. In 614, a Sasanian Persian army took Jerusalem, with Jewish support in the expectation of restored access to the site; churches were damaged, and the relic of the True Cross was carried off. There was, according to some sources, an attempt at renewed Jewish worship on the platform. Byzantine rule returned in 630 when Heraclius recovered the city and the relic. Eight years later, Jerusalem surrendered to the armies of the caliph ʿUmar.

The traditional account of that surrender turns on exactly the condition described here. Muslim sources report that ʿUmar asked to be shown the site of the temple, found it buried under refuse, and had it cleared, working with his own hands to begin. Whether or not the details are accurate, the story only makes sense if the platform really was a rubbish tip — which is what Christian sources, hostile to the site for entirely different reasons, had been saying for three hundred years. Two independent traditions with opposite motives describing the same condition is about as good as evidence gets when there is nothing to excavate.

The drawing for this period is therefore mostly empty: the Herodian enclosure intact, a hatched area of debris across the interior, and the rock marked as the one feature the sources keep returning to. It is worth pausing here before moving the slider. The next period transforms this space into one of the most sophisticated architectural ensembles of the early medieval world — and it does so, essentially, on an abandoned rubbish tip inside walls that were already six hundred years old.

The pilgrim literature is worth taking seriously as evidence, with care. Travellers such as the Bordeaux Pilgrim, Egeria and later Piacenza pilgrim wrote practical itineraries: how many miles between towns, which liturgies to attend, what could be seen. Their descriptions of Jerusalem match the archaeology of the Byzantine city closely enough that historians trust their basic topography. When those same writers pass over the temple platform in a few sentences of rubble and ruin, the silence is informative. Pilgrims described what was worth a detour, and the platform was not.

The Sasanian interlude of 614–630 is often treated as a dramatic reversal, and the sources on it are highly partisan on all sides — Christian accounts of massacre and desecration, later Jewish texts of messianic hope. What archaeology shows is damage and repair in several churches and monasteries around the city, and burials that some excavators connect to the conquest. Whether any construction or renewed worship took place on the platform during those sixteen years is unknown; nothing survives to test it. When Heraclius restored Byzantine rule in 630 the arrangement lasted less than a decade, and the platform passed to Muslim rule still in the condition the pilgrims had been describing for three centuries.

It is also worth noting what the Byzantine city did with the enclosure's edges. The walls remained standing and continued to define the eastern side of the city; roads ran alongside them, and buildings were put up against them in places. The southern wall area saw Byzantine houses and workshops, excavated in the Ophel. So the platform was not sealed off — people lived and worked right up against it. It was the surface on top, and only that surface, which was left as ruin, and the choice was deliberate rather than accidental.

The evidence chain for the Byzantine platform is short and unusually consistent. Step one is the walls, which continued to stand and still do. Step two is the absence of any excavated Byzantine floor on the esplanade, which means nothing can be positioned by stratigraphy. Step three is a set of pilgrim accounts — most usefully the Bordeaux pilgrim of 333 and the Piacenza pilgrim later — which describe an open, ruined area with a pierced stone visited by Jews once a year. Step four is inference from what Byzantine Jerusalem built elsewhere: the great churches went up on the western hill and at the Holy Sepulchre, not here. The drawing therefore shows heaps and hollows rather than structures, and the only marked feature is the rock itself. There is no reconstructed height in this period because there is nothing whose height could be argued.

Jerusalem came under Muslim rule in 638. Within about fifty years, the neglected esplanade became al-Ḥaram al-Sharīf, the Noble Sanctuary — and the building put at its center is still standing, still substantially original, and dated by its own inscription. After eight periods of reconstruction from texts and fragments, this is the moment when the visualization can finally show a building that actually exists.

The site's significance in Islam rests on its identification with al-Masjid al-Aqṣā, "the farthest mosque," named in Qurʾan 17:1 as the destination of the Prophet Muhammad's Night Journey, and on its association with the ascension, the miʿrāj. It is also understood as the place of the earlier prophets, including Solomon and David, whose sanctuary tradition Islam inherits. The three names matter for reading this map: the whole enclosure is al-Aqsa or the Haram; the silver-domed congregational mosque at the south is also called al-Aqsa or the Qibli Mosque; the golden-domed building over the rock is the Dome of the Rock, which is a shrine, not a mosque.

