How we read the architect's signature

The architect of the 1899 Rochussenstraat 169 permit drawings is Nicolaas Tinbergen; that is confirmed¹. The architect's own deep dive explains who he was and how he came to be both the architect on the elevation and the head of the builder's firm named on the application. This note deals with method. Read the signature on those 1899 drawings the easy way and it can give you the wrong answer.

What the signature does when you read it alone

At the foot of drawing Nº II is a cursive signature, with the word architect written beneath it. Lifted out of the sheet and read alone, it does not plainly say "Tinbergen". The sheet is a faded cyanotype: pale lines on a dark blue ground, in a quick and heavily flourished hand. Read in isolation, the ornate capital can pass for almost any letter, and it has. An early reading, since refuted, took the name to be "Molenaar", and an attribution was built on it before anyone found that the attribution was circular². Out of context, the opening capital reads as readily as an M or a W as the letter it really is. Settle the name that way and you mis-attribute the whole building.

Why the easy reading fails, and what rescues it

The isolated reading compares the wrong things. A signature has no single fixed shape, and Tinbergen, like many men of his trade, signed a formal document one way and a working drawing another. In his 1877 marriage act his name stands in full, careful, deliberate script, every letter articulated³. On his project drawings he wrote a quick, abbreviated autograph. There the surname's capital is an ornate looped T whose long horizontal crossbar sweeps flat across the whole word. Degraded on a faded blueprint, the capital's strokes beneath that crossbar are exactly what the eye takes for an M or a W. Beside the careful 1877 hand, the rapid 1899 flourish can look like a different person's writing altogether. They are simply the two ways one man signed, and reading the project signature against the formal one is the trap. The plate below sets the 1899 signature above the 1877 marriage autographs so the mismatch is plain. It shows the comparison to avoid, not the one the reading rests on:

The right test compares like with like: the 1899 project signature against another of his project signatures, not against his formal marriage hand. Two Rotterdam permits of 1902 carry his autograph, and the same writing runs through the body of every office sheet in that small group⁴. One of them, a drawing for verandas at the Gesticht Bethesda, is signed N. Tinbergen at the foot in that same quick project hand⁵:

Put the 1899 corner signature next to that 1902 permit signature. The crossbar-capital, the pinched medial b, the compressed -ber- run and the trailing flourish line up as one hand⁶. The plate below lays the two project signatures side by side for comparison. It is a working figure assembled from the primary scans, and its authority lies in the scans it crops, not in the montage:

What the reading rests on, and where it stops

The identification of the architect does not rest on the 1899 flourish at all, which is why the method is worth writing down. It rests on several records that agree and never needed the signature. The city archive's own catalogue attributes a run of 1902–03 permits in the same series to N. Tinbergen⁷. An 1892 Rotterdam birth act enters him as an architect at 1e Lombardstraat 30⁸. And the 1899 address book prints the builder's firm and the architect at that one door in the year of the corner's permit⁹. The signature corroborates what they establish; it is not the foundation.

The limit is the 1899 signature itself. It is degraded and cannot be read letter by letter in isolation, so any single-glyph reading of it is a finding aid, never the settled value. What the record holds is the agreement of those sources and the project-to-project match of the hand. What it rules out is the early misreading: the architect is not "Molenaar", a name built on a circular reading and long since set aside¹⁰.