Marillion: The Doug Irvine Era

Author and musician Nathaniel Webb wrote the book Marillion In The 1980s. The format and purpose of the book didn't leave enough space for his focused and extensive research on the recordings the band made in their earliest years, before Diz Minnit and Fish joined. This chapter had turned into an essay on its own. If titles Alice, Lady Fantasy, The Haunting Of Gill House, Herne The Hunter, Scott's Porridge, or Close mean anything to you, this is the story behind those.

DPRP.net is proud to be able to publish and share this essay with you now. Read about The Doug Irvine Era!

This essay has seen several revisions before being published here, September 2021. New insights, new people spoken to, new evidence popping up. There will no doubt be revisions to this in the future as well!

Nathaniel Webb

Author's Note

I wrote this article as part of the research process for my book Marillion In The 1980s, from Sonicbond Publishing. The book follows the band from the founding of Silmarillion to the arrival of Steve Hogarth and the release of Seasons End, and includes track-by-track analysis of Marillion's five studio albums and many singles from the 1980s.

While learning about the Doug Irvine days of 1979 and 1980, I was curious to reconstruct the band's songwriting process. I read everything I could find and sought out copies of the many bootlegs of that period. (Thanks are due here to Mark Abbott and Peter Goodfield of Skyline Drifter, Andre Kreutzmann of Marillion Setlists 1980-1988, Claus Nygaard, and Jon Collins, author of Marillion: Separated Out... Redux.)

What I learned surprised me: there was no agreement whatsoever about when, where, and what exactly Irvine, Mick Pointer, Steve Rothery, and Brian Jelleyman recorded. Moreover, the bootlegs often made obvious errors in how they credited the songs they contained.

There's no room in a 65,000-word manuscript – intended to be friendly to casual readers, no less – for a painstaking discussion of six songs from before the band even had a real lead singer. But the problem still bothered me: when I state in Marillion in the 1980s that a certain song was recorded at a certain time and place, I want to be right! Barring that, I want to have a solid foundation for whatever educated guess I put down—especially since some claims I stake are counter to decades of received wisdom among Marillion fans and biographers.

Luckily, you don't have to take my word for everything. I've assembled a few short audio clips that demonstrate some claims I make below.

The following audio clip gives audio comparisons of Alice from two different lineages, showing that in some cases, recordings said to be from different sessions are actually the same tape at different speeds.

(If you cannot play this inside this page, you can follow this link.)

The next file demonstrates my claim that Marillion did not, as commonly stated, record a new version of the song Close in November 1980 (see Sessions, below).

(If you cannot play this inside this page, you can follow this link.)

These files, a record certification chart, and notes and errata for the book Marillion In The 1980s are all freely available here.

Please share this link as you please. I hope you enjoy the essay and audio, and if you think I'm wrong about anything in here, I would love to hear why! I also hope you will consider purchasing a copy of Marillion in the 1980s.

Nathaniel Webb
Portland, Maine, USA
August 2021

For the book, check out Burning Shed, Amazon UK, Amazon US, Amazon Sweden, Amazon Germany. Read the review on DPRP.net here.


The issue of Marillion's demos before new bass player Diz Minnitt and singer Fish arrived early January 1981 is a thorny one. We are attempting to reconstruct a recording history that began over 40 years ago, in the summer of 1979, with the arrival of guitarist Steve Rothery at Doug Irvine and Mick Pointer's band house in Long Marston. Marillion's wonderfully punctilious gigging records didn't really get going until Diz and Fish joined up—the Irvine days are much hazier.

We are also untangling decades of bootlegger assumptions, perhaps even lies. From the very beginning, Marillion was a band with fervent fans, eager to get their ears around anything its members had touched. This jumpstarted a cottage industry of bootleggers that persists to this day – but in the 20th century, there was no metadata tagging of FLAC files, no Google to consult. If a fan bought a copy of Forgotten Songs or Haunters Having Lots Of Fun from the back of a van, they simply trusted the tracks were in fact what the label claimed.

Meanwhile, the band's recording sessions from this period have come down to us as little more than an oral history. Marillion biographies tend to reference each other and make claims or assumptions without citing evidence. This makes a true history difficult to determine.

Finally, there are multiple lineages of bootlegs. The same original tapes seem to have been copied multiple times very early on, with slight variations, then passed around without regard or reference to other bootlegs. This means the same track can sound quite different, depending on where you find it.

All this has led to a tangled web of copies, citations, and confusion about what was recorded, where, and when. This article attempts to clarify this mess as best as is humanly possible, using a combination of historiography and deep listening to categorize and date the many extant recordings.

Silmarillion At Amersham

A brief detour, given in the interest of completeness, before we begin in earnest. In his manuscript In Shades Of Green Through Shades Of Blue, Claus Nygaard gives a terrific quotation from Mick Pointer about the early Silmarillion rehearsals when Neil Cockle and Martin Jenner were still part of the band: “Where we used to play was in Amersham, it was a rehearsal studio in Amersham. All the Silmarillion stuff was done in Amersham, and the tape of Silmarillion was just done on a tape-machine in the corner of the studio. I was a great fan of Camel at that time, and Doug [Irvine] was a great fan of Camel, and they were rehearsing next door to us in this rehearsal studio, and they were watching us one day playing what we thought was similar sort of music as Camel. And [Camel drummer] Andy Ward went on and took my place.”