The first mosque was modest. The Gallic bishop Arculf, who visited around 670 and whose account was written down by Adomnán, describes a rectangular building of beams and boards at the southern end of the platform, large enough for a very large congregation. That is a plausible description of an early hypostyle mosque built quickly of reused materials. Nothing of it has been identified, though excavations under the present mosque floor by R. W. Hamilton in 1938–42, during structural repairs, revealed earlier walls and phases.

The Dome of the Rock is another matter entirely. Completed in 691 or 692 under the caliph ʿAbd al-Malik, it is an octagonal building with two concentric ambulatories around the rock and a tall wooden dome on a drum, about 20 meters across, carried on piers and columns. The interior is covered in glass mosaic — vegetal scrolls, jewels, crowns and vessels, without human or animal figures — and a mosaic inscription band over 240 meters long runs around the arcades, quoting the Qurʾan and naming the patron and the date. Its geometry is exceptionally refined: the plan is generated from a small number of related measurements, and the proportions of dome, drum and octagon are tightly coordinated.

Its purpose has been debated for a century. Older theories held that it was built to substitute for the Kaʿba during a period when a rival caliph controlled Mecca — a reading most scholars now reject. More persuasive arguments emphasize the inscriptions, which address Christian doctrine directly, and the building's commanding position in a city full of domed churches: an assertion, in the visual language of the region, of the new order's presence and permanence. The choice of the rock also connects the building to the site's earlier sacred history.

The Dome of the Rock does not stand on the general esplanade but on a raised inner platform, reached by flights of steps topped with free-standing arcades, the qanāṭir. When that raised platform was created is not securely known, since it too has never been excavated. Nearby, the small Dome of the Chain — an open, domed structure on columns, roughly at the geometric center of the esplanade — is Umayyad in origin but has been rebuilt so many times that little original fabric can be identified. Its function has been read as a scale model, a treasury, or a place of judgement.

The congregational mosque was rebuilt in permanent form under al-Walid I in the early eighth century, in stone with a wide hypostyle hall. Administrative papyri found at Aphrodito in Egypt record workers and materials dispatched to the mosque in Jerusalem — the sort of dry bureaucratic evidence historians prize, because nobody forges a payroll for theological reasons.

South and southwest of the enclosure, outside the walls, excavations directed by Benjamin Mazar and Meir Ben-Dov in the 1960s and 70s uncovered a group of large Umayyad courtyard buildings, one of them connected by a bridge to a gate in the Haram wall. These are usually interpreted as administrative palaces or governmental complexes, and they are fully excavated with stratigraphy and dating evidence — an important corrective to the idea that nothing around this site is well documented. They show the Haram in its Umayyad setting: not an isolated shrine but the ceremonial core of a governing complex.

In 749, a major earthquake struck the region. The palace complex was damaged and largely abandoned; the mosque required substantial rebuilding, the first of many. With the Abbasid caliphate's move to Baghdad, Jerusalem lost political centrality, though patronage continued: caliphs repaired the buildings, and the geographer al-Muqaddasi, a native of the city writing in the tenth century, describes the Haram with local pride. Another earthquake in 1033 brought more rebuilding under the Fatimids, which is when much of the present mosque's plan took shape.

The Dome of the Rock also shows how Umayyad builders worked with what the city already contained. Its columns and capitals are reused Roman and Byzantine pieces, gathered and matched by height with the help of extra bases and shims. The dome is a double timber shell, light enough for the drum to carry, a technique with Byzantine precedent. The mosaics use glass tesserae in the eastern Mediterranean tradition, almost certainly laid by craftsmen trained in that tradition, but the imagery avoids figures — a choice that in itself marks a new religious setting. The building is thus both continuous with local practice and clearly declaring something different.

