Building the Tabernacle in Your Mind

The description of mishkan (Tabernacle) in Exodus is often read as a set of instructions. At first blush, this reading fits the plot: as the story goes, these are the instructions our forebears received to build just such a structure, and build it they did. But as many readers over the generations have noticed, once you get into the details, this text is difficult to use as instructions.

Some of the difficulty is because it uses esoteric vocabulary; some of it is the near-hypnotizing level of repetition that can make it read more like poetry than directions to be executed. In addition to these literary issues, there are difficulties with the details the text provides—or doesn’t provide—especially in places where component parts of the structure are to be joined together. Indeed, for all the dioramas this text has inspired, William Propp of UC San Diego notes that “nobody has yet definitively calculated the Tabernacle dimensions.”^[1]

Ambiguity at Points of Juncture in the Mishkan Text

Three parts of the wooden frame for the mishkan are left sufficiently unclear in the text that generations of scholars have been left to disagree on the structure’s dimensions: the thickness of the planks of the frame, the assembly/attachment of the planks, and the way the corner pieces fit into the whole.

How thick are the planks that frame the structure? The Tabernacle is a rectangular-cuboid structure, whose frame is made of acacia-wood planks:

^{שמות כו:טו} וְעָשִׂיתָ אֶת הַקְּרָשִׁים לַמִּשְׁכָּן עֲצֵי שִׁטִּים עֹמְדִים. ^כו:טז עֶשֶׂר אַמּוֹת אֹרֶךְ הַקָּרֶשׁ וְאַמָּה וַחֲצִי הָאַמָּה רֹחַב הַקֶּרֶשׁ הָאֶחָד.

^{Exod 26:15} You shall make the planks for the Tabernacle of acacia wood, upright. ^26:16 The length of each plank shall be ten cubits and the width of each plank a cubit and a half.

The Torah gives us the length and width of these planks, but says nothing about the third dimension, the depth – a detail you cannot ignore if you are trying to build this structure. Close readers of this text have filled in this gap quite differently. For example, Propp assumes that the thickness is sufficiently small such that it did not affect the calculation of the Tabernacle’s size. ^[2] This could explain why the detail is absent from the biblical text, though it seems structurally precarious: the planks stand about 15 feet tall (10 cubits) and support multiple layers of covers.

Opting for a more solid construction, the Babylonian Talmud (b. Shabbat 98a) images the planks were a full cubit thick—about 18 inches!^[3] The difference between the minimal and maximal thicknesses that these interpreters imagine would indeed affect the dimensions of the structure, but not nearly so much as the following issues.

How do the framing planks fit together? The Torah tells us how many planks frame the structure on each side: twenty 1.5-cubit planks on the long side (the north and south sides) and six 1.5-cubit planks on the short side (the west side).

^{שמות כו:יח} וְעָשִׂיתָ אֶת הַקְּרָשִׁים לַמִּשְׁכָּן עֶשְׂרִים קֶרֶשׁ לִפְאַת נֶגְבָּה תֵימָנָה... ^כו:כ וּלְצֶלַע הַמִּשְׁכָּן הַשֵּׁנִית לִפְאַת צָפוֹן עֶשְׂרִים קָרֶשׁ.... ^כו:כב וּלְיַרְכְּתֵי הַמִּשְׁכָּן יָמָּה תַּעֲשֶׂה שִׁשָּׁה קְרָשִׁים.

^{Exod 26:18} Of the planks of the Tabernacle, make twenty planks on the south side… ^26:20 and for the other side wall of the Tabernacle, on the north side, twenty planks… ^26:22 And for the rear of the Tabernacle, to the west, make six planks.

The standard assumption is that the planks are placed side by side, flush against each other, in which case one could calculate the area with straightforward multiplication (# of planks x 1.5cu for the dimension of each side). But the text is not explicit about this, and scholars Yohanan Aharoni, Richard Elliott Friedman, and Michael Hayutin have all argued that the planks overlap by a quarter cubit edge on each side, ^[4] cutting the overall dimensions of the mishkan by almost a third.

The joined planks on the longer side of the rectangle, then, have been reconstructed to be 30 cubits long (1.5cu x 20 planks) or 20 cubits long (1cu x 20 planks), and the planks of the shorter side as 9 cubits (1.5cu x 6 planks) or 6 (1cu x 6 planks). But don’t start doing calculations yet – we still have the corner pieces to figure in.