First band T-shirt, designed by Darrell Foster based on a draft by Mick and Doug. First band T-shirt, designed by Darrell Foster based on a draft by Mick and Doug.

While not the focus of the story, this does raise the question of what became of that Silmarillion rehearsal tape. A record of the band before Steve Rothery joined would be remarkable indeed! It would also have musicological value: in a 2019 interview with Mark McCormac, Pointer maintained that a number of beloved Marillion tracks had their genesis in Silmarillion's long instrumentals. Mentioning Grendel, He Knows You Know, The Web, and Garden Party, he said, “You listen to the drum parts – they came first.”

Luckily, Pointer also gives us the answer, though it is bittersweet indeed: “I also had a tape of all the Silmarillion music, and about two days before Doug left the band, he said: 'Oh, have you got that tape, 'cause I want to listen to it.' And I handed him the only tape of Silmarillion music. I remember giving it to him. If only I had that tape. A lot of the early stuff from that period found its way onto Script For A Jester's Tear. Doug left I think it was around November 1980, I remember handing him this Silmarillion tape. I never saw him again after that day.”

In fact Mick and Doug did eventually reconnect – Mick invited Doug to an Arena show in 2018 – but there was no discussion of the lost Silmarillion demos.


Moving from Silmarillion to Marillion, we begin with a list: what songs, exactly, are we talking about? Luckily, history and the fandom agree on this point.

There are seven songs at play.

1. Alice

This was the first song to get vocals from Doug Irvine, a few lines based on Alice In Wonderland tucked inside its lengthy instrumental. It became Snow Angel under Fish, then ultimately got recycled into the long ending vamp of Forgotten Sons. The track is called You'll Never Return on a few bootlegs, from a lyric near the end.

2. Close

Another song featuring Irvine's singing, Close became The Web immediately upon Fish's arrival. The arrangement and instrumentation stayed mostly unchanged up to its appearance on Script For A Jester's Tear. The song is referred to as Anyway in at least one place, a reference to its opening lyrics.

3. The Haunting Of Gill House

Under Fish, this gothic instrumental became a tune called Skyline Drifter, but it was ultimately scrapped – with the exception of a keyboard riff that became the intro to Garden Party. (It was also briefly called Silver Drifter, as well as a scratched-out title on Fish's original lyric sheet, Machine Hawks.) Live, it was an excuse for sound engineer Christopher “Privet” Hedge to set off pyrotechnics.

The song is called The Haunting Of Gill House on most bootlegs. Fish agrees: his Skyline Drifter lyric sheet notes that his words are set to “Marillion's Gill House,” and he refers to it as The Haunting Of Gill House in his 1997 liner notes to Script For A Jester's Tear (though he calls it an early version of Herne The Hunter, which is wrong).

Nevertheless, the tune occasionally bears other different names. In Mick Wall's authorized 1987 biography Market Square Heroes, Steve Rothery calls it The Haunting Of Your House. This is probably a simple error, but the track's spoken-word intro does include the phrase “your house is burned to the ground,” and at the end, “But still, our very song gives rise to the legend of your house.” This last line leads into the real start of the song, burdening it with importance.

Adding to the confusion, Diz refers to the song as The Haunting Of Hill House in a 2013 interview with a Facebook fan group. It's also given that name on a bootleg or two. Most likely, Diz and the bootleggers are confusing the song with the classic 1959 horror novel of that name. (Written by Shirley Jackson, best known for her short story The Lottery which has upset generations of high school English classes in America, the book was made famous by its 1963 film adaptation The Haunting.) But it's also plausible that the song, with its gothic tones, was inspired by the novel (though the titular Hill House doesn't burn to the ground). If so, any reasoning behind the name change from “Hill” to “Gill” has been lost to time.

4. Herne The Hunter

Another instrumental, which gained lyrics from Fish but didn't survive long in the band's live set. Nor did it provide any parts to later songs, though it does share some similarities with Grendel and Three Boats Down From The Candy.

In Market Square Heroes, Rothery calls it Hare And The Hunter and refers to it as the band's attempt at a "folk legend". Herne the Hunter is a figure in English folklore, a ghost associated with Windsor Forest and Great Park in Berkshire county – and the horse-riding figure on Marillion's very first T-shirts, self-produced in 1980, is commonly understood to be Herne. So it's likely Mick Wall just misheard the title when interviewing Rothery. Fish's original lyric sheet agrees: it's titled Herne The Hunter and has quite mythical lyrics.

5. Lady Fantasy

A very 1970s track, with a very seventies vocal from Irvine. It became Madcap's Embrace under Fish but was ultimately abandoned altogether. On one bootleg it's titled Lady Ramsay.

6. Scott's Porridge

Occasionally referred to as Scotch Porridge, presumably because it's a jam on Scotland The Brave, this is a snippet of instrumental improvisation. It would soon be refined into the barn-burning encore Margaret during the Fish days. It was also briefly called Margaret Gets Her Oats – in both cases the title is a reference to fledgling Marillion's green Commer van.