The wider Haram was organized as well. Gates in the walls were reopened or rebuilt, cisterns and channels maintained, and the aqueduct that had supplied the enclosure since antiquity kept in use. A stair and bridge connected the palace complex to the southern gates, so the governor could enter the sanctuary directly — an arrangement that says a good deal about how religious and political authority were meant to be seen together.

Later writers give us the place as it was used. Al-Muqaddasi, a native of Jerusalem writing in the tenth century, describes the buildings, the gates, the cisterns and the crowds with the precision of someone who grew up walking there, and complains about the city's Christian majority and high prices. Nasir-i Khusraw, visiting in 1047, measured and counted. After the 1033 earthquake, Fatimid rebuilding narrowed and reshaped the mosque into roughly the form still standing. By the eve of the Crusades the Haram already looked, in outline, like the site on this map.

The early Islamic period is the first in the sequence where the chain runs the whole way to a measured height, and the difference is worth naming. The Dome of the Rock stands; its plan, its two concentric ambulatories, its column and pier spacings, its dome diameter and its interior mosaic inscription with the date 691/692 are all directly measurable, so nothing here is reconstructed. Al-Aqsa is harder: the present building is later, but the 1938–42 repairs recorded earlier phases beneath the floor, and those records — a rescue excavation inside a working mosque, not a survey — are the only stratigraphy the platform has ever produced. The Umayyad buildings at the southwest corner are excavated, outside the enclosure, by Mazar and Ben-Dov. So in this period the drawing is mostly measurement, and the conjectural parts are the connective tissue: paving lines, minor arcades, and how the courtyard was subdivided.

On 15 July 1099, the armies of the First Crusade broke into Jerusalem after a five-week siege. The massacre that followed, in which large numbers of the city's Muslim and Jewish inhabitants were killed, is described in horrified detail by both Muslim chroniclers and the crusaders' own writers, some of whom recorded it as a triumph. For the next eighty-eight years, the Haram was the ceremonial heart of a Latin Christian kingdom.

The Crusaders read the site through their own scriptural map. The Dome of the Rock was identified as the Temple of the Lord — Templum Domini — believed to be either the temple of Solomon itself or its direct successor, the building Jesus knew. The al-Aqsa Mosque became the Templum Salomonis, Solomon's palace. Neither identification is historically correct, but both shaped what happened to the buildings.

The Dome of the Rock was consecrated as a church and given to a community of Augustinian canons. A gold cross was fixed above the dome. The rock itself was paved over with marble, an altar was built on it, and an iron grille was set around it — partly for liturgy and partly to stop pilgrims chipping off pieces to take home, a real and documented problem. Interior fittings, painted decoration and Latin inscriptions were added. Remarkably, the Umayyad mosaics were largely left in place, which is a substantial reason they survive.

The building's silhouette travelled. Because Europeans believed it was Solomon's Temple, its centralized octagonal plan was copied and echoed in churches across Europe, most famously by the Templar order, and it appears on seals and in manuscript illustration as the standard image of "the Temple." A great deal of European visual tradition about the Jerusalem temple derives, by way of this misidentification, from a seventh-century Islamic shrine.

The al-Aqsa Mosque first served as the royal palace of the kings of Jerusalem. In 1119 King Baldwin II granted it to a new order of knights sworn to protect pilgrims; from their quarters in the supposed Temple of Solomon they took the name Poor Fellow-Soldiers of Christ and of the Temple of Solomon — the Knights Templar. Over the following decades the order expanded the complex substantially, adding halls, workshops, storerooms, cloister ranges and a church to the west and south of the mosque. Standing Crusader vaults, capitals, carved doorways and rib-vaulted rooms survive both in and beside the building; some Crusader-carved capitals were later reused in Muslim buildings on the platform.

The vast vaulted space under the southeastern corner was used for horses and equipment. This is the origin of the name Solomon's Stables, which stuck for eight centuries and misled generations of visitors into thinking the vaults were tenth-century BCE work. They are not: the substructure is Herodian in origin, and the vaulting as it now stands is largely medieval — which is why the key tags them "inferred," not "excavated."