What exactly is a “corner piece”? The two planks described as corner pieces are especially difficult to understand. Exodus simply tells us:

^{שמות כו:כג} וּשְׁנֵי קְרָשִׁים תַּעֲשֶׂה לִמְקֻצְעֹת הַמִּשְׁכָּן בַּיַּרְכָתָיִם׃ ^כו:כד וְיִהְיוּ תֹאֲמִם מִלְּמַטָּה וְיַחְדָּו יִהְיוּ תַמִּים עַל־רֹאשׁוֹ אֶל־הַטַּבַּעַת הָאֶחָת כֵּן יִהְיֶה לִשְׁנֵיהֶם לִשְׁנֵי הַמִּקְצֹעֹת יִהְיוּ׃

^{Exod 26:23} Make two planks for the corners of the Tabernacle at the rear. ^26:24 They shall match at the bottom, and terminate alike at the top inside one ring; thus shall it be with both of them: they shall form the two corners.

But what do these corner pieces look like, and how do they fit together with the rest of the frame?

Option 1: The corner pieces could be just like the planks of the frame – flat pieces with the same dimensions, placed alongside the planks that are considered part of the frame. This Propp’s view.^[5] He imagines that they would add their full width (1.5 cu) to the measurement of the short (rear) wall.

Why wouldn’t such planks be considered part of the initial frame, then? How are they different than the other 6 planks on the back wall? Propp argues that these planks are called “corner pieces” rather than being counted as part of the framing pieces because they are permanently attached to one the 20 planks that frames the longer, north/south sides, but the planks themselves are not fundamentally different, and they add length only on the short (rear) side of the structure.

Option 2: The corner pieces could be just like the planks of the frame, as in option one, but could overlap the width of the north/south frame, meaning that only part of their width would add to the dimensions of the frame. This is the view of the Talmud, and the traditional assumption amongst rabbinic commentators.^[6] Recall that the Talmud imagines that the planks are a full cubit thick to begin with. Here, it imagines that both corner pieces overlap with the full thickness of north/south sides, thereby each adding only ½ cubit in length to the short (rear) side of the structure.

Option 3: A third option is that the corner piece is not a flat plank at all, but rather is a L-shaped combination of two planks, thereby contributing 1.5 cubits to both the long north/south sides and the short rear side of the Tabernacle. This is the argument of Michael Homan.^[7]

Playing only with the variable of how the frame is assembled and how the corner pieces fit in, here are possible sizes for our Mishkan:

Relatively small gaps in information about areas that would hardly be visible to the eyes of a person walking through the Tabernacle have now left us with a structure that could be anywhere from 120 sq cubits to over three times that size.

The Awkwardly Fitting Fabric Overlay

After our troubles with the dimensions of the frame, you’ll be delighted to know that the Torah is quite clear about the measurement of the fabric that goes over its top.

^{שמות כו:א} וְאֶת הַמִּשְׁכָּן תַּעֲשֶׂה עֶשֶׂר יְרִיעֹת... ^כו:ב אֹרֶךְ הַיְרִיעָה הָאַחַת שְׁמֹנֶה וְעֶשְׂרִים בָּאַמָּה וְרֹחַב אַרְבַּע בָּאַמָּה הַיְרִיעָה הָאֶחָת מִדָּה אַחַת לְכָל הַיְרִיעֹת.

^{Exod 26:1} The Tabernacle itself you shall make with ten curtains… ^26:2 The length of each curtain shall be twenty-eight cubits and the width of each curtain four cubits; all the curtains shall be of the same size.

Here we have a rectangle of fabric that is 28 x 40(4x10), to be laid over the top of the structure, presumably covering the short (rear) wall and the long (north/south) walls but leaving the other side open. Unfortunately, since we aren’t sure how big the frame of the Tabernacle should be, we cannot quite picture how it would look to drape fabric of a particular size over its top. Does it reach the ground? If not, how big of a gap is left? The answer, of course, will vary with the dimensions of the structure. But I want to raise a different issue entirely: the shape of the overlay.

If a rectangular cover is laid over a rectangular cuboid, there will always be considerable extra “bulk” at the corners. It takes some spatial intelligence to be able to predict this – or else some experience wrapping gifts or laying formal tablecloths.

In the diagram above, the larger rectangle represents the curtains, underneath which I have “shadowed” the Tabernacle structure. The measurements in the diagram follow roughly the proportions of Propp’s model, 30x12, but the problem I am highlighting will come up using any of the permutations, and will apply equally to the additional layer of goat hair cover that the Torah instructs us to place over the curtains:

As the fabric cover is laid, each side and the back hang down against the walls, but in the corners where there is no wall to cover, the fabric demarcated here within dotted lines will have nowhere to spread out. Covering a rectangular cuboid with a rectangular cover leaves considerable extra material to either bunch up on the ground or billow out behind the back corners.

If we imagine (following both Rashi and Propp) that the fabric does not hang evenly across all 3 sides to begin with—it is imagined to reach the ground on the short side, but hang a cubit or two above the ground on the long sides—this ends up being quite an odd looking divine dwelling place indeed. As Propp has noted, “the whole thing seems esthetically strange, like a poorly fitting garment.”^[8] It seems more likely that the builders would likely have cut out the area between the dotted lines above, creating a fabric cover in a cruciform shape rather than a rectangle.