7. The Tower

This lost epic was recycled into Grendel upon Fish's arrival, but if it was ever put to tape, that recording never leaked and is assumed long lost. Some descriptions put it at twenty-one minutes long, though apparently without any source. Fish describes it as eight minutes in his 1997 Script liner notes. As he's consistently said that sections were added to The Tower to make Grendel, it's reasonable to suppose it was shorter than Grendel, which is about seventeen minutes.

Live at Civic Centre, Birkhamsted. Photographer unknown but photo credited to Steve Rothery.


The next issue to tackle is how many performances of each song were actually put down. It's at this point that bootleggers start muddying the waters. Different bootleg lineages (see the Lineages section below) get confused with different performances, even assigned to different sessions (see Sessions below), when in fact they're just different copies of the same recording. Luckily there's not much material overall, which allows us to easily analyze the canon.


There are three performances of Alice.

Two are very similar. They feature identical arrangements, and are often called the “instrumental” and “vocal” versions. I will refer to them as Alice (A) and Alice (B) respectively. They can quickly be distinguished by the drumming in the first 20 seconds. In (A), Pointer doesn't play any fills until the transition to the guitar lead, which he flubs pretty badly. In (B), he plays fills throughout the beginning. The endings are also quite different. (B) goes into an extended wrap-up, lots of noodling and flourishes that are reminiscent of the end of a live set. Irvine even says something like “Thank you, goodbye” just before the end!

Both Alice (A) and (B) actually do feature vocals from Doug Irvine, beginning at about 3:20 in A and about 3:30 in B. In both cases, the vocals are muddy and quiet, making them hard to discern. They're nearly inaudible in (A), probably leading to its common mis-designation as an instrumental.

Alice (C) is easily distinguished from the other two. The recording quality is far superior, as is the performance, and it begins immediately with a synthesizer lead over the opening chords. This version also includes vocals, with a stronger performance from Irvine.


Conventional wisdom holds that Marillion recorded Close twice, once with Irvine and once without. Certainly some version of the song was sent to Fish and Diz prior to their audition, and Fish overdubbed his vocals promptly upon joining, making it into a demo of The Web. Contrary to popular belief, however, I don't think Marillion ever rerecorded Close in its entirety. Rather, they seem only to have stripped out Irvine's vocal and improved a few spots while keeping the rhythm track in place. This is discussed at length below.

The Haunting Of Gill House, Herne The Hunter, Lady Fantasy, Scott's Porridge

Only one recording of each of these songs survives. Some sources claim that there is an instrumental version of Lady Fantasy, but if so, I've never heard it.

The Tower

No known recording of The Tower exists.

Brian Jelleyman, Mick Pointer, Doug Irvine, Steve Rothery, and an unknown person. The trees and coats suggest this is Autumn so might be just before Doug left. Credits for this photo differ but are often given to either Steve Rothery and Stef Jeffrey Depolla.


Now we enter disputed territory. In my research, I've come upon five supposed recording dates on which these demos were produced. Some are more likely than others, but there are countless ways to read the evidence.

Nonetheless, it's reasonable to start by splitting the demos into two groups: those with low production values and those with high. In the first category we can put Alice (A) Alice (B), The Haunting Of Gill House, Herne The Hunter, and Scott's Porridge. In the second go Alice (C), Lady Fantasy, and Close.

The latter three have much more in common with each other than their cousins. The recording quality is better, brighter and clearer. The playing is tighter and more polished as well – and in the case of Alice (C), the band has mastered the song, which they flub in versions (A) and (B). These are also the three tracks featuring clear, solid vocals from Doug Irvine. They must have been recorded in a multi-track studio where Irvine could overdub his singing.

It's safe to say the low-production demos were recorded earlier than the pro-sounding group. Beyond that, we might split our two categories into three, or even more, individual recording sessions. Most bootlegs place Close alongside the other two high-production tracks (often in the “summer sessions” in June) but as I'll explain at length later, I'm confident it was recorded on its own in late November. This sets Alice (C) and Lady Fantasy apart, which agrees with most of the written history.

A thornier question is whether the low-production demos were recorded all at once or at different times. It seems likely Scott's Porridge was recorded together with Alice. (Andre Kreutzmann first put this to me, and I've warmed to the idea.) On many digital bootlegs, Alice (B) includes the first 15 or 20 seconds of Scott's Porridge at the end of the track. The performances themselves don't exactly run together, but neither is there empty silence between them. In the brief pause between the end of Alice and the start of Scott's Porridge there's a sort of whistling tone that sounds like room resonance and can also be heard as Alice is wrapping up. So it's possible the band went straight from Alice (B) to Scott's Porridge when making this tape. There's a certain logic to this: Alice (B) has that big live-style ending, and Scott's Porridge evolved into the encore Margaret, so perhaps Marillion was practicing the end of their setlist.

It's also possible Alice (B) was just digitized sloppily from a cassette that happened to have Scott's Porridge next. In that case, the feedback that runs over from Alice is either a tape artifact or simply coincidence. To my ear, Scott's Porridge does sound a bit more competent than Alice (A) (B), suggesting a more practiced group recording at a later date. But this is a subjective judgment, and given the arrangement differences between (A) and (B), it's possible the group was still working out Alice and this accounts for any sloppiness.