Muslims and Jews were generally barred from living in the city under Crusader rule, though the picture became more complex over time, with pilgrimage and trade continuing and some communities returning. The Spanish Jewish poet and philosopher Judah Halevi and the traveller Benjamin of Tudela both engaged with Jerusalem in this era, and Benjamin's twelfth-century itinerary records a small Jewish presence in the city.

The kingdom's defeat at the Horns of Hattin in July 1187 destroyed its field army, and Saladin took Jerusalem that October. The terms were negotiated rather than stormed: inhabitants could ransom themselves, and there was no general massacre. The cross was removed from the dome, the Christian additions inside were taken out, the rock was uncovered again, and the buildings were ritually washed with rosewater before being restored to Muslim worship. The grille around the rock, however, stayed in place for centuries — one of many small pieces of Crusader fabric absorbed into the site.

The Crusader layer is unusually well evidenced compared with most periods here, because so much of it is standing architecture that can be examined directly, and because both sides produced detailed written accounts. It is also a useful reminder for reading the whole sequence: the same stones have been continuously reinterpreted by each community that held them, and several of the site's most familiar names — including "Solomon's Stables" — are the fossils of interpretations that turned out to be wrong.

The Crusader Haram was also a working economic and institutional space. Pilgrims paid for candles, guides and relics; the canons of the Templum Domini held property across the kingdom; the Templars ran workshops, stores and stables and controlled one of the city's most valuable enclosures. Written charters, seals and building accounts document all of this, which is why we can describe the period's administration in more detail than its predecessors — parchment survives where excavation is impossible.

Something of the Crusader Haram travelled outward, too. The dome's octagonal image spread through European art and architecture; carved capitals and sculptural fragments made in Jerusalem workshops of this period are among the finest Romanesque work in the region, and some were later dismantled and reused in Muslim buildings on and around the platform. For the communities excluded from the city during much of this period, the site remained a focus of writing, prayer and pilgrimage from a distance. When Saladin took the city, the change of hands was fast; the physical fabric absorbed the eighty-eight years and kept going.

For archaeologists, the Crusader century is unusually legible on this site precisely because so much of it is still in place. Standing rib vaults, moulded doorways, carved capitals, masons' marks and rebuilt gate passages can be examined, measured and compared with dated Crusader work elsewhere in the kingdom — at Acre, Nazareth and Belvoir. The result is that for one period out of twelve, we can talk about construction sequence and workshop practice on the platform itself with real confidence.

That confidence has a limit worth naming. What can be read is the fabric of two buildings and the additions around them, not the ground beneath. Nobody has excavated the Crusader floors, drains or refuse deposits on the esplanade, so questions about how the space was actually used day to day — where pilgrims queued, what was sold, where the stables' fittings sat — are answered from charters and chronicles rather than from soil.

For the Crusader century the chain has an unusual middle link: images. The walls and the two main buildings still stand, so the frame is measured. The changes the Crusaders made — an altar and choir screen inside the Dome of the Rock, a wrought-iron grille around the rock, an eastern porch, the conversion of al-Aqsa into the Templars' headquarters and the vaults into stables — are described in Latin chronicles and pilgrim accounts, and several survive as physical fragments: the grille itself was in place until the twentieth century, and Crusader capitals and sculpture from the platform are catalogued. Twelfth-century seals of the Templars and maps of Jerusalem show the Templum Domini as a domed building, which corroborates identification without adding dimensions. Nothing here needs a reconstructed height, because the buildings whose heights matter were already standing.

The seven centuries between Saladin's conquest in 1187 and the end of Ottoman rule in 1917 gave the Haram most of the appearance it has today. Little of it involved new construction at the center; almost all of it was restoration, refinement, and a gradual filling-in of the edges. Because these buildings are standing and most carry dated inscriptions, this is the best-documented period in the whole sequence.

