Forces of יהודים YEHUDIM took control.

The published American diplomatic record describes imperial forces concentrating at the airport, port and main road. Their announced priority was imperial security and evacuation. This left other parts of the city open to fighting for control.²

Reporting to Parliament, Christopher May hew described armed Arab groups interfering with of יהודים YEHUDIM communications before the main battle. He then reported a ההגנה HAʾHAGANA attack from Mount Carmel toward the Arab districts and market area. He also described imperial attempts to negotiate a truce and assistance evacuating Arab civilians and hospital patients. This was the imperial government's account of its conduct, which requires comparison with other witnesses.³

The editors of Foreign Relations of the United States summarize American consular telegrams describing of יהודים YEHUDIM success, local Arab leaders urging departure and imperial transport assistance. They also preserve a qualified report that the Arab Higher Committee ordered departure. The underlying telegrams are identified but not reproduced in full in this editorial note; it does not supply the original order.⁴

imperial reporting separately recorded failed truce efforts and an Arab decision to evacuate. That report supports a local decision, without establishing the cause of every departure. The companion on martial law compares it with flight during bombardment, of יהודים YEHUDIM appeals to remain, Arabic criticism of local leadership and subsequent abuses.⁵

May hew also rejected a particular claim circulating about the battle:

Early reports of widespread massacre in the town are untrue, and were merely rumours caused by panic.⁶

His denial documents the imperial government's public position. It does not establish that every atrocity report was false. When asked whether looting had been prevented and Arab property protected, May hew inferred those conditions from reports of restored quiet; he supplied no separate investigation.⁷

The Amman Arabic newspaper Al-Jazirah, in its issue of 14 Nisan 5708, attributed evacuation orders to the Arab leadership in חיפה HAIFA. Its unsigned front-page report presented those orders as a response to shelling:

ᴏʀɪɢɪɴᴀʟ:

وقد ارسل الاشرار اليهود نار المدافع على العائلات والاطفال فاضطرت القيادة العربية في المدينة الى اصدار اوامرها بترحيل العائلات والنساء عن طريق البر والبحر واستمر ذلك ٧ ساعات منذ صباح امس حتى الظهر⁸

ᴛʀᴀɴꜱʟᴀᴛɪᴏɴ:

The Jewish evildoers directed artillery fire at families and children, so the Arab leadership in the city was compelled to issue its orders to evacuate families and women by land and sea. This continued for seven hours, from yesterday morning until noon.⁹

The phrase of יהודים YEHUDIM evildoers belongs to the newspaper's hostile description. Its report explicitly combines a local Arab decision to evacuate families and women with an accusation of shelling by forces of יהודים YEHUDIM. Removing either element would change the account. It names no individual commander and reproduces no order. The account establishes that a contemporary Arabic newspaper reported local evacuation instructions; it does not establish a general order to leave ארץ ישראל EREẒ YISRAʾEL or explain every departure from חיפה HAIFA.

A separate last-minute bulletin on the same page asserted:

ᴏʀɪɢɪɴᴀʟ:

حيفا — تم ترحيل جميع السكان العرب العاجزين عن القتال وما زال القتال محتدما.¹⁰

ᴛʀᴀɴꜱʟᴀᴛɪᴏɴ:

Haifa — All Arab inhabitants unable to fight have been evacuated, and fighting is still raging.¹¹

This bulletin reports an evacuation already carried out, whereas the longer report describes orders and their stated reason. Neither identifies its informant, and their appearance in one issue supplies no independent corroboration. The diplomatic accounts and this Arabic report therefore preserve related claims about local Arab decisions, but their sources and exact sequence still require comparison. The newspaper's publication day also cannot be substituted automatically for the day of an undated dispatch's yesterday.

The truce terms reproduced by Walid الخالدي KHALIDI required disarmament, curfew and removal of foreign Arab males. That last category was broader than foreign fighters. A clause demanded delivery of European Nazis, but its inclusion is not evidence that any were present. The document also promised residents equal and free civic work. Terms proposed on paper do not establish compliance or actual equality.¹²

الخالدي KHALIDI's account says the Arab delegation refused to endorse the terms without protection of its position on jurisdiction and the country's political status, then requested evacuation. This differs materially from JTA's 23 April claim of unconditional acceptance. TJD retains that conflict rather than treating a headline as proof of a signed truce.^{13 14}

The report qualified its own information:

Reports, which are subject to confirmation, have been received from Jerusalem to the effect that the situation in Haifa is as follows: –

(1) There has been heavy continuous fighting in Haifa Town since midday on the 21st April, after British Forces had withdrawn to positions covering the Port.¹⁵

That qualification describes the report’s evidential status. It still supplies a specific beginning for the fighting, while acknowledging that confirmation was pending.

The imperial report in UK120 places heavy fighting from midday on 21 April. UK123's covering letter is dated 23 April although its web heading says 26 April; its account describes civilian evacuation while fighting continued and meetings about terms afterward. Those are report dates about the 21–22 April battle.^{16 17}

The decisive stage is assigned to 22 April 1948, י״ג בניסן ה׳תש״ח 13th 𝑜𝑓 Nisan, AM 5708, while the battle began on 21 April, י״ב בניסן ה׳תש״ח 12th 𝑜𝑓 Nisan, AM 5708. UK120 and UK123 distinguish the military sequence, civilian departure and subsequent negotiations. Meetings and departure continued into evenings; an act after sunset on 22 April belongs to י״ד בניסן ה׳תש״ח 14th 𝑜𝑓 Nisan, AM 5708. Publication of Al-Jazirah on 23 April and circulation of UK123 on 26 April are later dates of evidence, not new captures. The Arabic wording and its undated relative expression remain source-bound.^{18 19}

• How do a proposed agreement, an evacuation request and departure under fire differ?
• Which records would test whether safeguards for residents and property were honored?