I could see Herne The Hunter and The Haunting Of Gill House being recorded a bit later. Alice (A) and (B) are sloppy, with mistakes from both Pointer and Rothery. They also lack the intro synth that Jelleyman plays on Alice (C), but he plays confidently on Herne and Gill. One possible explanation is that these tracks are rehearsal-room recordings, made on different days. Another is that they were recorded together at The Lodge, but Alice was a new song that still needed practice. It's less likely they would be recorded at The Lodge in separate sessions, as that would have been a poor use of time and money.

The final question is when exactly each group of demos was recorded. Putting precise dates on the sessions is impossible, but there are two basic timelines that could fit the evidence. Which you prefer depends on how you read a particular clue.

Marillion - A History, EMI band info, probably Autumn 1982

The document titled MARILLION - A HISTORY, which I understand to be from the band's early press kit, states, “Their first step was to record a two-track demo tape for their own use at the Enid's studio in Hertford and, encouraged by the results, they played their first gig at Berkhamstead Civic Centre on 1st March 1980.” This places Marillion's first session at The Lodge prior to March, but it contains a word with multiple meanings: “two-track.”

One obvious reading of “two-track demo tape” is that the demo contained two songs: Alice and Lady Fantasy. Less obvious, but also possible when discussing demos made long before the advent of digital recording, is that “two-track” refers to the recording process – namely that the songs were recorded directly to a tape machine with two inputs. These two inputs would write to the two magnetic tracks on the tape, corresponding to the left and right stereo channels. This would enable the band to document their performance with better fidelity than a live bootleg or a consumer tape deck set up in the rehearsal room – but could not approach the clarity of a multi-track recording, where each instrument is recorded and mixed separately. In other words, two-track recording would produce something roughly like the low-production demos that have survived.

Timeline 1: Everything Except Close Recorded Before March 1980

This timeline is generally accepted by Marillion biographers, including Jon Collins and Claus Nygaard. The dates shift around, but the story goes that Marillion went to The Lodge in late 1979 or early 1980 to record two songs, Alice and Lady Fantasy, as a sort of test. Pleased with the results, they used the demo to book gigs around Aylesbury, starting with their March 1st debut.

If you read “two-track” to mean two songs, this timeline seems probable. There's a lot to like in those songs, and one can easily imagine booking agents responding eagerly to a band capable of such professional recordings. In this case, the earlier, rougher demos weren't made at The Lodge at all, but rather recorded by the band during practice sessions.

There is some sonic evidence for this interpretation. The ending of Alice (B), with its live-style outro, suggests a practice tape rather than a paid studio recording. The surviving snippet of Scott's Porridge, which began life as a rehearsal-room jam, fits the rehearsal vibe as well.

If this timeline is correct, the so-called “spring” and “summer” demos of bootleg tradition are both absolute phantoms, red herrings, invented by early bootleggers for reasons unknown. The other possible analysis, laid out in Timeline 2 below, was guided in part by an assumption the bootleg dates are based on something, even if the particulars are wrong – but it's possible they're simply nonsense.

Timeline 2: High-Production Demos Recorded in March or Later

This is the timeline I give in my book Marillion in the 1980s, reflecting my best guess based on the evidence I had at the time. It relies on reading “two-track” to refer to the recording process, as well as giving some credence to the dates found on countless bootlegs, which refer to a “Summer Session” in June 1980. In this interpretation, the low-production demos are a result of an initial trip to The Lodge, where Marillion put down rough versions of the songs in question, probably playing live in a room with two microphones set up. This was the demo that encouraged them to seek gigs. Then, after a few months of gigging, they returned to The Lodge to make professional demos of Lady Fantasy (which had perhaps not been written by March) and Alice.

This timeline mainly hinges on the question of what Marillion was capable of producing on their own. If the low-production demos are merely rehearsal-room recordings, made for reference during practice sessions, then the band had no need to visit The Lodge until they were ready for multi-track. In that case they would have spent only one session there – before March, as described above.

But despite their flaws, to my ear those rough songs sound better than a consumer tape deck in the corner of the practice room could produce. In those situations, drums tend to dominate, followed by amped instruments like guitars and keyboards. The rough demos are better balanced than that, and Irvine's vocals in both takes of Alice are audible (though muddy and faint), an achievement in an amateur recording environment. Perhaps Marillion had access to better equipment, or maybe they returned to Amersham or somewhere like it, but I've seen no reference to either option.

I'll also note from personal experience that with a decent demo in hand, getting entry-level gigs is not so difficult for a competent band. Certainly my own group had no trouble finding shows, even in our earliest days. While I can't claim to know the scene outside Aylesbury in 1980, I suspect it wouldn't have taken Marillion long to book gigs on the strength of even their roughest demos. I'll certainly allow that the high-production demos would be better for booking shows, but I don't think the earlier tracks are automatically disqualified. Too, we should note the quality of the gigs Marillion played in 1980 – the notorious street fair and retirement home shows certainly don't suggest discriminating booking agents.

With all that said, let's take a look at the five dates that have been attached to Marillion's demos at one point or another, and see what truth they might contain.