Saladin's own program was one of restoration. The Dome of the Rock was returned to its Islamic form and repaired. In the al-Aqsa Mosque he installed a carved wooden minbar, or pulpit, that had been commissioned decades earlier by Nur al-Din of Aleppo in anticipation of Jerusalem's recovery. It was a masterpiece of interlocking joinery — assembled from thousands of pieces of wood and ivory without nails — and it stood in the mosque for nearly eight centuries until it was destroyed by an arson attack in 1969. A painstaking replica, reconstructed from photographs and measured drawings, was installed in 2007. The Ayyubids also added arcades and refurbished gates, and the mosque's mosaic-decorated drum was restored.

Under the Mamluk sultans, who ruled from Cairo from 1250 to 1517, Jerusalem became a major center of religious learning and pious endowment. This produced the single most characteristic feature of the Haram's present appearance: the dense ring of buildings along its western and northern edges. Dozens of madrasas — colleges of law and religious study — plus ribāṭs for pilgrims and Sufi lodges, tomb-chambers for their founders, and sabils, or public fountains, were built directly against the enclosure walls, opening onto the esplanade through richly carved portals.

Mamluk architecture in Jerusalem is instantly recognizable: alternating courses of light and dark stone called ablaq, elaborately carved stalactite vaulting called muqarnas over doorways, and long carved foundation inscriptions naming the patron and the exact year. Those inscriptions are why this period is so securely dated: the buildings state their own construction dates in their own stonework. Among the best known are the Ashrafiyya Madrasa, described by a contemporary as the third jewel of the Haram, and the Tankiziyya. Together they created a kind of urban wall of institutions along the western edge, mediating between the city streets and the open sacred space.

The finest freestanding Mamluk work on the platform is Sabīl Qāytbāy, built in 1482 for the sultan of that name, just west of the raised platform. It is a domed fountain over a cistern, and its dome is carved with an intricate arabesque relief — the only stone dome of its type in the city, and generally considered the high point of Mamluk stonework in Jerusalem. The arcades at the heads of the stairways onto the upper platform, the qanāṭir, were built and rebuilt across the Ayyubid and Mamluk periods; each set frames a view of the Dome of the Rock and structures the approach to it.

The Ottomans took Jerusalem in 1517, and under Süleyman the Magnificent the city received the works that most define it today. Between 1537 and 1541, the city walls were rebuilt on largely medieval lines — the walls tourists walk now are Süleyman's, not Herod's or Hadrian's, though they incorporate earlier fabric and reused blocks. He also built or restored a series of public fountains and rebuilt water supply infrastructure.

The most visible Ottoman intervention on the Haram came in 1545, when the decaying exterior mosaics of the Dome of the Rock were replaced with glazed ceramic tiles. The blue, turquoise, white and yellow revetment that makes the building so recognizable in photographs is Ottoman, not Umayyad; it has been repaired and renewed repeatedly since, including a full replacement in the twentieth century. Windows were reglazed with stucco-and-glass grilles, and inscription bands in Ottoman-period calligraphy were added.

Alongside all this, the site's other communities lived with a much more constrained arrangement. Access to the Haram itself was restricted to Muslims, and Jewish prayer concentrated at the exposed stretch of the Herodian western retaining wall, in a narrow alley within the Mughrabi Quarter — the origin of the Western Wall as a fixed place of prayer in its modern form. Ottoman-era documents record repeated disputes over rights and furnishings in that alley, disputes that would harden in the nineteenth and twentieth centuries into the arrangements known as the Status Quo.

In the nineteenth century, European surveyors arrived. Charles Wilson's Ordnance Survey of Jerusalem in 1864–65 produced the first accurate measured plans; Charles Warren's shafts and tunnels of 1867–70 reached bedrock at the foot of the walls and mapped cisterns and substructures. Their drawings remain primary evidence, because much of what they recorded is now inaccessible. Almost everything modern scholarship knows about the enclosure's foundations begins with those two surveys.