How can we understand the intent of a text that positions itself as instructions for a building project, attends to considerable detail (e.g. by painstakingly differentiating between the fabrics and metals to be used in different spaces within the structure), but also leaves us unable to decipher how to frame the building, and lays the exquisitely woven fabrics over the frame in such an ill-fitting way?

Another Option: Construction In Meditatio

In my “Repetition and the Tabernacle: Eternity in the Face of Change” (TheTorah 2014), I noted that relaxing our genre-specific expectations and allowing ourselves to experience the repetition in this text the way we might in a poem, for example, opens up a new world of possibility for how we can interact meaningfully with this text. The same approach can offer a way forward when faced with the text’s construction problems.

When this text is read as instructions for a building project, the technical issues outlined above are paramount. But if we are willing to consider the possibility that the text serves to support a ritualized envisioning of this most holy space, or a construction in meditatio and not an actual construction, the problems disappear. A comparison with the Eastern practice of mandala construction offers a model for thinking about the living practices that may have surrounded a text such as ours.

Tabernacle Texts and Mandala Texts

A mandala is commonly understood as a two- or three-dimensional representation of a god’s palace and its grounds, and like the Temple in biblical tradition, it is understood to be a microcosm of the universe.^[9] Also like the biblical God’s dwelling places (mikdash), frequently referred to with the root קדש, the Sanskrit word mandala, writes Elizabeth ten Grotenhuis, “suggests a …sacred center that is marked off, adorned, or set apart.”^[10]

The construction of physical mandalas, a practice that occurs across religious traditions in India and elsewhere in the far east, commonly uses colored sand on a horizontal surface or, occasionally, ink on fabric, though they are sometimes produced in three-dimensional form as well.

If you have had the good fortune to see such an artistic endeavor, you could easily imagine that it is the creation of this stunning piece of art, however temporary (as in the case of colored sand), that is the motivating factor: that the creation of the physical mandala is the end goal. But scholar Richard Kohn explains that in the Tantric notions of ritual that inform mandala construction, there is a sense that there is a different plane of existence—a more mystical one, if you will—where the real work of mandala construction is happening:

Ritual reality is manipulated through visualization, to which the external objects used in the ritual are but props or supports.^[11]

In other words, these physically constructed models are not the most important part of this religious activity. These physical representations are intended to support the contemplative practice of envisioning a god’s palace in one’s mind, piece by piece, as though you were imaginatively constructing it. It is through visualization, not physical construction, that this other ritual plane of existence can be affected: the mandalas constructed in meditation—the ones the practitioners carefully envision—are the heart of the matter; they are the real mandalas.^[12]

There are texts that guide the construction of mandalas, though these are frequently memorized by practitioners and thus may not be referred to in written form during the construction. Of great interest for us, though, mandala texts witness similar patterns of repetition and lacunae, of intricate detail and mismatched points of juncture.

The types of detail that are missing or troubled in mandala texts, like the examples I have highlighted from the Tabernacle texts, are often at points of juncture: each component may be described fairly well, but the instructions tend to gloss over the details of how they come exactly together. Propp understand this feature of the Tabernacle text as a way to keep people from trying to reproduce one.^[13] Buddhist scholars, somewhat similarly, understand the presence of gaps in their mandala construction texts as reflective of the text’s goals: the purpose is not to facilitate the physical construction of three-dimensional structures.

When the end goal is visualization, the precise measurements involved in the integration of two components of a larger whole are simply not that important: a reader picturing all the details of the tabernacle as they unfold would not hold in their mind enough detail for this to matter. Our mind’s eye can smooth over the trouble spots.

Reading and Religious Experience: Text as a Doorway to the Mishkan

The knowledge that there is a type of text within another religious tradition that shares some of the Tabernacle text’s most unusual features—another text that describes the dwelling place of the divine, no less—helps us understand what it is we are supposed to do with the Tabernacle texts. Though couched as instructions, the Tabernacle texts are not meant to instruct us in how to actually build such a structure; practically speaking, we readers don’t need instructions. What we do need, though, is some way to experience this most holy space.

Both now and in the past, the vast majority of people with access to this text and the ability to read it would not have had have access to the inside of the Tabernacle (or the later Temple), even if/when those structures stood: holy spaces were restricted to the personnel who served within them.

To put a finer point on it, readers generally don’t need building instructions, but do want for access, imaginative or otherwise, to this most holy space. This text may serve as a guide for envisioning the space that is so intimately connected to God, offering its readers some contemplative experience of it?

For this purpose, the impact of the corner pieces on the length and width of the frame seems less important than the cherubim woven into the curtains, or the careful attention to the difference between pure gold and gold in the different holy spaces. It may be that these are poor building instructions, but the text suits precisely the needs we actually have.