Session: October 1979

This date is given by a couple sources, most notably Claus Nygaard's manuscript In Shades Of Green Through Shades Of Blue. As he tells it, the band needed some sort of demo in order to begin finding gigs. Supposedly, this session featured instrumental versions of Alice and Lady Fantasy, but as noted above I'm skeptical that an instrumental Lady Fantasy actually exists. In my own conversations with Nygaard he's agreed this supposed date is suspect.

So where did it come from? October 1979 is given for Alice and Lady Fantasy on numerous bootlegs, but the bootleggers are likely steering us wrong. On these discs, Alice is Alice (C), and Lady Fantasy is the same as elsewhere. The songs just sound different from other tracks of the same name because they come from a different tape lineage (see Lineages below).

There is, if you squint, an almost-primary source for this date: The Web #1, the first issue of Marillion's fanzine. Published in February 1982, it includes a band history, which states (with typos and punctuation intact):

Marillion were formed in September 1979 by Doug Irvine (Bass/Vocals) and Mick Pointer (Drums) out of the ashes of Silmarillion. They recruited Steve Rothery (Guitars) from Whitby and Brian Jelliman (keyboards) from Aylesbury. The band recorded their first demos at this time, at the Enids studio's in Hertford, the tracks were lady Fantasy and a song called Alice. They practiced and rehearsed for six months and the bands first gig was on Saturday 1st March 1980 at Berkhamsted Civic Centre.

This does seem to be saying that the band was formed in September and promptly recorded at The Lodge, which could suggest an October recording date. But the many other errors (Rothery joined in August, Jelleyman probably joined in October) suggest this document shouldn't be taken too seriously. Certainly there's no chance the polished demos were recorded this early, since Jelleyman played on the earlier recordings.

As noted earlier, I think it's possible Alice (A) and (B) and Scott's Porridge were put to tape before Herne and Gill House. If so, perhaps it really was this early on. In that case, Jelleyman's intro synth is missing from Alice because it hadn't been written yet!

Session: February 1980

This date appears in only one source, the biography Marillion: Separated Out... Redux by Jon Collins. His recounting of the early days matches closely with Nygaard's manuscript and The Web: he has Alice and Lady Fantasy recorded at The Lodge, notes the session was prior to the band's first gig, and doesn't mention possible March and June sessions at all. But only Collins places the session in February. Nevertheless, it may be the date best supported by the evidence.

I was initially skeptical, but I had the good fortune to ask Collins directly how he'd come up with February. At first he was unsure, but he very generously dug out his notes from the first edition of Separated Out in 2002. They are somewhat sporadic in nature, but include two clues. Collins specifically asked Brian Jelleyman whether the demo was recorded before or after Marillion's debut gig at Berkhamsted Civic Centre. His notes on Jelleyman's reply read, “First gig! We'd done the demo!” Meanwhile, in Collins's interview notes from Steve Rothery, we have, “Doug + Mick big fans of the Enid, they fancied working there! [Q]uite exciting - first experience of recording.”

So while the month of February goes unmentioned, it seems irrefutable the first demo was recorded at The Lodge sometime before March 1st, 1980. Jelleyman confirms the demo came before any gigging, and Rothery states the Enid's studio was his first recording experience. (There's no mention in these notes of which songs were recorded, so they're no help in choosing between timelines.)

Collins and I have theorized that perhaps something in the phrasing of one of his interview subjects suggested February as a recording date: “just before the first gig”, “a few weeks after we did the demos”, anything Collins might not have written down explicitly, but placed the first Lodge session in the month prior to Marillion's live debut. In favor of February, then, we have Collins' general attention to detail and determination to get the facts right.

There was also a photo posted recently to a Marillion fan group, of a cassette bootleg from the eighties. It contains the three high-quality demos alongside some early Fish recordings. Tantalizingly, it is labeled “1-2-80.” To my American eye, this recalls (if incorrectly) Fish and Diz's January 2nd, 1981 audition and recording session. But in the European style, the date is February 1st, 1980.

Session: March 18th, 1980

There's general bootleg agreement that what are called the “Spring Demos” were recorded in March 1980. I clung to this date for some time, then decided it was nothing more than a particularly tenacious mistake. After all, at least one session at The Lodge must have taken place prior to March 1st, and the leap in skill from the rough demos to the polished ones is so great that months of rehearsing probably passed between them. So unless Marillion made rough demos at The Lodge very early on, then returned in March after only a gig or two to record their high-quality tracks, this date had to be bogus.

At least that was where I left it until December 2020, when consummate Marillion collector Mark McCormac posted a new item to his Facebook page: Marillion's invoice from recording at The Lodge, dated 18 March 1980! The transcription reads:

18 March 1980
From The Lodge Studios
To Marillion
Time 100 -
Tape 12 50
Sandwiches [?] 1 25
6 Meals 6 00
Total 119 75
(Signature unreadable)

There are a couple fun mysteries embedded in this receipt (does it really say “sandwiches?” Who ate the two extra meals?) but the biggest question is the date. My first thought was that the band recorded in February but didn't pay until March. Certainly that would be a welcome arrangement for a group of broke musicians. But from the perspective of The Lodge, waiting a month to try and get paid sounds like magical thinking. One would think they would at least write up the invoice on the last day of recording, even if they gave Marillion some time to pay it.