Behind almost every building of these centuries stands the waqf, the Islamic charitable endowment. A patron would assign the income of shops, farmland or a bathhouse permanently to a madrasa, fountain or tomb, funding its staff and repairs. Endowment deeds survive in quantity and name founders, salaries, teaching duties and maintenance obligations. That system explains both the density of institutions around the Haram and their long survival: each came with its own income stream. It also means the site's history in this period can be read in legal documents as well as in stone.

Practical upkeep mattered as much as patronage. The platform's cisterns and the aqueduct from the south had to be cleaned and repaired; the arcades, paving and stairs needed periodic rebuilding; the Dome's timber structure, lead and later tiles required constant attention, and earthquakes in this long span forced several campaigns of repair. Inscriptions record many of these works with their dates, which is how a site that has never been excavated nevertheless has a fairly detailed building history for its last eight hundred years — the buildings tell you themselves.

The Haram in these centuries was also a lived-in landscape rather than a monument. Sources and travellers' accounts describe teaching circles in the arcades, water sellers, gardeners tending the olive and cypress trees on the esplanade, custodians with keys, and the daily and Friday rhythms of prayer that organized the whole space. Families held hereditary offices connected to the sanctuary for generations. None of this leaves a footprint on a plan, which is one reason a drawing like this one, however carefully tagged, will always understate what the site actually was.

These seven centuries are the best-documented in the whole sequence, and for once the documents are administrative rather than literary. The buildings along the northern and western edges — madrasas, ribats, the Ashrafiyya, the arcades and fountains — are standing, dated by their own foundation inscriptions, and cross-checked against waqf endowment deeds that name patrons, dates and functions. That is a chain with no conjectural link in it: inscription, deed, standing wall. The Ottoman contribution is equally documented, notably the 1545 tile revetment of the Dome of the Rock, which is dated by its own inscriptions. What remains uncertain is not the buildings but the ground under them: no excavation accompanies any of this construction, so what these foundations sit on, and what they cut through, is unknown. The drawing shows the buildings at their measured heights and says nothing about what is beneath.

The modern history of the Haram is inseparable from the political history of Jerusalem, and it is where any honest visualization has to be most careful. What follows describes what each community holds and what happened, without adjudicating claims that remain live.

Under the British Mandate after 1917, the buildings were surveyed and repaired. The architect Ernest Richmond documented the Dome of the Rock in detail, and structural work on the al-Aqsa Mosque in the late 1930s and early 1940s, supervised by R. W. Hamilton, gave the only opportunity to date to examine earlier phases beneath the mosque floor — the closest thing to an excavation the platform has ever had, and it was a rescue operation inside one building, not a survey of the site. The period also saw serious intercommunal violence connected to the site, including the 1929 riots that followed disputes over prayer arrangements at the Western Wall.

From 1948 to 1967 the Old City, including the Haram, was under Jordanian control, and the Islamic Waqf administered the site under Jordanian custodianship. During those years Jews were in practice unable to reach the Western Wall. The major physical change of the period was the dome itself: the old lead sheathing had decayed badly, and between 1959 and 1964 it was replaced with a covering of anodised aluminium-bronze alloy, which gave the building a duller bronze colour familiar from photographs of the era. In 1993 it was recovered again, in gold-plated copper funded by King Hussein of Jordan, producing the bright gold surface seen today.

In June 1967, Israel captured the Old City. Within days, the Mughrabi Quarter in front of the Western Wall was demolished and its residents relocated, creating the large open plaza that exists now. Administration of the Haram itself was left with the Waqf under an arrangement usually described as the Status Quo: the Waqf manages the site and Muslim worship there; Israel controls external security and access; non-Muslim visitors may enter at set times but non-Muslim prayer on the platform has been prohibited under the arrangement. The Status Quo has been repeatedly contested and repeatedly strained, and disputes over its interpretation have triggered major episodes of violence. Sovereignty over the site is one of the central unresolved issues of the Israeli–Palestinian conflict.