Too, as I noted in Marillion In The 1980s, the name change from Silmarillion to Marillion occurred later than biographers have generally thought. As late as March 1st, 1980, Melody Maker ran a gig listing for a misspelled “Silmerillion” on March 8. If the name change didn't occur until the end of February, then an invoice addressed to the new name must have been written later.

Whatever the import of the invoice, it's the first piece of hard evidence to support the bootleg tradition of placing a recording session in March. Prior to the document resurfacing, all I could offer is that March 1980 marked Marillion's arrival on the scene via their first gigs, so perhaps it became associated with the birth of the band in the minds of early fans. It's an open question whether those bootleggers have known something we didn't all along.

Session: June 6th, 1980

As is the case with the others, there's no real source for the date given for the so-called “Summer Demos” except the long tradition of what's written on bootlegs. Still, the timeline is at least plausible. If the band made their first, rough demos on the cheap at The Lodge in early 1980, they could have returned that summer to record something more professional on a multi-track system.

Placing Alice (C), Lady Fantasy, and Close in June agrees with widespread bootlegs such as Forgotten Songs. But there's a larger question around Close, which we'll now attack.

Session: November 1980

There's little dispute about the last recording session Marillion undertook before Fish and Diz arrived. The group had booked time at Leyland Hill Farm Studio in Gawcott, intending to record another version of Close. Then Doug Irvine suddenly announced his departure from the band and retirement from music. Rothery, Pointer, and Jelleyman had no intention of giving up, so they showed up at Leyland Hill Farm and put down an instrumental of Close with Rothery playing Irvine's bass part as well as his own guitars. This tape was sent to Fish and Diz in December of 1980, and the next month, overdubbed by Fish with his lyrics for The Web.

Or at least that's how the story goes...

But I don't think the story is correct. The broad outline likely is – it's been repeated enough times, by enough people, and it makes sense in the history of the band. As for the details, however, the tapes tell a different tale.

Attentive listening to Close and The Web – the version Fish recorded during his audition at Leyland Hill Farm on January 2nd, 1981 – shows they are in fact based on the same original track. The drum performance is the same. Contrary to legend, so is the bass.

But there are differences, too (in addition to the obvious change of vocalist). Keyboards have been added to the Fish version, most notably under the staccato stabs of the intro. Rothery's long guitar solo (about 3:53 to 5:21) was rerecorded as well, with no wah pedal, a richer tone, and more dynamic lines. (It's also worth noting that by the recording of Script For A Jester's Tear, the intro of The Web had shed the staccato delay guitar, which was suspiciously close to Rush's “2112.” The guitar solo is quite close to Rothery's second demo take, though!)

So what's the truth here? It's hard to know for sure. The basics seem incontrovertible: there was a session booked at Leyland Hill Farm in November, Irvine quit unexpectedly, the others kept the date and worked on Close.

But the order of operations is debatable. Let's start with the source for all these claims, the MARILLION - A HISTORY document. Here is its exact wording:

Marillion clocked up fourteen dates to November 1980. It was at this point that Doug decided to leave the band... Doug's departure coincided with the recording of a track at Leyland Studio, near Buckingham, which was destined to become 'The Web'. Both guitar and bass tracks, therefore, were recorded by Steve. Meanwhile, the ad was answered by, amongst others, two guys – a vocalist and a bass player – who were living in Scotland. The instrumental track was sent up to them to give them an indication of the sort of music Marillion were playing. They came down to an audition on 2nd January (confident enough to have brought their luggage with them) with lyrics already composed for the song! So, it was back to the studio to try it out and Fish and Diz Minnitt joined the merry throng.

Note the phrasing there: “Doug's departure coincided with the recording of a track.” It doesn't say he left before. Indeed, it could easily be read to mean he quit during the time at Leyland Hill Farm. Too, Andre Kreutzmann's website gives the date of Irvine's last gig with Marillion as November 14th. The famous ad seeking a bassist/vocalist appeared in the December 6th issue of Musicians Only, and Stef Jeffries' letter to Fish and Diz is dated December 10th. So Irvine must have quit in late November.

Hand-out at Doug's last gig, 14 November 1980

The ad that Fish and Diz responded too.

Therefore, I propose that the first and only recording of Close comes from this session. Irvine laid down vocals, but his bass playing either never made it to tape or got replaced by Rothery's after he quit.

There is further evidence in favor of this theory. A fragment of Stef's letter to Fish and Diz, which accompanied the tape including “Close,” has survived as a photo plate in Mick Wall's 1987 authorized biography Market Square Heroes. The complete transcription is as follows:

[Page 1]
Waddesdon 8104
53 Quainton Road

10 December 1980

Dear Derek + Diz (I think!?)
I'm writing this because nobody in the band can write – at least that's their excuse. When you hear this you'll understand why they've placed so much emphasis on the vocals!

We're hoping to lay down a tape with the present set on it to get to you before you come down. But this may not be possible as the old bassist has now refused to play through it with them and the tape made at the last gig did not come out (Basically because someone who shall remain nameless forgot to press the 'record' button!)

Anyway I hope this gives you a rough idea of the sort of thing we do and doesn't...

[...Page 2...]

[Page 3] ...'Songiest' ...ber we do. It was recorde... Enid's place at the be... year. Doug (old... bass on... has the... Som... Scrawl [or scrawny?]… it to...

Steff's letter to Fish and Diz.