For Muslims, the Haram is the third holiest site in Islam, the place of the Night Journey and ascension, and a functioning space of daily prayer for a living community. For Jews, the Temple Mount is the holiest site in Judaism, the location of both temples and the focus of prayer and mourning for two millennia; many rabbinic authorities forbid entry to the platform on grounds of ritual purity, while other groups advocate visiting. For Christians, the site carries the memory of the temple of the Gospels. These are not competing accounts of the same evidence — they are different kinds of claim, and archaeology does not arbitrate between them.

In 1999, the Waqf converted the vaulted southeastern substructure — the space long called Solomon's Stables — into the el-Marwani Mosque, a prayer hall for thousands of worshippers. Creating an emergency exit involved removing a large quantity of fill, reportedly hundreds of truckloads, which was dumped in the Kidron Valley. Israeli archaeologists protested that antiquities-rich soil had been removed without archaeological supervision; the Waqf's position was that the work was maintenance of an active religious building on a site under its administration. The episode is a compressed version of the whole problem: a living sanctuary and an archaeological site occupying exactly the same ground.

Out of that dispute came the Temple Mount Sifting Project, begun in 2004, which wet-sifts the discarded soil at an off-site facility, much of the labour done by volunteers. The results are genuinely valuable: tens of thousands of coins, pottery from every period, arrowheads, seal impressions, jewellery, glass, hundreds of thousands of mosaic tesserae, and a set of stone floor tile fragments that researchers reconstruct as Herodian-period opus sectile paving. But the material has no stratigraphy. It cannot be assigned to a layer, a building or a room. It can tell you what was on the mount; it cannot tell you where, when in relation to what, or with what.

That is the note to end on, and it is the reason this visualization is built the way it is. Move the slider back through twelve periods and watch what the confidence tags do. At the modern end and in the early Islamic period, most elements are standing buildings you could photograph. In the Herodian period, the walls and gates are excavated but the entire sacred complex on top is textual or contested. In the First Temple period, everything is textual. The shapes on the map look equally solid at every point in the sequence — that is a limitation of drawing, not a claim about knowledge. The tags, not the shapes, carry the truth.

The site's modern status is also formal. The Old City and its walls were listed as a UNESCO World Heritage Site in 1981 and placed on the List of World Heritage in Danger the following year, and the condition and management of the Haram's monuments have been the subject of international reporting ever since. Conservation work continues under the Waqf, with periodic structural surveys of the Dome of the Rock and the mosque, repairs to the arcades and paving, and specialist attention to the mosaics and the timber dome. Because the site is in daily religious use, conservation is a continuous, negotiated activity rather than a project with an end date.

Research under these conditions has become mostly non-invasive. Scholars work from measured surveys and photogrammetry, from historical drawings such as Wilson's and Warren's, from published records of the 1938–42 repairs, from the walls that can be reached from outside, and from sieved debris. Ground-penetrating radar and thermography have been tried and produce results too ambiguous to settle the questions people most want answered. Some things will simply remain unknown: what lies under the esplanade's paving, whether First Temple foundations survive, and exactly where the sanctuary stood.

That is not a failure of archaeology so much as a limit set by the site's own life. A place that has been continuously sacred, rebuilt and contested for three thousand years is not a stratified excavation waiting to be opened; it is a working sanctuary, a heritage monument, and a political fact at the same time. The most accurate visualization is one that says clearly which lines are measured, which are argued, and which are simply drawn because a text describes them.

The modern period closes the loop by making the evidence chain visible as a method rather than a result. Everything drawn is measurable: the enclosure, the two great buildings, the arcades, the plaza cut in 1967. The reconstruction problem has moved underground and off-site — to Warren's nineteenth-century shafts and galleries, to the 1938–42 records beneath al-Aqsa, to ground-penetrating radar that returns ambiguous readings, and to the sifted soil from the 1999 works, which yields material without position. Set the slider back through the twelve periods and the chain shortens as you go: at this end it runs walls → standing buildings → survey; in the Herodian period it runs walls → excavated gates → Josephus → reconstructed height; in the First Temple period it runs walls that did not yet exist → 1 Kings → comparison with other Levantine temples. Same drawing conventions, radically different warrant. That is the thing this tool is built to show.