Not much to go on, but aside from Stef's wonderful sense of humour it implies a couple interesting things. First, it suggests the band were still in touch with Irvine as of December 10th. This may be poetic license on Stef's part, as interviews have suggested that Irvine simply disappeared from Marillion's lives after he quit. But she certainly seems to suggest they were still hoping to get him back into the studio. Jumping back to MARILLION - A HISTORY for a moment, note this phrasing: “Both guitar and bass tracks, therefore, were recorded by Steve. Meanwhile, the ad was answered by, amongst others, two guys...” It might just be sloppy writing, but that “meanwhile” technically means the famed call from Diz came while the band was still working at Leyland Hill Farm.

Second is the question of how the tape suggests an “emphasis on the vocals.” We know from both Diz and Fish that the tape also included Lady Fantasy, a good candidate for “the songiest number we do” that “was recorded at the Enid's place at the beginning of the year.” But it seems likely that Close wasn't an instrumental, but the full song, with Irvine's singing and all. At first I thought the band must have just bounced down a vocal-less mix for Fish and Diz. But I noticed this in Fish's Script For A Jester's Tear liner notes: on January 2nd, “Diz got through his run-through while I recorded vocals for the first time singing my newly written lyrics entitled The Web over an existing band composition written and performed with Doug Irvine singing a few scant verses of lyric. The song was called Close and sounded to my ears like Camel. I'd received a tape in Scotland and worked one of my existing lyrics around it.”

This seems to suggest that Fish had heard Irvine's vocals on the tape he received. And fascinatingly, the lyrics to Close and The Web are light years apart in all but one aspect: they're both about the lonely aftermath of a breakup. Irvine's were fairly bland (“Any way, anyhow, all I need is you / if I told you in the night my loving is true / now you're gone, I'm so alone, well I have nothing to do / spend my time to remember, when I was close to you”) and Fish's are prolix and arcane, but in both cases a narrator sits alone pining for a lost love.

Isn't it just possible that Fish chose his lyric inspired by Irvine's?

There's another little piece of evidence in Clive Gifford's 1987 biography Marillion: The Script. Gifford tells us “Marillion sent a tape of their own containing versions of Close and Lady Fantasy. [Fish and Diz] were impressed by the sound but felt the vocals were weak.” He continues, about the January 2nd audition, “On the first day the instrumental version of Close was taken and overdubbed with a new guitar solo from Steve. Lyrics written and sung by Fish were added and the whole new package was renamed The Web.”

Here we have two things. First, a suggestion that perhaps the Close Fish and Diz heard did have vocals. Second, confirmation the instrumentation changes between Close and The Web were done concurrently with Fish's new vocals.

Jumping back to The Web #1 – and recognizing that last time I mentioned it, it was to call it unreliable – we have the fact its band history puts Alice and Lady Fantasy together, with no mention of Close. Perhaps this is, despite the zine's other inaccuracies, a reflection of the fact Close was indeed recorded separately.

Finally (I promise!) there's the fact the band went back to Leyland Hill Farm for the January 1981 audition/recording session, plus later rehearsals once Fish and Diz had joined. While there, they were able to do multiple track replacements on Close within the space of a day or so. This suggests the version of the song they were working from – which we know to be the version Irvine sang on – was also recorded at Leyland Hill Farm. It would have been far easier, cheaper, and more logical to remove and add tracks on tapes that had been originally recorded, and still resided, at Leyland Hill, than it would to scrap a Leyland Hill instrumental, take an earlier version of Close from The Lodge, and strip out the vocals and guitar solo.

So what exactly am I suggesting? Only this: that there's just one version of Close, recorded in November 1980 at Leyland Hill Farm, with Doug Irvine present at least part of the time. In keeping with tradition, Irvine could have quit after he put down vocals but before he did a final take of the bass line, forcing Rothery to play it instead. (Or perhaps the band decided to keep the vocals to show the “songiness” of Close, but saw no harm in replacing an ex-member's bass part.)

Against this we have two arguments. First, MARILLION - A HISTORY does state the “instrumental track [of Close] was sent up to” Fish and Diz. But it doesn't mention Lady Fantasy at all, so it's a bit unreliable. Even if it was an instrumental of Close, that's more evidence the song was recorded at Leyland Hill Farm in November. Just as it would be easier to overdub a recording at the studio where it was first put down, it would have been fairly easy for the band to bounce a vocal-less mix to tape to send to prospective singers if they were still in the studio with it – suggesting again that Irvine quit while the session for Close was still going on.

The second argument against my theory was put to me by Andre Kreutzmann, and it goes something like this: why the heck would a band with a singer/bassist go to the trouble of recording his not-so-good vocals but not his perfectly good bass playing? To this I have no real answer except to say that by all evidence, that's exactly what seems to have happened. Maybe Irvine did a bass track that got replaced after he quit. Maybe he hadn't written the part yet. Maybe he was having a good voice day! Recording sessions can be funny things.

The last thing to mention about Close is that a few years ago, a fan got in touch with Doug Irvine and played him the bootleg, presumably to try to answer some of the questions I've posed. Doug's response was that it was definitely him singing, but he had absolutely no memory of recording the track! One wonders exactly what the band was up to during the last days of November 1980.


This last area of study is arcane in the extreme, but I discuss it both out of a perverse impulse towards completeness and recognition of the fact it has been the source of much confusion in the bootlegging world. In short, I propose there are multiple bootleg lineages, perhaps based on different copies of the original multi-track tapes, played at slightly variant speeds.

Most of the surviving demos have two versions, one that's more or less at pitch and another that has been slowed down. These slow versions are also pitched a bit lower – somewhere around a quarter-tone, maybe a half-tone – resulting in them sounding like different recordings entirely at first listen!

I've heard fast and slow versions of all the 1980 demos except Scott's Porridge. As mentioned above, the early vocal version of Alice often includes the beginning few seconds of Scott's Porridge as a sort of accidental coda, and that snippet is actually higher and faster than the separate long version that survives. That suggests that Scott's Porridge comes from the slow lineage, but what it says about the recording date of the original Alice is hard to know. Perhaps it's more evidence that the first two Alice recordings were made at the same time as the other low-production demos, since there are fast and slow lineages of all of those – though not of those Alice takes.

There are also fast and slow versions of Fish's first overdub of The Web, based on Close.

I consider the fast versions of the songs to be more accurate. This assessment is based mostly on their tuning, of which the best example is “Lady Fantasy.” The arpeggiated acoustic guitar that opens the song is based on an open D major chord, a common enough choice for guitarists and a key of which Steve Rothery tends to be fond. The fast version is pretty close to D, whereas the slow version is about at D flat. It seems unlikely Rothery would down-tune his guitar a half-step for the song. Alice makes more sense in D as well – D minor in this case – since it stayed in that key when it was recycled into Forgotten Sons. Like Lady Fantasy, the fast version of Alice is in D or close to it, whereas the slow version approaches D flat. A final excellent piece of evidence comes from The Web: Fish's unique voice is immediately recognizable in the fast version, whereas it sounds unusually low in the slow version.

Early Demos And Sessions 1979 - 1981, bootleg Sessions Demos And Outtakes, bootleg

I'll end with a quick example of how much confusion these alternate bootleg lineages have caused. Probably the most readily available bootleg collection online is called Early Demos And Sessions 1979-1981 and has an air of authority, with its metadata declaring definitive dates for the various songs:

  1. Lady Fantasy 1979-10-XX
  2. Alice 1979-10-XX
  3. The Haunting Of Gill House 1980-03-XX
  4. Herne The Hunter 1980-03-XX
  5. Scott's Porridge 1980-03-XX
  6. Close 1980-06-06
  7. Lady Fantasy 1980-06-06
  8. Alice 1980-06-06
  9. Close 1980-11-XX
  10. The Web 1981-01-02

The relevant track listing should really look like this:

  1. Lady Fantasy Fast
  2. Alice (C) Fast
  3. The Haunting Of Gill House Fast
  4. Herne The Hunter Slow
  5. Scott's Porridge Slow
  6. Close Slow
  7. Lady Fantasy Slow
  8. Alice (C) Slow
  9. Close Fast
  10. The Web Fast

Note how the fast and slow lineages of songs like Alice and Lady Fantasy are purported to be not only different recordings, but from entirely different sessions. I suspect my point is clear: when we're not even sure when certain recording sessions even occurred, representing them using tape-warped versions of the same tracks muddies the waters immensely and creates the sort of confusion that, forty years on, one can only dream of unraveling!


The truth will never be known for certain, that much is obvious. But we can clear out decades' worth of cobwebs and earn ourselves a better view of the few treasures left behind after Doug Irvine's departure. The four big questions marks are:

  1. whether the rough demos were made in the practice room or at The Lodge
  2. when, prior to March 1980, Marillion went to The Lodge
  3. whether Scott's Porridge and the first two takes of Alice were recorded alongside the other rough demos, or if they really reach back to October 1979
  4. when Close was recorded

I'm wary about hazarding guesses, but for the sake of completeness, here are attempts at filling out our two timelines:

Timeline 1

  • October 1979 - January 1980: Alice (A) (B), Scott's Porridge, The Haunting Of Gill House, Herne The Hunter recorded during rehearsals
  • February or March 1980: Alice (C), Lady Fantasy recorded on multi-track at The Lodge
  • November 1980: Close recorded on multi-track at Leyland Hill Farm

Timeline 2

  • February or March 1980: Alice (A) and (B), Scott's Porridge, The Haunting Of Gill House, Herne The Hunter recorded on two-track at The Lodge
  • June 6th, 1980: Alice (C), Lady Fantasy recorded on multi-track at The Lodge
  • November 1980: Close recorded on multi-track at Leyland Hill Farm

Of course, new evidence may come to light at any point that would throw this all into the trash can. Jerry Van Kooten first pointed out that “two-track” might mean “two songs.” I was skeptical Alice (A) and (B) even existed until Andre Kreutzmann hunted them down for me – perhaps there really is an instrumental Lady Fantasy, too. I'd also love to hear the instrumental version of Close that may or may not have been sent to Fish and Diz.

Whatever their provenance, these tracks give us a glimpse into the past, a peek at the private history of a band that didn't just “hit the scene back in '81” fully formed, but put in the long, grueling hours of songwriting and rehearsal that made their later success seem so easy.